Widely regarded as one of the greatest and most influential bands in the history of popular music, there is plenty of available literature on the techniques and equipment used to record The Fab Four, including my personal favorite Recording The Beatles by Brian Kehew and Kevin Ryan.
Here are some examples of creative production techniques used by the group and the talented crew of engineers that helped create a catalog of albums that have sold well over two billion copies:
1. Recording at half-speed on In My Life”
On this song from 1965 Album Rubber Soul, all but the instrumental bridge section had been completed. John Lennon asked Producer George Martin to come up with something baroque sounding.”¹
Martin's Bach-inspired Piano solo was written at a tempo that Martin himself couldn't play. The engineers brilliantly recorded the part with the tape running at half-speed, so when it was played back at normal speed, the solo was an octave higher and twice as fast. Additionally, the timbre was altered, with the attack of the notes played on the piano becoming more prominent.
There are numerous other examples of the engineers using this technique on Beatles recordings, including extensive use on Strawberry Fields Forever.”
Using the varispeed” mode in Pro Tools' elastic audio can perhaps yield similar results.
2. Reverse tape effect on the guitar solo of I'm Only Sleeping”
On this John Lennon tune from Revolver (1966) George Harrison spent a reported five hours meticulously constructing a guitar part by having the engineers run the tape backwards as he composed a solo that would ultimately, when reversed, fit the dreamlike mood.”²
The part was double-tracked, once with fuzz, and once without. Conveniently, you can hear the entire master reversed, revealing what the original guitar recording sounds like at Beatles Bible
Essentially every modern DAW has a reverse audio capability, but actually taking the time to write out the performance before the effect is applied will definitely result in something unique.
3. Randomly splicing tape loops together on Being for the Benefit of Mr. Kite!”
For this song, which derived virtually all of its lyrical content from a 19th century circus poster owned by John Lennon, a carnival atmosphere” was desired for the production.
In the middle-eight, we have perhaps one of the most creative techniques used on The Beatles recordings, with engineer Geoff Emerick (as instructed by George Martin) taking tape recordings of fairground organs and calliope music, chopping the tape with scissors, throwing the pieces up into the air, and reassembling at random.⁴
The resulting effect is quite unique, and fits in perfectly with the rest of the psychedelic Sgt. Pepper's Lonely Hearts Club Band.
This type of effect (if desired) is not only much easier to do in a modern DAW, it is certainly cheaper than mangling sought-after analog tape.
4. Use of Tea Towels and other drum muffling techniques on multiple recordings
As early as 1962, Ringo can be seen using John's Harmonica to dampen his snare drum.
Throughout the years, it was quite common to use tea towels to muffle the sound of his drums. A collection of pictures and information on Ringo's kits and evolution can be seen at Ringo's Beatle Kits This became a sonic staple for the band.
Using tea towels or other muffling devices can allow for more control over the volume, attack and decay of individual drums. Especially considering it was common for the engineers to apply extreme compression on Ringo's kit with a Fairchild limiter, dampening the drums allowed for a tighter, more focused sound.
I highly suggest having Moongel dampening on hand, but don't be afraid to use something more aggressive. Native Instruments' Abbey Road 60's Drums even come with tea towel articulations, making it one of my go-to virtual instruments for vintage sounding drums.
Bonus: Practicing singing on Oh! Darling”
Performing a song until the performance sounds the way the artist wants it to — what a concept!
During the Abbey Road sessions Paul McCartney would come in early to the studio that this classic album shares a name with, arriving before the other band members, singing this retro-sounding song once per day attempting to capture the raw, strained quality that the production needed. Engineer Alan Parsons recalls:
Paul came in several days running to do the lead vocal on Oh! Darling. He'd come in, sing it and say, ‘No, that's not it, I'll try it again tomorrow.' He only tried it once per day, I suppose he wanted to capture a certain rawness which could only be done once before the voice changed. I remember him saying, ‘Five years ago I could have done this in a flash,' referring, I suppose, to the days of Long Tall Sally and Kansas City”⁴
There is not yet a plugin that can achieve this effect. This is just one of the countless examples of McCartney's complete (and sometimes obsessive) dedication to capturing a musical moment exactly as he envisioned it.
Conclusion
I've always quite enjoyed researching the production secrets of my absolute favorite band. Fortunately, there's a wealth of knowledge available on recordings, in books, and on various Internet sources.
Additionally, much credit to Waves, Softube, and Native Instruments whose lines of Abbey Road-inspired plugins and VSTs help preserve the sounds of this crucial time in recorded music for current engineers.
Showing posts with label production. Show all posts
Showing posts with label production. Show all posts
Tuesday, 27 June 2017
Tuesday, 20 December 2016
20 Vocal Production Tips
Achieving the perfect vocal recording is a complex business. Hollin Jones explains how to get the best out of your recording and your kit…
1: Keep It Dry
It's a good idea to always record your vocals dry unless you have a specific reason to do otherwise. While the character of a guitar sound might come from the amp and pedals, in 99 per cent of cases you'll be better served by recording a completely clean, dry vocal and then adding effects afterwards. There are several reasons for this.
Recording reverb or delay as part of the take can make comping and other edits much harder because you introduce effect tails to the signal which sound odd when cut and joined together. And although you may end up EQ'ing a lot of the frequencies in a vocal up or down, it's best to start with everything present then cut frequencies surgically, rather than starting with a top-heavy take and having to try to add mid and low end back in afterwards.
2: Create Space Pockets
Wherever you record vocals, be it in a spare room at home or in a studio, it's crucial to create a contained space for your vocal performance. Most people won't be recording in an optimum space so there can be a lot of reflections from walls and other surfaces that will colour your recording, and these can be time consuming to remove afterwards using EQ and other tools.
To avoid all of that, a great technique is to use a portable vocal isolation booth such as sE's Space or a similar product. These create a pocket of dead space for your recording which leads to a much cleaner take and also has the added benefit of helping the performer focus on the task at hand, blocking out the view of the room. They can also be used for recording guitars and other sounds, so they're not just for vocals.
3: Consider Mic Emulation
It's common sense that you should use the best mic you can lay your hands on for vocal recording, but not everyone will have a top-end model to call on. The best mics cost thousands but this is beyond the scope of most people's budgets. A good solution is to get a solid all-round performer that doesn't cost the earth: perhaps a Rode NT1 or NT2 or one of sE's many condenser mics or something similar, that has a fairly clean, neutral sound.
After recording you can then try using mic modelling to emulate the characteristics of different mic models without incurring quite so much cost. Waves makes a range of mic modeling plug-ins, and there's Antares' Mic Mod EFX, Universal Audio's Ocean Way plugin and more. You probably won't get exactly the same sound as using a £5000 mic but with some care and attention you can get a great effect.
4: Aim for DSP Powered Monitor Effects
One of the best ways to help a vocalist give a more confident and natural performance is to provide them with some reverb in their headphones. The contrast with trying to sing completely dry can be quite startling when you actually do it. However it's not always quite as easy as it sounds. The problem with using insert reverb for this task is that you will almost always incur at least a small amount of latency as the vocal is processed live through your DAW and sent back to the singer's headphones.
This can be incredibly off-putting for the vocalist, and the fiddling with audio buffer sizes to minimize it can be a hassle too. Some audio interfaces - typically medium or larger sized ones - have onboard DSP-powered effects that can be applied to the monitored signal and provide reverb in the headphones but not incur any latency along the way, as it's processed inside the interface and not the computer.
If you record a lot of vocals this is an important consideration when you're choosing an interface.
5: De-ess with Care
Sibilance is an issue with any vocal recording and although it can be mitigated with a popshield it's harder to block out than plosives. One solution is to use a de-esser plug-in after recording. These are specialised compressors that work on a very specific frequency range to rein in the ‘s' sounds in vocals - used correctly, they can be very effective at cleaning up your vocal track.
However it's vital not to overuse them because they can quickly start to suck the life and energy out of a vocal. If you dial in too much de-essing or set your processor to pull back too wide a range of frequencies you'll find that the vocal starts to sound artificial and overly restrained. The key is to find a happy medium of frequency, de-essing amount and the aggressiveness of the attack and decay, and take out the harshest of the ‘s' sounds without dulling the sound too much overall.
6: Record Everything
One trick that some producers use is to record everything a singer does, sometimes without telling them it's being captured. A good example is recording warm-up or test takes, when a vocalist feels less pressured and more relaxed because they don't think this one is ‘it' and may give a different kind of performance to the one they give when they think they are centre stage.
This doesn't just apply to amateur vocalists, it has been known to work with professionals as well. Sometimes after the session is over you go back to the warm ups or the tests and find whole takes or parts of takes that work better than the ‘real' ones, and then splice them in. Tracks and storage space are plentiful on modern systems so there's no reason not to do this.
7: Sweet Harmonies
Harmonies and backing vocals can add a lot to the overall fullness of a track and it's possible to create these after the fact if you're not able to do it on the same day. Obviously the ideal situation is to get your singer to perform as many of them as possible at the time and then you have more to work with afterwards. But even if all you have is a single lead vocal you still have options.
The simplest trick is to double or triple track the vocal and then mix, pan and EQ the copied versions to enhance the main track. You can also process the duplicates differently, perhaps adding chorus or unison, or other kinds of effects. Some DAWs let you generate harmonies from existing tracks: Cubase is able to do this, and there are plug-ins that can help too. It's also possible to pitch shift duplicated tracks to create harmonies, though this requires more care and effort since not every note will sit properly with the original take.
8: Loop and Comp
When it comes to getting the perfect take it's rare to find a singer who can do a whole song perfectly in one go. That's not a reflection on their talent, just that with modern production techniques there's no reason to settle for anything less than total perfection. You can set up loops to record over different sections of a track and your performer can have several goes at it without stopping, or indeed try variations as the track loops.
You can then use the comping tools that your DAW most probably has to find the best takes or best parts of takes and splice them together - with care - to get the perfect take. This may involve changing the levels of different takes if the singer has moved a little in relation to the mic at different times. You can also overdub of course, punching a singer in only at the specific point they need to replace a line or word, then punching them out.
9: Gate Post-Recording
Gating can be a useful tool in the producer's arsenal but it should be used carefully. Although it might be tempting to use a noise gate during recording, it's not often a good idea.
Gates work by cutting off the signal - closing - when the signal falls beneath a specified threshold. They open again when the signal rises above that threshold, with accompanying attack and release settings to govern how quickly they open and shut.
The problem with this is that since a live vocal performance is an organic thing, with your vocalist moving around a little and varying their volume without you necessarily realising, having a preset gate can result in words or parts of lines being completely missed if the thresholds that have been set don't respond to the actual performance.
A closed gate lets absolutely nothing through. A much better idea is to record everything, including any stray sounds or breaths between vocal lines, and then either apply the gate as an insert so you can tweak it after the recording, or physically cut the sections of silence out from the timeline
10: Pan Backing Vocals
Your lead vocal should always be front and centre, but you can get much more creative with your harmonies, backing vocals or hip hop ad libs. Think about placing them creatively, maybe using panning to send them a little off to the left or right. This makes the track sound much fuller and it also ensures that they complement rather than compete with the lead vocal in the centre.
It's quite rare to hard pan vocal sounds as this can sound a little odd especially on headphones, but with things such as hip hop ad libs it's common to send them off to 45 degrees left or right. To get even more creative, consider automating their pan settings so perhaps one backing line appears off to the left, and the next line off to the right, and so on.
1: Keep It Dry
It's a good idea to always record your vocals dry unless you have a specific reason to do otherwise. While the character of a guitar sound might come from the amp and pedals, in 99 per cent of cases you'll be better served by recording a completely clean, dry vocal and then adding effects afterwards. There are several reasons for this.
Recording reverb or delay as part of the take can make comping and other edits much harder because you introduce effect tails to the signal which sound odd when cut and joined together. And although you may end up EQ'ing a lot of the frequencies in a vocal up or down, it's best to start with everything present then cut frequencies surgically, rather than starting with a top-heavy take and having to try to add mid and low end back in afterwards.
2: Create Space Pockets
Wherever you record vocals, be it in a spare room at home or in a studio, it's crucial to create a contained space for your vocal performance. Most people won't be recording in an optimum space so there can be a lot of reflections from walls and other surfaces that will colour your recording, and these can be time consuming to remove afterwards using EQ and other tools.
To avoid all of that, a great technique is to use a portable vocal isolation booth such as sE's Space or a similar product. These create a pocket of dead space for your recording which leads to a much cleaner take and also has the added benefit of helping the performer focus on the task at hand, blocking out the view of the room. They can also be used for recording guitars and other sounds, so they're not just for vocals.
3: Consider Mic Emulation
It's common sense that you should use the best mic you can lay your hands on for vocal recording, but not everyone will have a top-end model to call on. The best mics cost thousands but this is beyond the scope of most people's budgets. A good solution is to get a solid all-round performer that doesn't cost the earth: perhaps a Rode NT1 or NT2 or one of sE's many condenser mics or something similar, that has a fairly clean, neutral sound.
After recording you can then try using mic modelling to emulate the characteristics of different mic models without incurring quite so much cost. Waves makes a range of mic modeling plug-ins, and there's Antares' Mic Mod EFX, Universal Audio's Ocean Way plugin and more. You probably won't get exactly the same sound as using a £5000 mic but with some care and attention you can get a great effect.
4: Aim for DSP Powered Monitor Effects
One of the best ways to help a vocalist give a more confident and natural performance is to provide them with some reverb in their headphones. The contrast with trying to sing completely dry can be quite startling when you actually do it. However it's not always quite as easy as it sounds. The problem with using insert reverb for this task is that you will almost always incur at least a small amount of latency as the vocal is processed live through your DAW and sent back to the singer's headphones.
This can be incredibly off-putting for the vocalist, and the fiddling with audio buffer sizes to minimize it can be a hassle too. Some audio interfaces - typically medium or larger sized ones - have onboard DSP-powered effects that can be applied to the monitored signal and provide reverb in the headphones but not incur any latency along the way, as it's processed inside the interface and not the computer.
If you record a lot of vocals this is an important consideration when you're choosing an interface.
5: De-ess with Care
Sibilance is an issue with any vocal recording and although it can be mitigated with a popshield it's harder to block out than plosives. One solution is to use a de-esser plug-in after recording. These are specialised compressors that work on a very specific frequency range to rein in the ‘s' sounds in vocals - used correctly, they can be very effective at cleaning up your vocal track.
However it's vital not to overuse them because they can quickly start to suck the life and energy out of a vocal. If you dial in too much de-essing or set your processor to pull back too wide a range of frequencies you'll find that the vocal starts to sound artificial and overly restrained. The key is to find a happy medium of frequency, de-essing amount and the aggressiveness of the attack and decay, and take out the harshest of the ‘s' sounds without dulling the sound too much overall.
6: Record Everything
One trick that some producers use is to record everything a singer does, sometimes without telling them it's being captured. A good example is recording warm-up or test takes, when a vocalist feels less pressured and more relaxed because they don't think this one is ‘it' and may give a different kind of performance to the one they give when they think they are centre stage.
This doesn't just apply to amateur vocalists, it has been known to work with professionals as well. Sometimes after the session is over you go back to the warm ups or the tests and find whole takes or parts of takes that work better than the ‘real' ones, and then splice them in. Tracks and storage space are plentiful on modern systems so there's no reason not to do this.
7: Sweet Harmonies
Harmonies and backing vocals can add a lot to the overall fullness of a track and it's possible to create these after the fact if you're not able to do it on the same day. Obviously the ideal situation is to get your singer to perform as many of them as possible at the time and then you have more to work with afterwards. But even if all you have is a single lead vocal you still have options.
The simplest trick is to double or triple track the vocal and then mix, pan and EQ the copied versions to enhance the main track. You can also process the duplicates differently, perhaps adding chorus or unison, or other kinds of effects. Some DAWs let you generate harmonies from existing tracks: Cubase is able to do this, and there are plug-ins that can help too. It's also possible to pitch shift duplicated tracks to create harmonies, though this requires more care and effort since not every note will sit properly with the original take.
8: Loop and Comp
When it comes to getting the perfect take it's rare to find a singer who can do a whole song perfectly in one go. That's not a reflection on their talent, just that with modern production techniques there's no reason to settle for anything less than total perfection. You can set up loops to record over different sections of a track and your performer can have several goes at it without stopping, or indeed try variations as the track loops.
You can then use the comping tools that your DAW most probably has to find the best takes or best parts of takes and splice them together - with care - to get the perfect take. This may involve changing the levels of different takes if the singer has moved a little in relation to the mic at different times. You can also overdub of course, punching a singer in only at the specific point they need to replace a line or word, then punching them out.
9: Gate Post-Recording
Gating can be a useful tool in the producer's arsenal but it should be used carefully. Although it might be tempting to use a noise gate during recording, it's not often a good idea.
Gates work by cutting off the signal - closing - when the signal falls beneath a specified threshold. They open again when the signal rises above that threshold, with accompanying attack and release settings to govern how quickly they open and shut.
The problem with this is that since a live vocal performance is an organic thing, with your vocalist moving around a little and varying their volume without you necessarily realising, having a preset gate can result in words or parts of lines being completely missed if the thresholds that have been set don't respond to the actual performance.
A closed gate lets absolutely nothing through. A much better idea is to record everything, including any stray sounds or breaths between vocal lines, and then either apply the gate as an insert so you can tweak it after the recording, or physically cut the sections of silence out from the timeline
10: Pan Backing Vocals
Your lead vocal should always be front and centre, but you can get much more creative with your harmonies, backing vocals or hip hop ad libs. Think about placing them creatively, maybe using panning to send them a little off to the left or right. This makes the track sound much fuller and it also ensures that they complement rather than compete with the lead vocal in the centre.
It's quite rare to hard pan vocal sounds as this can sound a little odd especially on headphones, but with things such as hip hop ad libs it's common to send them off to 45 degrees left or right. To get even more creative, consider automating their pan settings so perhaps one backing line appears off to the left, and the next line off to the right, and so on.
Sunday, 18 December 2016
7 Advanced Guitar Mixing Techniques For An Amazing Guitar Production Audio Issues
Building upon one of my older articles, Top 10 Guitar Mixing Techniques , here are even more tips and tricks to make your electric guitar sound better in your mix.
1. Long Release on Chords - For a longer, more sustained sound for chords, use a long release on your compressor. Think if it this way: the longer the release, the longer the sustain of the chords.
2. Fast Attack to Reduce String Sound - Compression is used to create shape, it's not just used to control the volume. A really fast attack really pushes down on the initial strum, resulting in a more pad-like sound.
If you combine a fast attack with a fast/medium release you'll end up with a chord that seems to wash over you, especially if you hit the chords(and your compressor) hard.
3. Re-amp if Needed - Re-amping means routing your performance from your DAW into an amp and recording it again. If you recorded a great performance through a guitar emulator and you're not satisfied with the sound, re-amping is a great way to give your electric guitar more depth. It's a perfect way to tweak a great performance that has a lackluster sound.
4. Cut at 800 Hz - I always seem to dip a little at 800 Hz, it takes the cheap sound away. Especially if I'm working with cheap guitars.
5. EQ the reverb - Reverb is the number one culprit when it comes to a muddy mix. This trick is applicable to any instrument, but make sure to filter out some of the low-end of your reverb send when mixing electric guitar. It'll still give your guitar space and depth, but without the unnecessary muddiness.
6. Use subgroups - If you're working with many similar guitar track, grouping them together makes mixing a lot easier. This is one of the most important advice I can give. It's simple but extremely important. Bus, route and group your tracks and suddenly your tracks become more manageable.
7. Use the Right Reverb - Take your time to go throw your reverbs. Choosing the right reverb depends on a lot of things: genre of music, speed of song, type of feel.
You can't get a large open vocal sound with a small jazzy -club preset. You're better off with a hall setting. Also, a fast rock song with a cathedral reverb is going to muddy up the mix immediately. So take your time to use the right reverbs for the song.
1. Long Release on Chords - For a longer, more sustained sound for chords, use a long release on your compressor. Think if it this way: the longer the release, the longer the sustain of the chords.
2. Fast Attack to Reduce String Sound - Compression is used to create shape, it's not just used to control the volume. A really fast attack really pushes down on the initial strum, resulting in a more pad-like sound.
If you combine a fast attack with a fast/medium release you'll end up with a chord that seems to wash over you, especially if you hit the chords(and your compressor) hard.
3. Re-amp if Needed - Re-amping means routing your performance from your DAW into an amp and recording it again. If you recorded a great performance through a guitar emulator and you're not satisfied with the sound, re-amping is a great way to give your electric guitar more depth. It's a perfect way to tweak a great performance that has a lackluster sound.
4. Cut at 800 Hz - I always seem to dip a little at 800 Hz, it takes the cheap sound away. Especially if I'm working with cheap guitars.
5. EQ the reverb - Reverb is the number one culprit when it comes to a muddy mix. This trick is applicable to any instrument, but make sure to filter out some of the low-end of your reverb send when mixing electric guitar. It'll still give your guitar space and depth, but without the unnecessary muddiness.
6. Use subgroups - If you're working with many similar guitar track, grouping them together makes mixing a lot easier. This is one of the most important advice I can give. It's simple but extremely important. Bus, route and group your tracks and suddenly your tracks become more manageable.
7. Use the Right Reverb - Take your time to go throw your reverbs. Choosing the right reverb depends on a lot of things: genre of music, speed of song, type of feel.
You can't get a large open vocal sound with a small jazzy -club preset. You're better off with a hall setting. Also, a fast rock song with a cathedral reverb is going to muddy up the mix immediately. So take your time to use the right reverbs for the song.
Sunday, 20 November 2016
96 Vocal Production Tips
01
We generally recommend that if you're aiming for really pro-sounding vocals it's easier to have a real preamp on the go, keeping good signal levels on the way in. If you haven't got access to one of these, however, you can always try throwing in something that provides the same sort of flavour. Plug-ins such as PSP's Mix Saturator 2 offer some handy preamp settings that quite convincingly mimic outboard saturation and compression.
02
It's vital to get the best mic position for your vocalist. While it's usually best to start from right in front of the mic, some people's voices may sound better a little closer or further away, or even to the side. If your singer's voice has a nasal quality, for example, you might want to place the mic slightly above their nose rather than in front of or below it. Experiment until you find the sweet spot.
03
Many vocalists sing better when they tilt their head back a little, as it opens up their vocal chords. The easiest way to take advantage of this is to place the mic slightly higher than the singer's mouth, angled down towards them - this will encourage them to tilt their head back as they sing.
04
It might sound a little obvious, but before doing a vocal take, it's a good idea to make sure your singer has a clear idea of where the lines should be placed. There's nothing wrong with having them scribbled down on a piece of paper, but generally speaking, a singer who doesn't know the lyrics probably isn't that sure how to sing them either.
05
Make sure you keep track of your singer's position (even if your singer is you). You will often find yourself taking breaks, or coming back to re-record a line, and it's very important to make sure the vocals are coming from the same position each time - if not, they might not sound the same. It goes without saying that you should keep track of where the mic was, too.
Artificial stereo spread effects can push a vocal back in the mix, so try to keep them for backing or quieter layered vocals.
06
Adding an artificial stereo spread effect to a vocal can be a great way of giving it real presence, but it also has the effect of taking away some of the vocal's punch and causing it to lose some of its prominence in a mix, so is often not ideal on the lead vocal. Instead, try using it for backing vocals or layered words, or for specific effect on certain words.
07
Chorus effects are great, but one of the best ways to achieve chorusing is to actually record multiple takes of the vocal. You can then take a few of these and layer them - try panning a couple slightly and leaving one centred, or thinning out the underlying two a little with EQ and more heavily compressing the central one for added impact.
08
You can thicken up a thin-sounding vocal by using a synth layered underneath it that subtly plays the same notes. You can even use a sidechained gate on the synth, triggered by the vocal, to ensure they only play together. Alternatively, try using something like Waves Tune, which will export the notes of your vocal as a MIDI file. Then compress the synth and vocal as one.
09
There are plug-ins out there, like Antares Punch, which are designed to accentuate the more punchy parts of a vocal, but you can also achieve something similar using gating or even an expander on your compressor. With a gate, for example, you just need to set it up so that only the punchy parts of your vocal open the gate, and then apply a small amount of gain reduction.
10
One of the great ways to keep a vocal punchy but still interesting is to use some of your additional takes to layer key words. These could be words that are significant in a lyrical sense, or that work well with the groove, or that enhance the arrangement. Just be sure not to overdo it - use them sparingly enough that they keep their impact when they appear.
Try chopping your vocal up and loading the parts into a sampler, then applying some groove quantise to them.
11
We often talk about adding groove quantise to musical and rhythmic parts, but it can also be very effective on a vocal line. How you do this is up to you - Live users might try putting warp markers on the individual words of the track and then applying quantise, but we like putting the vocal in a sampler so the quantise can affect the length of the words as well as their placement.
12
A vocal line can sound a little jarring when it first appears, so try introducing it by taking the first word, reversing it and playing it into the normal version. You can use this effect anywhere, in fact - and don't forget that you can also timestretch the length of the reversed part to fit the space. Reverse reverb within your vocal can have a similar effect.
13
Once you've spent ages carefully ridding your vocal track of background hum, the cha-chinks of ice in a glass, coughs, comments and all the other bits of assorted noise that you only notice on a vocal track once it's been pumped with compression, it's often easy to forget how cool some of them can sound when sequenced into your track. Try it!
14
The easiest way to get rid of breath noise is to use a gate. You want to set it up very carefully to make sure you aren't losing any important parts of the words - a useful rule of thumb is to keep your attack and release times relatively slow. When in doubt, take it easy and manually remove any problem parts. There are de-breathing plug-ins that can help, too.
15
Breath noise can be problematic, but it's also vital on some vocals - completely removing it can make a vocal sound artificial and over-processed. Sometimes the words and breaths will slide between each other, so make sure you set any gain reduction carefully to leave some of the vocal in the track.
Whether manually or with a plug-in like Vocal Rider, try using old-fashioned volume controls to balance your vocal.
16
We often tell you to compress a vocal to achieve a balanced level, but before you do that, you want to ensure that the volume is level too. What's the difference? Well, compressing squashes the signal and makes it more powerful, but it also changes the overall character of the sound. Try riding the fader manually, changing the levels of individual word clips or using something like Waves' Vocal Rider to achieve a more natural-sounding result.
17
When recording a vocal, have your lyrics written out on a sheet and have three different- coloured pens to hand. When the vocalist sings a part of the vocal right, draw a line through that part with a pen. When they sing the same part well again on a subsequent take, strike the part through with a different colour. Don't stop recording until you have all three colours through every line!
18
Sometimes when you're on a deadline, have been looking forward to recording a vocal or have been planning it for ages, you might be tempted to go ahead with a recording even if your singer has a cold or otherwise isn't in the best condition. Check with the vocalist first - they know their voice, and they'll be able to make the best assessment of whether they're up to the job. Singing with a cold or sore throat is at best pointless, and at worst harmful. You'll all get frustrated, and waste your time.
19
We think it's always worth trying a vocal in a different key, just to see how it sounds. It can be very easy to get stuck in your ways, but it only takes a minute to sing a line in a different key. You might decide it sounds horrible, but you could also be pleasantly surprised. You might even decide to put a cheesy (sorry, 'epic') key change in the track.
20
You can add a lot of extra excitement to key bars or phrases in your track by adding some panning effects. This works on anything, but with vocals, it's ideal used on backing vocals or double-tracked words. The strength and speed of the panning are entirely up to you, but one thing that's often worth doing is softening the extremity of the pan.
Applying a little creative processing on key lines of a vocal is a great way to add emphasis and effect.
21
A lot of vocal processing is about achieving a consistent sound throughout, but that doesn't mean you can't accentuate some key lines. Try using some extreme processing - for example, you can put one phrase through an EQ and make it sound like it's coming down a phone line (cut below 300Hz and above 3kHz), to make it stand out.
22
To make a vocal sound more intimate, use very strong compression. The sounds you are trying to pull out are the breath and mouth noise of your vocal, as these are the sounds we associate with somebody speaking very close to our ears. With this in mind, it's also a good idea to keep your vocals quite dry too, for a more natural sound.
23
Remember that your vocal is usually not just another part of your track - it is your track. There will be musical elements that people like, and riffs that people love, but it's best to treat your vocal as the hook in almost every instance, as that is quite simply how most other people will view it.
24
Processing is great for adding emphasis to a vocal, but you can add interest by using some real gear too. For example, try recording a version of your vocal into an iPhone (you can set this up to run at the same time as you're capturing the vocal with a proper mic), or you might consider recording through a megaphone or guitar amp.
25
Try capturing vocals in different places. Generally dry is good, but if you have a great echoing hallway, or even a roof or a field… why not? When recording outside your main enemy will be wind, so you'll need either wind baffles or a foam cover, but if you want a completely dry vocal, free from reverb and reflection, it's hard to beat the middle of a field on a misty morning!
Singing slightly flat is a common problem among vocalists, so we often pitchshift the whole part up a few cents.
26
Most vocalists you are likely to encounter will sing a little flat rather than sharp, and this can often sound quite noticeable and unpleasant. One quick fix for a lot of vocal parts is simply to nudge the whole thing up by 5 to 25 cents. You'll be amazed how many vocals this will help.
27
Try different mics, if you have them, including cheap and old ones. Obviously if you're capturing an opera singer's voice, this could be considered something of a waste of their gift, but sometimes the character and flaws of the mic can really add something, even if only used as a layering effect.
28
While it is true that you can get awesome results with a relatively cheap microphone, good singers deserve a decent mic, and one that suits them. If you have access to a few different mics, try them out on different singers and you'll often hear little plosives or mouth noises that aren't audible on other mics. When aiming for expensive-sounding vocals, get a mic that suits your singer.
29
With automation available on almost everything in the studio now, it's incredibly easy to become a control freak with delay and other processing - so go for it! We like adding some delay to a few key words or parts of our vocal to create excitement. It can work with reverb and other effects too - and for added impact, try those effects on a bus and add an auto-panner.
30
Vocoded vocals can be quite hard to make intelligible. There are any number of more modern plug-ins out there now that are designed to make things clearer, and they usually do a great job, but it can still be really useful to ask your vocalist to enunciate clearly and to try to put some unnatural emphasis on the plosives and individual syllables.
Giving your vocalist a little reverb in their headphones will help them tell whether they are singing in key.
31
You should always try to give your singer some reverb in their cans while they're singing (unless they don't want it). The idea Giving your vocalist a little reverb in their headphones will help them tell whether they are singing in key is to keep their confidence high by giving them a good impression of their own voice, and to give them something they can pitch to.
Your vocalist can't listen and sing at the same time - at least, not like the engineer can listen - but with reverb, they can hear the pitch.
32
If you have the opportunity, and know you are working with a great singer, leave the mic running while they warm up (after checking with them that it's OK to do so). You could strike gold, even if you only catch a little phrase to drop in at the start of a track (think Amy Winehouse). Failing that, you might get some great archive material to use when VH1 run a special show to commemorate the 30th anniversary of your first album…
33
For a thicker vocal sound, don't just add reverb and delay on busses. Try adding them directly to the channel and then following that with some compression to really pull the elements together. Slap-back delay (with very fast delay times) is a perfect candidate for this kind of thing. It's also important to keep release and feedback times short to avoid making a mess.
34
You should try to have one set of headphones for the singer, and another set for you to listen out for any problems on the recording. It's far too easy, when you're just listening to a singer without cans, to miss a little bump from a ring or button on the mic stand that can completely ruin a recorded phrase.
35
No matter how carefully you've set your initial input level, singers will sometimes stray into louder territory than you planned, so even if you've tested their loudest note and left ample headroom, keep a stern eye on the incoming level to watch for clipping.
Grouping vocals helps pull different takes together and makes it easier to create separation between sections.
36
It's usually worth grouping the various parts of your vocal track so that you have one group (or channel) for the lead vox, one for doubled words and breath effects and one for the backing vocals, and then adding compression. This will pull all the vocal parts together nicely and make them seem more coherent, and will make it easier for the listener to differentiate the vocals from other elements of the track.
37
Generally speaking, the place for a vocal is in the centre of a mix. It sounds obvious, but listen to almost any great rock or pop recording and you'll hear that while the instruments might be dotted all over the place, the vocals are usually centred, and it's no accident. The vocal is normally the key element, and centring it makes it, well, central. We're also used to people speaking to us from the front, so it feels more personal.
38
Be very wary of anything that makes noise when recording. Even quiet noise will be amplified, including the slightest rustle of paper - this is why it's handy for the vocalist to have the lyrics memorised. Other culprits include rattling bracelets, necklaces and change or keys in pockets. Get rid of them before you start!
39
It's a good idea to always have a dictaphone or similar portable recording device ready. You never know when inspiration might strike, and even in a studio full of equipment, nothing quite beats the immediacy of hitting one button for instant recording. It can be digital, analogue… anything! They key is having instant, reliable recording at the touch of a button.
40
Although various kinds of, ahem, stimulants have been considered beneficial to the creative process by many a great musician, generally speaking, people sing better sober. Stick to a nice cup of tea - it will warm the singer's throat and wake them up. Try to steer clear of milk, though, as it acts as a coagulant in the throat and can have a really adverse effect on a singer's voice.
Chopped-up vocals will almost always sound better with a little octave-wide pitchshifting on individual words.
41
When chopping up parts of your vocal line and pitchshifting them, it's often hard to know where to begin. A good place to start is shifting an octave up. Alternatively, play a scale and pick notes that work in quick sequence. And whenever you repeat a syllable or vocal noise a few times, try shifting one up or down an octave. Experiment with the pitchbend, too.
42
It's usually worth grouping the various parts of your vocal track so that you have one group (or channel) for the lead vox, one for doubled words and breath effects and one for the backing vocals, and then adding compression. This will pull all the vocal parts together nicely and make them seem more coherent, and will make it easier for the listener to differentiate the vocals from other elements of the track.
43
One of the best ways to get vocals to cut through a mix is to use sidechain compression on competing musical parts. Instead of turning up the vocal, insert a compressor on the competing parts and let your vocal control the sidechain to duck the music. Don't overdo it - you shouldn't hear the music pump, but with careful attack and release times your vocal will magically appear in the mix.
44
One of the most common sources of noise on a vocal recording is the backing track itself, bleeding out of the headphones. One simple way to avoid this is to use closed over-the-ear headphones. If your singer likes to have only one cup over their ear, try panning the headphone signal so it's not bleeding into the mic.
45
If you don't have access to a singer but need to get a vocal down, you could try using a voice-altering plug-in, speech box or vocoder and recording your own version. You might be pleasantly surprised at how often you can come up with interesting, if classically imperfect, results.
Sidechain compression can be used to help individual words or phrases cut through the main vocal.
46
Sidechaining isn't just a good way to get your vocal to cut through a mix, it's also a good way to get a vocal to cut through… your other vocals. You might want, for example, one bit of backing vocal - or perhaps a bit of a spoken word or vocal noodling - to cut through. Apply a sidechain-enabled compressor to your main vocal group, and then use the one you want to stand out to duck it (just make sure it's not in the same group, ducking itself!).
47
We've all been here before: you're comping and editing a vocal, and you realise you haven't followed your own advice about keeping track of the vocal takes to make sure you've got a decent version of every line. So you're tweaking away, trying to get that one word to sound right. Stop. Call the singer and ask if they can either come into the studio to redo it, or record a new take and send it over.
48
When writing a song, get a guide vocal down as soon as possible - it needn't be the definitive take, but it will help you get on with constructing the whole song, which will provide you with clues about how to handle your vocals. Quite often, though, you'll capture something nice and raw - it's surprising how often a rough take becomes the final version.
49
Break up your vocal by using a doubled layer at the end of every verse or half-verse. If you've already double-tracked your vocals, this will mean adding another layer - one of the reasons why we suggest always getting at least three takes of a vocal. When recording, try getting a whole take sung as a harmony, ready to place wherever you need it.
50
This tip may raise the ire of the traditional songwriters out there, but in this day and age, the chorus of your song doesn't need to sound that great in its raw form. Modern music (pop, soundtrack and club alike) is replete with examples of pedestrian phrases or words that have been made into a chorus by being chopped, processed and generally mucked about with.
So if you have a great verse but can't nail that chorus, just record anything and play around with it. After all, what have you got to lose?
If you've chopped up your vocal and loaded it into a sampler, try applying some octave-wide pitchbend.
51
Whenever you're using chopped-up, pitch-effected vocals, it's always worth trying a bit of pitchbend action, particularly on held notes. Set the pitchbend range to 12 semitones, play the vocal back and sweep it up and down. Naturally, for this to work, you need a very powerful Auto-Tune-style device, or to chop up the phrase into individual words in a sampler - and of course, there's nothing to stop you using both!
52
Auto-tuning devices are powerful tools, and they're also very versatile. If you tend to work with very fast response time settings, try making them slower for a natural effect. If you normally go slow, try some faster, T-Pain-style changes. And don't forget that you can automate the effects throughout the track. Sometimes, less is more.
53
When asking your singer to do double takes of a vocal, it's worth having at least one running in the background as a guide - otherwise, if you decide to layer them, the differences in timing can make the vocals sound disjointed. Plug-in tools such as VocALign can do a great job of matching the timing of vocals, but it's always easier to get it right from the start.
54
If you want some variation on your backing vocals but only have one singer, try doing them yourself. Usually even the most tone-deaf person can sing monotone, and you'll be amazed how much a monotone version of a vocal, on the root note of each passage, can thicken up and add interest to a vocal part.
55
It's great to take chances and experiment, but if you're going for a specific type of track, chances are there will be a specific type of processing that goes best with it. Spend a bit of time listening to the types of effects that are used in comparable tracks - don't just use the same old effects on every vocal.
When using voice synths for harmonies, keep things simple to avoid an voerly Kraftwerk-like vocal sound.
56
When using harmonising plug-ins such as Neptune's voice synthesizer, it's usually best just to use one note at a time for energetic pop. Playing chords will have the effect of making your track sound like a choir or acapella group, or like an old Kraftwerk-style vocoded vocal. Single-note harmonies sound like reinforcing backing vocals.
57
If you haven't got many backing vocals - or sometimes even if you do - it's often worth finding one key word or phrase that you can repeat. Ideally this will have some resonance with the rest of the song, but it needn't. The key is to make sure it sounds punchy and catchy. More often than not, you'll find that this is the listener's favourite bit.
58
When processing your vocals, it's tempting to bounce them down with the de-essing and breath removal effects applied, just to save some CPU cycles. Fair enough, but keep in mind that you might want to change these later, as when other effects such as compression and EQ are applied, they might disappear or get accentuated further. Wait until you're sure before doing anything that you can't undo.
59
Sometimes a singer just isn't in the mood. Learn to recognise the signs, and suggest a break or a rain check. Don't get us wrong - if your singer spends more time sulking than singing then it might be time to look elsewhere, but vocalists can often be quite highly strung (it takes a certain type of person to get on stage and sing), so it helps if you know how to placate them..
60
If you're using any kind of Auto-Tune/Neptune-esque device, keep an eye out for a vibrato control. Subtle amounts will sound natural, heavy amounts not so much, but either one can sound fantastic if used in the right way and the right context. As with all types of processing, we suggest trying automation to switch it on and off in different places.
Add some excitement to a track by using ascending harmonies and a harmonising device like Record's Neptune.
61
A great way to get anticipation and excitement to a build in a track - when moving from verse to chorus, for example - is to layer some ascending harmonies. Pick a good harmony note, then while the main vocal stays on one level, take the harmonies up in musically fitting increments (these will depend on the track, of course). If your chorus is in a different key, try making your last harmony note meet the root note of the chorus.
62
Pay careful attention to the order of your effects. It seems obvious, but if you have a compressor before a de-esser or a breath removal plug-in, it will make it much harder for that plug-in to distinguish what is meant to be there and what isn't. This is particularly important when dealing with plug-ins that rely on dynamics, but it applies to all effects.
63
Sometimes the oldest tricks are the best, and slap-back delay has been around for quite a while. Use a very small delay time, with no feedback, and you'll get a thickening sound that's reminiscent of the best of The Beatles. Unlike a lot of vocal effects, this can often be used on the main vocal without it losing its grounding and punch.
64
As an alternative to thickening your vocal track by layering on another sound and treating them as one, try playing a musical riff that follows your vocal. It needn't be prominent in the mix. If you have a tool that will create a MIDI file from your vocal, try doing that. It will have a load of smaller notes from the slides in it, though - you'll need to ditch them and stick to the core notes.
65
Compression is key with vocals. If you want your vocal to cut through a mix, you'll need some level of compression to get consistent weight - how much depends on the track. A soft vocal needs some dynamic range, but if your singer was belting it over a shredding backing track, don't be afraid to push the compression all the way up to limiting.
Delay makes a great reverb subsitute (or adjunct) as, unlike reverb, it needs not completely fill out your mix.
66
Reverb is a great tool for giving extra space and width to vocals, but it can also be a very dominating effect. As an alternative option, try some subtle delay. The key to not overpowering your vocal is to ensure that the delays are EQ'd - if your delay doesn't have EQ, put it on a bus and apply some EQ from there. A little bit of top- and bottom-end cut will thin it out and sound like reverb, but without the muddying tail.
67
Don't just use sidechaining to help your vocal pop through. If you have a musical part (say a piano or guitar) that's playing the same notes as the vocal, try using a sidechain gate on your vocal, taking its input from the musical part, to chop your vocal with the music. It's like our layering suggestion, only more obvious and extreme.
68
When you are making a vocal track, make sure the vocal comes to the fore and is the most prominent part of the mix - this usually means a certain amount of level. Singers always want their vocals louder, while the producer is often focused on the music. The genius is usually somewhere in the middle. Listen to other songs on your studio speakers and we reckon you'll be surprised just how loud the vocals are in most of them.
69
If you don't have a vocalist but want to get used to working with vocals, try experimenting with acappellas. It will give you some pointers about recording good vocals, even if only through observation, and it will definitely teach you a lot about placing vocals and making the most of what you have.
70
Timing is everything. If your vocal lacks energy, try applying a little pre-delay to it. Or better yet, chop up your vocal and add delay or pre-delay to certain parts, to create a tailored groove. And try using a bigger amount than with percussion - generally 1 to 3ms will be very noticeable with drums, but with vocals, you're going to need to hit about 3 to 5ms or more to even notice the difference.
De-breathing plug-ins are a great quick fix, but manually editing the breaths can be much more effective and creative.
71
De-breathing plug-ins can be fantastic for reducing the overall level of breath noise, but for real punch and effect, it's worth chopping your vocal parts manually, taking out certain breaths or finishing words in a more clipped way. Fit your vocal to the melody and groove of the music.
72
We often talk about using sidechain pumping on a bassline to give it groove and fit it into the mix better, and the same thing can be done with a vocal part. How much bounce you add depends entirely on the track and the vocal, but even when used subtly, it can pull a vocal right into the track without it disappearing into the mix entirely.
73
You might want to try using glitch effects on certain parts of a vocal for emphasis. There are a couple of ways to do this. Sometimes we like to simply bounce a version of our vocal with our mangling plug-in of choice and use the parts that work; other times we automate the wet/dry balance. Either way, it's often also worth trying to mix some of the original signal over the results.
74
If you have four versions of the same chorus line, try taking the last two, layering them under the first two and then repeating both for the second half of eight bars. Not only will this generate a nice chorusing effect, but it will also create a subtle repeating variation, giving the listener something to latch onto and keeping them from feeling like things are on a static loop.
75
As we've said before, you should always get three takes of a vocal. Try offsetting your third vocal in relation to the main vocal, perhaps a half-bar or a bar later. Now let the vocals play all the way through and wherever the extra vocal sounds great filling in the gaps, leave it in. Then just strip out all the other parts you aren't using.
Dropping your sidechain kick on a vocal can help certain words or phrases stand out, particularly with chopped vocals.
76
If you are using ducking on your vocal, try taking the sidechain off for key phrases to help the vocal stand out (if it's set up properly you can make it get louder when the sidechain kick is removed). This can be particularly useful when you have some rhythmically chopped vocals, or harmonies that are bouncing off each other.
77
Vocals have a lot of messy frequencies in them, particularly if you've recorded them yourself. Get used to trimming out all the bottom end you don't need with EQ. Just be sure not to take the weight out of the voice. You want it to sound natural, but not like the deep, soothing baritone-style vocal you get on talk radio.
78
Rather than boosting your vocal with EQ to make it pop through the mix, try cutting out some of the corresponding frequencies in the competing parts (guitars and synths are the usual culprits). We can't stress enough that boosting should be done in moderation to add a little edge - cut should always be your first port of call when trouble-shooting frequency issues.
79
As you'll no doubt be aware, recording and comping vocals can be a pretty messy and confusing business on the screen. For this reason it's always a good idea to bounce down an instrumental track and place it in a fresh project before you start. This makes recording easier and more responsive (no latency or glitching) and makes it far easier to comp the vocals, ready to import into your actual project file.
80
Parallel compression can work really well with vocals. It preserves the natural rhythm and expression, but still lets you fatten things up a little. Or if you have two layered vocals, try compressing one of them heavily, while leaving the other quite natural.
Bit-crushing plug-ins are ideal for adding presence and bite to any part of a mix, and vocals are no exception.
81
One way to give your vocal some sparkle and presence without having to add EQ (which can just add to the mess), is to use a bit-crushing plug-in such as the free TAL-Bitcrusher. Whether you apply this with extreme settings or just use a smidgen is up to you, as is whether you apply it to a solo vocal line, layer or entire group.
82
If you have a multiband compressor that allows an external sidechain, try making only some of the frequency range pump rhythmically. You can preserve the top end of the vocal and still create the impression of pumping - this tends to keep vocals audible in the mix while giving things bounce.
83
Many vocalists find it easier to hit the right notes if they sing powerfully. If your singer is regularly missing notes, try asking them to sing with more force. Just remember that this will tire out their voice much faster, and none of the subsequent takes are likely to match the early ones, so it is generally best saved until towards the end of a session.
84
Reverb fills out a vocal section and can give it much more character and life, but it can also really obscure it. Try using a sidechain compressor on your vocal's reverb (or indeed any reverb bus) to duck the reverb while the vocals are sounding. You usually want pretty fast attack times, but try to get the release to open in time with your track.
85
When you are comping vocals, it's best to be decisive. If your sequencer offers automatic comping facilities, start by making a copy of a take, and then be utterly ruthless. Don't worry about what you might need later - just get stuff you're happy with. If you find a line missing, you always have the back-up, and if you can't get it resung, you can slather on the processing to disguise the flaws!
Improve the presence of your vocal by centring any lower frequencies with an application like Sonalksis' StereoTools.
86
If you have a stereo enhancing tool with the ability to centre frequencies below a certain point, try adding an artificial stereo widener to your vocal and following it with a stereo enhancer plug-in, then setting the crossover to about 1kHz. You're aiming to keep the vocal centred and weighty, but also adding a little width to the space at the top end.
87
It sounds obvious, but make sure your singer is singing in their ideal range. If you've already made a track and they've come in to do a vocal on it, and the vocal is good but the range is a strain, try shifting the key of the track. These days this is usually easy to do (unless your track was played and recorded live), and often it can transform a track even before you've put the vocal in.
88
When you have two layered vocals, you can add some stereo spread and excitement by adding a tremolo or other rhythmic panning tool to one of the layers. Make sure it's lower in the mix, to keep things grounded, and try thinning the panned sound out a little with some EQ to make sure the track stays centred.
89
In most cases you want your vocal nicely panned, but sometimes, when the lyrics are recounting (or actually are) a conversation, it can be a good idea to pan the alternating lines from one side to the other. For an absolutely exquisite example of this, have a listen to Level by The Raconteurs , Jack White's superlative-defying side project.
90
Reverse and gated reverb are usually used on drums, but with the right timings they can add enormous impact to a vocal. You generally don't want them running all the way through a track, but judicious use can transform a lonely-sounding vocal.
Chorus and doubling effects can be used on backing vocals without taking away too much punch and presence.
91
Doubler and chorus effects are great for thickening vocals, but they also take away from the punch. Try using them on a bus or on one of two layers to ensure that at least one part of the vocal is strong and dominant. The same goes for phasing and flanging. Keep the levels on the dry side to start with, then raise them slowly during the final mix session.
92
A powerful vocal, suddenly removed, can leave a perceived empty space in the busiest of tracks. Set up a delay bus and send some of your vocal channel to it over the course of the last line before the vocal drops out, and let the delays ease the vocals out gently. Using a panner or some kind of pitch modulating effect (such as Audio Damage's Discord 3) will add further interest.
93
Sometimes a live vocal performance can sound amazing, but once you record it, you're stuck with it. To get a similar effect on a studio-recorded vocal, play it back through some speakers, then record that playback and layer it underneath your original. Since this is almost exactly what happens with live feedback, the result will be strikingly similar.
94
For some real bite on your vocals, you can't beat a bit of distortion, and almost every DAW now comes with some kind of guitar pedal and amp simulator, so make use of them. The trick, as ever, is to strike the right balance between bite and intelligibility - unless you just want a mangled vocal, in which case, go for your life!
95
Whenever you are making a track with vocals, it automatically becomes a 'song', and when producing songs, it's almost always a good idea to start with the radio edit. This flies in the face of many electronic producers' instincts, but it is much easier to spin out a good club cut from a radio edit than working the other way around. More importantly, it helps you concentrate on the power of your vocal.
96
All these tips are guidelines, and there's no need to think that you should apply all, or any, of them on every occasion. Rules are there for a reason, but they are made to be broken, and at least half the pearls of wisdom here were discovered by accident - indeed, many of them were heresy when they first arrived on the scene.
Don't be afraid to ignore all of our advice if you think your stuff sounds great - just make sure it sounds great to other people too!
We generally recommend that if you're aiming for really pro-sounding vocals it's easier to have a real preamp on the go, keeping good signal levels on the way in. If you haven't got access to one of these, however, you can always try throwing in something that provides the same sort of flavour. Plug-ins such as PSP's Mix Saturator 2 offer some handy preamp settings that quite convincingly mimic outboard saturation and compression.
02
It's vital to get the best mic position for your vocalist. While it's usually best to start from right in front of the mic, some people's voices may sound better a little closer or further away, or even to the side. If your singer's voice has a nasal quality, for example, you might want to place the mic slightly above their nose rather than in front of or below it. Experiment until you find the sweet spot.
03
Many vocalists sing better when they tilt their head back a little, as it opens up their vocal chords. The easiest way to take advantage of this is to place the mic slightly higher than the singer's mouth, angled down towards them - this will encourage them to tilt their head back as they sing.
04
It might sound a little obvious, but before doing a vocal take, it's a good idea to make sure your singer has a clear idea of where the lines should be placed. There's nothing wrong with having them scribbled down on a piece of paper, but generally speaking, a singer who doesn't know the lyrics probably isn't that sure how to sing them either.
05
Make sure you keep track of your singer's position (even if your singer is you). You will often find yourself taking breaks, or coming back to re-record a line, and it's very important to make sure the vocals are coming from the same position each time - if not, they might not sound the same. It goes without saying that you should keep track of where the mic was, too.
Artificial stereo spread effects can push a vocal back in the mix, so try to keep them for backing or quieter layered vocals.
06
Adding an artificial stereo spread effect to a vocal can be a great way of giving it real presence, but it also has the effect of taking away some of the vocal's punch and causing it to lose some of its prominence in a mix, so is often not ideal on the lead vocal. Instead, try using it for backing vocals or layered words, or for specific effect on certain words.
07
Chorus effects are great, but one of the best ways to achieve chorusing is to actually record multiple takes of the vocal. You can then take a few of these and layer them - try panning a couple slightly and leaving one centred, or thinning out the underlying two a little with EQ and more heavily compressing the central one for added impact.
08
You can thicken up a thin-sounding vocal by using a synth layered underneath it that subtly plays the same notes. You can even use a sidechained gate on the synth, triggered by the vocal, to ensure they only play together. Alternatively, try using something like Waves Tune, which will export the notes of your vocal as a MIDI file. Then compress the synth and vocal as one.
09
There are plug-ins out there, like Antares Punch, which are designed to accentuate the more punchy parts of a vocal, but you can also achieve something similar using gating or even an expander on your compressor. With a gate, for example, you just need to set it up so that only the punchy parts of your vocal open the gate, and then apply a small amount of gain reduction.
10
One of the great ways to keep a vocal punchy but still interesting is to use some of your additional takes to layer key words. These could be words that are significant in a lyrical sense, or that work well with the groove, or that enhance the arrangement. Just be sure not to overdo it - use them sparingly enough that they keep their impact when they appear.
Try chopping your vocal up and loading the parts into a sampler, then applying some groove quantise to them.
11
We often talk about adding groove quantise to musical and rhythmic parts, but it can also be very effective on a vocal line. How you do this is up to you - Live users might try putting warp markers on the individual words of the track and then applying quantise, but we like putting the vocal in a sampler so the quantise can affect the length of the words as well as their placement.
12
A vocal line can sound a little jarring when it first appears, so try introducing it by taking the first word, reversing it and playing it into the normal version. You can use this effect anywhere, in fact - and don't forget that you can also timestretch the length of the reversed part to fit the space. Reverse reverb within your vocal can have a similar effect.
13
Once you've spent ages carefully ridding your vocal track of background hum, the cha-chinks of ice in a glass, coughs, comments and all the other bits of assorted noise that you only notice on a vocal track once it's been pumped with compression, it's often easy to forget how cool some of them can sound when sequenced into your track. Try it!
14
The easiest way to get rid of breath noise is to use a gate. You want to set it up very carefully to make sure you aren't losing any important parts of the words - a useful rule of thumb is to keep your attack and release times relatively slow. When in doubt, take it easy and manually remove any problem parts. There are de-breathing plug-ins that can help, too.
15
Breath noise can be problematic, but it's also vital on some vocals - completely removing it can make a vocal sound artificial and over-processed. Sometimes the words and breaths will slide between each other, so make sure you set any gain reduction carefully to leave some of the vocal in the track.
Whether manually or with a plug-in like Vocal Rider, try using old-fashioned volume controls to balance your vocal.
16
We often tell you to compress a vocal to achieve a balanced level, but before you do that, you want to ensure that the volume is level too. What's the difference? Well, compressing squashes the signal and makes it more powerful, but it also changes the overall character of the sound. Try riding the fader manually, changing the levels of individual word clips or using something like Waves' Vocal Rider to achieve a more natural-sounding result.
17
When recording a vocal, have your lyrics written out on a sheet and have three different- coloured pens to hand. When the vocalist sings a part of the vocal right, draw a line through that part with a pen. When they sing the same part well again on a subsequent take, strike the part through with a different colour. Don't stop recording until you have all three colours through every line!
18
Sometimes when you're on a deadline, have been looking forward to recording a vocal or have been planning it for ages, you might be tempted to go ahead with a recording even if your singer has a cold or otherwise isn't in the best condition. Check with the vocalist first - they know their voice, and they'll be able to make the best assessment of whether they're up to the job. Singing with a cold or sore throat is at best pointless, and at worst harmful. You'll all get frustrated, and waste your time.
19
We think it's always worth trying a vocal in a different key, just to see how it sounds. It can be very easy to get stuck in your ways, but it only takes a minute to sing a line in a different key. You might decide it sounds horrible, but you could also be pleasantly surprised. You might even decide to put a cheesy (sorry, 'epic') key change in the track.
20
You can add a lot of extra excitement to key bars or phrases in your track by adding some panning effects. This works on anything, but with vocals, it's ideal used on backing vocals or double-tracked words. The strength and speed of the panning are entirely up to you, but one thing that's often worth doing is softening the extremity of the pan.
Applying a little creative processing on key lines of a vocal is a great way to add emphasis and effect.
21
A lot of vocal processing is about achieving a consistent sound throughout, but that doesn't mean you can't accentuate some key lines. Try using some extreme processing - for example, you can put one phrase through an EQ and make it sound like it's coming down a phone line (cut below 300Hz and above 3kHz), to make it stand out.
22
To make a vocal sound more intimate, use very strong compression. The sounds you are trying to pull out are the breath and mouth noise of your vocal, as these are the sounds we associate with somebody speaking very close to our ears. With this in mind, it's also a good idea to keep your vocals quite dry too, for a more natural sound.
23
Remember that your vocal is usually not just another part of your track - it is your track. There will be musical elements that people like, and riffs that people love, but it's best to treat your vocal as the hook in almost every instance, as that is quite simply how most other people will view it.
24
Processing is great for adding emphasis to a vocal, but you can add interest by using some real gear too. For example, try recording a version of your vocal into an iPhone (you can set this up to run at the same time as you're capturing the vocal with a proper mic), or you might consider recording through a megaphone or guitar amp.
25
Try capturing vocals in different places. Generally dry is good, but if you have a great echoing hallway, or even a roof or a field… why not? When recording outside your main enemy will be wind, so you'll need either wind baffles or a foam cover, but if you want a completely dry vocal, free from reverb and reflection, it's hard to beat the middle of a field on a misty morning!
Singing slightly flat is a common problem among vocalists, so we often pitchshift the whole part up a few cents.
26
Most vocalists you are likely to encounter will sing a little flat rather than sharp, and this can often sound quite noticeable and unpleasant. One quick fix for a lot of vocal parts is simply to nudge the whole thing up by 5 to 25 cents. You'll be amazed how many vocals this will help.
27
Try different mics, if you have them, including cheap and old ones. Obviously if you're capturing an opera singer's voice, this could be considered something of a waste of their gift, but sometimes the character and flaws of the mic can really add something, even if only used as a layering effect.
28
While it is true that you can get awesome results with a relatively cheap microphone, good singers deserve a decent mic, and one that suits them. If you have access to a few different mics, try them out on different singers and you'll often hear little plosives or mouth noises that aren't audible on other mics. When aiming for expensive-sounding vocals, get a mic that suits your singer.
29
With automation available on almost everything in the studio now, it's incredibly easy to become a control freak with delay and other processing - so go for it! We like adding some delay to a few key words or parts of our vocal to create excitement. It can work with reverb and other effects too - and for added impact, try those effects on a bus and add an auto-panner.
30
Vocoded vocals can be quite hard to make intelligible. There are any number of more modern plug-ins out there now that are designed to make things clearer, and they usually do a great job, but it can still be really useful to ask your vocalist to enunciate clearly and to try to put some unnatural emphasis on the plosives and individual syllables.
Giving your vocalist a little reverb in their headphones will help them tell whether they are singing in key.
31
You should always try to give your singer some reverb in their cans while they're singing (unless they don't want it). The idea Giving your vocalist a little reverb in their headphones will help them tell whether they are singing in key is to keep their confidence high by giving them a good impression of their own voice, and to give them something they can pitch to.
Your vocalist can't listen and sing at the same time - at least, not like the engineer can listen - but with reverb, they can hear the pitch.
32
If you have the opportunity, and know you are working with a great singer, leave the mic running while they warm up (after checking with them that it's OK to do so). You could strike gold, even if you only catch a little phrase to drop in at the start of a track (think Amy Winehouse). Failing that, you might get some great archive material to use when VH1 run a special show to commemorate the 30th anniversary of your first album…
33
For a thicker vocal sound, don't just add reverb and delay on busses. Try adding them directly to the channel and then following that with some compression to really pull the elements together. Slap-back delay (with very fast delay times) is a perfect candidate for this kind of thing. It's also important to keep release and feedback times short to avoid making a mess.
34
You should try to have one set of headphones for the singer, and another set for you to listen out for any problems on the recording. It's far too easy, when you're just listening to a singer without cans, to miss a little bump from a ring or button on the mic stand that can completely ruin a recorded phrase.
35
No matter how carefully you've set your initial input level, singers will sometimes stray into louder territory than you planned, so even if you've tested their loudest note and left ample headroom, keep a stern eye on the incoming level to watch for clipping.
Grouping vocals helps pull different takes together and makes it easier to create separation between sections.
36
It's usually worth grouping the various parts of your vocal track so that you have one group (or channel) for the lead vox, one for doubled words and breath effects and one for the backing vocals, and then adding compression. This will pull all the vocal parts together nicely and make them seem more coherent, and will make it easier for the listener to differentiate the vocals from other elements of the track.
37
Generally speaking, the place for a vocal is in the centre of a mix. It sounds obvious, but listen to almost any great rock or pop recording and you'll hear that while the instruments might be dotted all over the place, the vocals are usually centred, and it's no accident. The vocal is normally the key element, and centring it makes it, well, central. We're also used to people speaking to us from the front, so it feels more personal.
38
Be very wary of anything that makes noise when recording. Even quiet noise will be amplified, including the slightest rustle of paper - this is why it's handy for the vocalist to have the lyrics memorised. Other culprits include rattling bracelets, necklaces and change or keys in pockets. Get rid of them before you start!
39
It's a good idea to always have a dictaphone or similar portable recording device ready. You never know when inspiration might strike, and even in a studio full of equipment, nothing quite beats the immediacy of hitting one button for instant recording. It can be digital, analogue… anything! They key is having instant, reliable recording at the touch of a button.
40
Although various kinds of, ahem, stimulants have been considered beneficial to the creative process by many a great musician, generally speaking, people sing better sober. Stick to a nice cup of tea - it will warm the singer's throat and wake them up. Try to steer clear of milk, though, as it acts as a coagulant in the throat and can have a really adverse effect on a singer's voice.
Chopped-up vocals will almost always sound better with a little octave-wide pitchshifting on individual words.
41
When chopping up parts of your vocal line and pitchshifting them, it's often hard to know where to begin. A good place to start is shifting an octave up. Alternatively, play a scale and pick notes that work in quick sequence. And whenever you repeat a syllable or vocal noise a few times, try shifting one up or down an octave. Experiment with the pitchbend, too.
42
It's usually worth grouping the various parts of your vocal track so that you have one group (or channel) for the lead vox, one for doubled words and breath effects and one for the backing vocals, and then adding compression. This will pull all the vocal parts together nicely and make them seem more coherent, and will make it easier for the listener to differentiate the vocals from other elements of the track.
43
One of the best ways to get vocals to cut through a mix is to use sidechain compression on competing musical parts. Instead of turning up the vocal, insert a compressor on the competing parts and let your vocal control the sidechain to duck the music. Don't overdo it - you shouldn't hear the music pump, but with careful attack and release times your vocal will magically appear in the mix.
44
One of the most common sources of noise on a vocal recording is the backing track itself, bleeding out of the headphones. One simple way to avoid this is to use closed over-the-ear headphones. If your singer likes to have only one cup over their ear, try panning the headphone signal so it's not bleeding into the mic.
45
If you don't have access to a singer but need to get a vocal down, you could try using a voice-altering plug-in, speech box or vocoder and recording your own version. You might be pleasantly surprised at how often you can come up with interesting, if classically imperfect, results.
Sidechain compression can be used to help individual words or phrases cut through the main vocal.
46
Sidechaining isn't just a good way to get your vocal to cut through a mix, it's also a good way to get a vocal to cut through… your other vocals. You might want, for example, one bit of backing vocal - or perhaps a bit of a spoken word or vocal noodling - to cut through. Apply a sidechain-enabled compressor to your main vocal group, and then use the one you want to stand out to duck it (just make sure it's not in the same group, ducking itself!).
47
We've all been here before: you're comping and editing a vocal, and you realise you haven't followed your own advice about keeping track of the vocal takes to make sure you've got a decent version of every line. So you're tweaking away, trying to get that one word to sound right. Stop. Call the singer and ask if they can either come into the studio to redo it, or record a new take and send it over.
48
When writing a song, get a guide vocal down as soon as possible - it needn't be the definitive take, but it will help you get on with constructing the whole song, which will provide you with clues about how to handle your vocals. Quite often, though, you'll capture something nice and raw - it's surprising how often a rough take becomes the final version.
49
Break up your vocal by using a doubled layer at the end of every verse or half-verse. If you've already double-tracked your vocals, this will mean adding another layer - one of the reasons why we suggest always getting at least three takes of a vocal. When recording, try getting a whole take sung as a harmony, ready to place wherever you need it.
50
This tip may raise the ire of the traditional songwriters out there, but in this day and age, the chorus of your song doesn't need to sound that great in its raw form. Modern music (pop, soundtrack and club alike) is replete with examples of pedestrian phrases or words that have been made into a chorus by being chopped, processed and generally mucked about with.
So if you have a great verse but can't nail that chorus, just record anything and play around with it. After all, what have you got to lose?
If you've chopped up your vocal and loaded it into a sampler, try applying some octave-wide pitchbend.
51
Whenever you're using chopped-up, pitch-effected vocals, it's always worth trying a bit of pitchbend action, particularly on held notes. Set the pitchbend range to 12 semitones, play the vocal back and sweep it up and down. Naturally, for this to work, you need a very powerful Auto-Tune-style device, or to chop up the phrase into individual words in a sampler - and of course, there's nothing to stop you using both!
52
Auto-tuning devices are powerful tools, and they're also very versatile. If you tend to work with very fast response time settings, try making them slower for a natural effect. If you normally go slow, try some faster, T-Pain-style changes. And don't forget that you can automate the effects throughout the track. Sometimes, less is more.
53
When asking your singer to do double takes of a vocal, it's worth having at least one running in the background as a guide - otherwise, if you decide to layer them, the differences in timing can make the vocals sound disjointed. Plug-in tools such as VocALign can do a great job of matching the timing of vocals, but it's always easier to get it right from the start.
54
If you want some variation on your backing vocals but only have one singer, try doing them yourself. Usually even the most tone-deaf person can sing monotone, and you'll be amazed how much a monotone version of a vocal, on the root note of each passage, can thicken up and add interest to a vocal part.
55
It's great to take chances and experiment, but if you're going for a specific type of track, chances are there will be a specific type of processing that goes best with it. Spend a bit of time listening to the types of effects that are used in comparable tracks - don't just use the same old effects on every vocal.
When using voice synths for harmonies, keep things simple to avoid an voerly Kraftwerk-like vocal sound.
56
When using harmonising plug-ins such as Neptune's voice synthesizer, it's usually best just to use one note at a time for energetic pop. Playing chords will have the effect of making your track sound like a choir or acapella group, or like an old Kraftwerk-style vocoded vocal. Single-note harmonies sound like reinforcing backing vocals.
57
If you haven't got many backing vocals - or sometimes even if you do - it's often worth finding one key word or phrase that you can repeat. Ideally this will have some resonance with the rest of the song, but it needn't. The key is to make sure it sounds punchy and catchy. More often than not, you'll find that this is the listener's favourite bit.
58
When processing your vocals, it's tempting to bounce them down with the de-essing and breath removal effects applied, just to save some CPU cycles. Fair enough, but keep in mind that you might want to change these later, as when other effects such as compression and EQ are applied, they might disappear or get accentuated further. Wait until you're sure before doing anything that you can't undo.
59
Sometimes a singer just isn't in the mood. Learn to recognise the signs, and suggest a break or a rain check. Don't get us wrong - if your singer spends more time sulking than singing then it might be time to look elsewhere, but vocalists can often be quite highly strung (it takes a certain type of person to get on stage and sing), so it helps if you know how to placate them..
60
If you're using any kind of Auto-Tune/Neptune-esque device, keep an eye out for a vibrato control. Subtle amounts will sound natural, heavy amounts not so much, but either one can sound fantastic if used in the right way and the right context. As with all types of processing, we suggest trying automation to switch it on and off in different places.
Add some excitement to a track by using ascending harmonies and a harmonising device like Record's Neptune.
61
A great way to get anticipation and excitement to a build in a track - when moving from verse to chorus, for example - is to layer some ascending harmonies. Pick a good harmony note, then while the main vocal stays on one level, take the harmonies up in musically fitting increments (these will depend on the track, of course). If your chorus is in a different key, try making your last harmony note meet the root note of the chorus.
62
Pay careful attention to the order of your effects. It seems obvious, but if you have a compressor before a de-esser or a breath removal plug-in, it will make it much harder for that plug-in to distinguish what is meant to be there and what isn't. This is particularly important when dealing with plug-ins that rely on dynamics, but it applies to all effects.
63
Sometimes the oldest tricks are the best, and slap-back delay has been around for quite a while. Use a very small delay time, with no feedback, and you'll get a thickening sound that's reminiscent of the best of The Beatles. Unlike a lot of vocal effects, this can often be used on the main vocal without it losing its grounding and punch.
64
As an alternative to thickening your vocal track by layering on another sound and treating them as one, try playing a musical riff that follows your vocal. It needn't be prominent in the mix. If you have a tool that will create a MIDI file from your vocal, try doing that. It will have a load of smaller notes from the slides in it, though - you'll need to ditch them and stick to the core notes.
65
Compression is key with vocals. If you want your vocal to cut through a mix, you'll need some level of compression to get consistent weight - how much depends on the track. A soft vocal needs some dynamic range, but if your singer was belting it over a shredding backing track, don't be afraid to push the compression all the way up to limiting.
Delay makes a great reverb subsitute (or adjunct) as, unlike reverb, it needs not completely fill out your mix.
66
Reverb is a great tool for giving extra space and width to vocals, but it can also be a very dominating effect. As an alternative option, try some subtle delay. The key to not overpowering your vocal is to ensure that the delays are EQ'd - if your delay doesn't have EQ, put it on a bus and apply some EQ from there. A little bit of top- and bottom-end cut will thin it out and sound like reverb, but without the muddying tail.
67
Don't just use sidechaining to help your vocal pop through. If you have a musical part (say a piano or guitar) that's playing the same notes as the vocal, try using a sidechain gate on your vocal, taking its input from the musical part, to chop your vocal with the music. It's like our layering suggestion, only more obvious and extreme.
68
When you are making a vocal track, make sure the vocal comes to the fore and is the most prominent part of the mix - this usually means a certain amount of level. Singers always want their vocals louder, while the producer is often focused on the music. The genius is usually somewhere in the middle. Listen to other songs on your studio speakers and we reckon you'll be surprised just how loud the vocals are in most of them.
69
If you don't have a vocalist but want to get used to working with vocals, try experimenting with acappellas. It will give you some pointers about recording good vocals, even if only through observation, and it will definitely teach you a lot about placing vocals and making the most of what you have.
70
Timing is everything. If your vocal lacks energy, try applying a little pre-delay to it. Or better yet, chop up your vocal and add delay or pre-delay to certain parts, to create a tailored groove. And try using a bigger amount than with percussion - generally 1 to 3ms will be very noticeable with drums, but with vocals, you're going to need to hit about 3 to 5ms or more to even notice the difference.
De-breathing plug-ins are a great quick fix, but manually editing the breaths can be much more effective and creative.
71
De-breathing plug-ins can be fantastic for reducing the overall level of breath noise, but for real punch and effect, it's worth chopping your vocal parts manually, taking out certain breaths or finishing words in a more clipped way. Fit your vocal to the melody and groove of the music.
72
We often talk about using sidechain pumping on a bassline to give it groove and fit it into the mix better, and the same thing can be done with a vocal part. How much bounce you add depends entirely on the track and the vocal, but even when used subtly, it can pull a vocal right into the track without it disappearing into the mix entirely.
73
You might want to try using glitch effects on certain parts of a vocal for emphasis. There are a couple of ways to do this. Sometimes we like to simply bounce a version of our vocal with our mangling plug-in of choice and use the parts that work; other times we automate the wet/dry balance. Either way, it's often also worth trying to mix some of the original signal over the results.
74
If you have four versions of the same chorus line, try taking the last two, layering them under the first two and then repeating both for the second half of eight bars. Not only will this generate a nice chorusing effect, but it will also create a subtle repeating variation, giving the listener something to latch onto and keeping them from feeling like things are on a static loop.
75
As we've said before, you should always get three takes of a vocal. Try offsetting your third vocal in relation to the main vocal, perhaps a half-bar or a bar later. Now let the vocals play all the way through and wherever the extra vocal sounds great filling in the gaps, leave it in. Then just strip out all the other parts you aren't using.
Dropping your sidechain kick on a vocal can help certain words or phrases stand out, particularly with chopped vocals.
76
If you are using ducking on your vocal, try taking the sidechain off for key phrases to help the vocal stand out (if it's set up properly you can make it get louder when the sidechain kick is removed). This can be particularly useful when you have some rhythmically chopped vocals, or harmonies that are bouncing off each other.
77
Vocals have a lot of messy frequencies in them, particularly if you've recorded them yourself. Get used to trimming out all the bottom end you don't need with EQ. Just be sure not to take the weight out of the voice. You want it to sound natural, but not like the deep, soothing baritone-style vocal you get on talk radio.
78
Rather than boosting your vocal with EQ to make it pop through the mix, try cutting out some of the corresponding frequencies in the competing parts (guitars and synths are the usual culprits). We can't stress enough that boosting should be done in moderation to add a little edge - cut should always be your first port of call when trouble-shooting frequency issues.
79
As you'll no doubt be aware, recording and comping vocals can be a pretty messy and confusing business on the screen. For this reason it's always a good idea to bounce down an instrumental track and place it in a fresh project before you start. This makes recording easier and more responsive (no latency or glitching) and makes it far easier to comp the vocals, ready to import into your actual project file.
80
Parallel compression can work really well with vocals. It preserves the natural rhythm and expression, but still lets you fatten things up a little. Or if you have two layered vocals, try compressing one of them heavily, while leaving the other quite natural.
Bit-crushing plug-ins are ideal for adding presence and bite to any part of a mix, and vocals are no exception.
81
One way to give your vocal some sparkle and presence without having to add EQ (which can just add to the mess), is to use a bit-crushing plug-in such as the free TAL-Bitcrusher. Whether you apply this with extreme settings or just use a smidgen is up to you, as is whether you apply it to a solo vocal line, layer or entire group.
82
If you have a multiband compressor that allows an external sidechain, try making only some of the frequency range pump rhythmically. You can preserve the top end of the vocal and still create the impression of pumping - this tends to keep vocals audible in the mix while giving things bounce.
83
Many vocalists find it easier to hit the right notes if they sing powerfully. If your singer is regularly missing notes, try asking them to sing with more force. Just remember that this will tire out their voice much faster, and none of the subsequent takes are likely to match the early ones, so it is generally best saved until towards the end of a session.
84
Reverb fills out a vocal section and can give it much more character and life, but it can also really obscure it. Try using a sidechain compressor on your vocal's reverb (or indeed any reverb bus) to duck the reverb while the vocals are sounding. You usually want pretty fast attack times, but try to get the release to open in time with your track.
85
When you are comping vocals, it's best to be decisive. If your sequencer offers automatic comping facilities, start by making a copy of a take, and then be utterly ruthless. Don't worry about what you might need later - just get stuff you're happy with. If you find a line missing, you always have the back-up, and if you can't get it resung, you can slather on the processing to disguise the flaws!
Improve the presence of your vocal by centring any lower frequencies with an application like Sonalksis' StereoTools.
86
If you have a stereo enhancing tool with the ability to centre frequencies below a certain point, try adding an artificial stereo widener to your vocal and following it with a stereo enhancer plug-in, then setting the crossover to about 1kHz. You're aiming to keep the vocal centred and weighty, but also adding a little width to the space at the top end.
87
It sounds obvious, but make sure your singer is singing in their ideal range. If you've already made a track and they've come in to do a vocal on it, and the vocal is good but the range is a strain, try shifting the key of the track. These days this is usually easy to do (unless your track was played and recorded live), and often it can transform a track even before you've put the vocal in.
88
When you have two layered vocals, you can add some stereo spread and excitement by adding a tremolo or other rhythmic panning tool to one of the layers. Make sure it's lower in the mix, to keep things grounded, and try thinning the panned sound out a little with some EQ to make sure the track stays centred.
89
In most cases you want your vocal nicely panned, but sometimes, when the lyrics are recounting (or actually are) a conversation, it can be a good idea to pan the alternating lines from one side to the other. For an absolutely exquisite example of this, have a listen to Level by The Raconteurs , Jack White's superlative-defying side project.
90
Reverse and gated reverb are usually used on drums, but with the right timings they can add enormous impact to a vocal. You generally don't want them running all the way through a track, but judicious use can transform a lonely-sounding vocal.
Chorus and doubling effects can be used on backing vocals without taking away too much punch and presence.
91
Doubler and chorus effects are great for thickening vocals, but they also take away from the punch. Try using them on a bus or on one of two layers to ensure that at least one part of the vocal is strong and dominant. The same goes for phasing and flanging. Keep the levels on the dry side to start with, then raise them slowly during the final mix session.
92
A powerful vocal, suddenly removed, can leave a perceived empty space in the busiest of tracks. Set up a delay bus and send some of your vocal channel to it over the course of the last line before the vocal drops out, and let the delays ease the vocals out gently. Using a panner or some kind of pitch modulating effect (such as Audio Damage's Discord 3) will add further interest.
93
Sometimes a live vocal performance can sound amazing, but once you record it, you're stuck with it. To get a similar effect on a studio-recorded vocal, play it back through some speakers, then record that playback and layer it underneath your original. Since this is almost exactly what happens with live feedback, the result will be strikingly similar.
94
For some real bite on your vocals, you can't beat a bit of distortion, and almost every DAW now comes with some kind of guitar pedal and amp simulator, so make use of them. The trick, as ever, is to strike the right balance between bite and intelligibility - unless you just want a mangled vocal, in which case, go for your life!
95
Whenever you are making a track with vocals, it automatically becomes a 'song', and when producing songs, it's almost always a good idea to start with the radio edit. This flies in the face of many electronic producers' instincts, but it is much easier to spin out a good club cut from a radio edit than working the other way around. More importantly, it helps you concentrate on the power of your vocal.
96
All these tips are guidelines, and there's no need to think that you should apply all, or any, of them on every occasion. Rules are there for a reason, but they are made to be broken, and at least half the pearls of wisdom here were discovered by accident - indeed, many of them were heresy when they first arrived on the scene.
Don't be afraid to ignore all of our advice if you think your stuff sounds great - just make sure it sounds great to other people too!
Wednesday, 24 August 2016
Recording The Bass Amp
Using a Beta 52 and SM 57 on the bass cabinet
Today everyone is conditioned to go direct with the bass guitar that many times miking a bass amp is completely overlooked. That's too bad because it can bring something to the track that you just can't get any other way. Here's an excerpt from my Audio Recording Basic Training book that provides an exercise for bass amp miking.
"Back in the 60's and 70's, the way engineers recorded the electric bass was by miking the bass amp. As direct boxes became more and more available, the trend eventually swung the other way, with most bass recording done direct. Today it's very common to record a bass using a combination of both an amp and direct, which provides the best of both worlds. While the bass will sound full and warm with a direct box, the amp can add just enough edge to help the bass punch through a mix.
When using a direct box, be aware that they're not all created equal in that some will not give you the low fundamental of the bass that you expect when recording this way. Active DIs do a better job at this than passive, although some passive boxes (like the ones made by Radial) do an excellent job because of the large Jensen transformer used in the circuit.
Depending on the sound that fits the track best, mix the amp track with a DI track. The sound will change substantially depending upon the balance of the DI and miked amplifier.
ALWAYS check the phase relationship between the amp and DI to make sure there's no cancellation of the low end. Flip the polarity switch to the position that has the most bottom.
Also remember that there's no rule that says that you have to use both tracks, so don't hesitate to use just a single track if it sounds best in the mix.
E7.4: Miking The Bass Amp
A) Listen closely to the amp as the bass player plays. If there are multiple speakers, find the one that sounds the best as in E7.1A.
B) Place a large diaphragm dynamic mic like D-112, RE-20 or Beta 52 a little off-center and a couple of inches away from a cone of the best sounding speaker in the bass cabinet.
C) Move the mic across the cone. Is there a spot where it sounds particularly good? Keep the mic at that spot. Is the sound balanced frequency response-wise? Can you hear any of the room reflections?
D) Move the mic towards the end of the cone? Is there more low end? Is it more distinct sounding?
E) Move the mic towards the center of the speaker? Is there more low end? Is it more distinct sounding?
F) Move the mic about a feet away from the speaker. Is there more low end? Is it more distinct sounding?
G) Move the mic about 2 feet away from the speaker. Is there more low end? Is it more distinct sounding? Can you hear more of the room? Does it work with the rest of the instruments?
H) Raise the cabinet about a foot off the floor. Is there more low end? Is it more distinct sounding?
I) I) Place the mic where it gives you the best balance of body and definition, and balance between the direct and ambient room sound."
---------
Saturday, 13 August 2016
Unorthodox Recording & Production Techniques
Guitar players are used to hearing guitars on their recordings much like the way they were heard in the room. Professional studios have recording of these sounds down to a science.
With the use of modern recording techniques, any sound desired by a player can be made audible. This however, wasn't always the case. In the ‘50s, ‘60s, and ‘70s guitarists and studio engineers had to get creative” with their production. This is a guide to the unorthodox recording techniques used in many famous recordings. In tandem with today's technology, these techniques can yield a wide variety of sounds, and can open up a new palate of tones.
Microphone Placement
Microphone placement is key to getting the desired sound out of a guitar. Today, most studios will use a Royer 121 ribbon microphone as well as a good dynamic microphone such as the Shure SM57 These two microphones could not be more different in terms of tonal quality, construction, and price. But when blended together, they produce a full tone that sounds excellent on recording.
Not many people would put a condenser microphone behind an open-backed speaker cabinet. Because a speaker cone travels in two directions, a microphone placed behind a cabinet will record a sound. The tone of this placement will be noticeably "boomier" and "bassier" with less articulation, brightness, and high-order harmonics. This tone can be used on its own for a more atmospheric effect, or blended with a microphone input from the front of the speaker for a very full sound. When using this blending technique, it is important to remember to reverse the phasing on one of the microphones to compensate for the speaker cones traveling in opposite directions.
Another unorthodox microphone placement is far out in front of the amplifier to capture the sound of the room. Because condenser microphones often have a cartioid polar pattern, they will capture sounds from many directions. They are also very sensitive, so lower-amplitude frequencies will be more audible. It is important to remember that this placement will not give you the brightness and treble response of a microphone on the speaker cone. It will, however, give your recorded tone a natural reverb because it will pick up the reverb from the room.
Tape Machines
Any audiophile will be able to tell you that analog tape sounds very different from digital files. Analog tape, in general, is noticeably warmer than a digital recording. Additionally, the distortion achieved when an analog tape machine clips the signal is not as harsh as digital clipping. Analog tape distortion can add a vintage flair to vocals and guitars (Think Aretha Franklin or Janis Joplin vocals).
Recording to analog tape is horribly inconvenient. In the world of editing, digital is far and away superior. For those looking to recreate the tone of an analog tape on their digital recording, there is a solution. Digital signals can be run onto an analog tape machine and then run back to the computer. One advantage to doing this is that once the perfect take is caught, the unreliability of magnetic tape will not be a problem. The level of the signal sent to the tape can be fine-tuned to get the perfect amount of tape distortion, and the warmer signal may be more pleasing to the ears of musicians who prefer the vintage vibe for their sound.
Do your research before investing in a reel-to-reel tape machine to be sure that it is compatible with your recording setup, and that it is operating at a useable capacity. Analog Tape machines are not mass produced as they were before music recording switched to digital formats, so a vintage tape machine may be your only option.
Unorthodox Distortion
While we're on the topic of distortion, guitar amplifiers can be used to distort instruments that we are not used to hearing distorted. One famous but often unrecognized example of this technique is with Justin Timberlake's vocal track during the bridge of his 2006 hit "SexyBack". Timberlake's vocals were recorded through a guitar amplifier to deliver a fuzzy, grimy tone. Jack White has also been known to deliver dirty vocals through a guitar amplifier using a harmonica microphone. Guitar amplifiers can be used for nearly any instrument.
Like the analog tape machine, audio tracks can be sent out to a guitar amplifier and returned to the digital audio workstation via the microphone.
Don't have a guitar amplifier handy? No problem, try any of these software plugins that simulate amps
Playing with Voltage
It is common knowledge that Jimi Hendrix would run his guitar pedals on dying batteries because the reduced voltage would produce a warmer tone. In the same way, Eddie Van Halen would run his Marshall stacks at a lower voltage to achieve more distortion. By reducing voltage, we reduce the efficiency at which our gear is operating, producing more distortion. I would not recommend imitating Eddie Van Halen's techniques for the safety of you and your gear. Pedals however, are a different story. Danelectro produces a power supply called the Danelectrode which can vary voltage from 3-9 volts. This can yield a wide variety of tones for your recording or live performance. I would not suggest using this device on digital pedals because they will not operate properly, or may not operate at all. Overdrive, Fuzz, Distortion, and analog modulation and delay pedals with dropped voltage may give you the vintage tones which you have been searching for.
Playing with Time
Recording a song faster than performance tempo and slowing it down to drop the pitch used to be quite common in the 1970s. This technique can be heard on many Led Zeppelin recordings, including Hey, Hey, What Can I Do.” The opposite of this was done to Bruce Springsteen's 1980 hit, Hungry Heart.” On Hey Hey What Can I Do,” Plant's vocals were recorded separately from the band, so Plant's voice sounds comparable to his voice on other Led Zeppelin recordings. On Hungry Heart,” however, Springsteen's voice was sped up with the rest of the song, giving it a brighter timbre. Most digital audio workstations will allow you to play with time and pitch independent from one another. When one is dependent on the other, however, new, old, and interesting sounds can be achieved.
At the end of the day, your goal should be creating the sound that's best for you. Given the myriad of recording techniques at your fingertips, this article should only scratch the surface of all possibilities. Be sure to equip yourself with all the proper gear, and you'll be inventing a sound that's unique, and specific to your music style.
With the use of modern recording techniques, any sound desired by a player can be made audible. This however, wasn't always the case. In the ‘50s, ‘60s, and ‘70s guitarists and studio engineers had to get creative” with their production. This is a guide to the unorthodox recording techniques used in many famous recordings. In tandem with today's technology, these techniques can yield a wide variety of sounds, and can open up a new palate of tones.
Microphone Placement
Microphone placement is key to getting the desired sound out of a guitar. Today, most studios will use a Royer 121 ribbon microphone as well as a good dynamic microphone such as the Shure SM57 These two microphones could not be more different in terms of tonal quality, construction, and price. But when blended together, they produce a full tone that sounds excellent on recording.
Not many people would put a condenser microphone behind an open-backed speaker cabinet. Because a speaker cone travels in two directions, a microphone placed behind a cabinet will record a sound. The tone of this placement will be noticeably "boomier" and "bassier" with less articulation, brightness, and high-order harmonics. This tone can be used on its own for a more atmospheric effect, or blended with a microphone input from the front of the speaker for a very full sound. When using this blending technique, it is important to remember to reverse the phasing on one of the microphones to compensate for the speaker cones traveling in opposite directions.
Another unorthodox microphone placement is far out in front of the amplifier to capture the sound of the room. Because condenser microphones often have a cartioid polar pattern, they will capture sounds from many directions. They are also very sensitive, so lower-amplitude frequencies will be more audible. It is important to remember that this placement will not give you the brightness and treble response of a microphone on the speaker cone. It will, however, give your recorded tone a natural reverb because it will pick up the reverb from the room.
Tape Machines
Any audiophile will be able to tell you that analog tape sounds very different from digital files. Analog tape, in general, is noticeably warmer than a digital recording. Additionally, the distortion achieved when an analog tape machine clips the signal is not as harsh as digital clipping. Analog tape distortion can add a vintage flair to vocals and guitars (Think Aretha Franklin or Janis Joplin vocals).
Recording to analog tape is horribly inconvenient. In the world of editing, digital is far and away superior. For those looking to recreate the tone of an analog tape on their digital recording, there is a solution. Digital signals can be run onto an analog tape machine and then run back to the computer. One advantage to doing this is that once the perfect take is caught, the unreliability of magnetic tape will not be a problem. The level of the signal sent to the tape can be fine-tuned to get the perfect amount of tape distortion, and the warmer signal may be more pleasing to the ears of musicians who prefer the vintage vibe for their sound.
Do your research before investing in a reel-to-reel tape machine to be sure that it is compatible with your recording setup, and that it is operating at a useable capacity. Analog Tape machines are not mass produced as they were before music recording switched to digital formats, so a vintage tape machine may be your only option.
Unorthodox Distortion
While we're on the topic of distortion, guitar amplifiers can be used to distort instruments that we are not used to hearing distorted. One famous but often unrecognized example of this technique is with Justin Timberlake's vocal track during the bridge of his 2006 hit "SexyBack". Timberlake's vocals were recorded through a guitar amplifier to deliver a fuzzy, grimy tone. Jack White has also been known to deliver dirty vocals through a guitar amplifier using a harmonica microphone. Guitar amplifiers can be used for nearly any instrument.
Like the analog tape machine, audio tracks can be sent out to a guitar amplifier and returned to the digital audio workstation via the microphone.
Don't have a guitar amplifier handy? No problem, try any of these software plugins that simulate amps
Playing with Voltage
It is common knowledge that Jimi Hendrix would run his guitar pedals on dying batteries because the reduced voltage would produce a warmer tone. In the same way, Eddie Van Halen would run his Marshall stacks at a lower voltage to achieve more distortion. By reducing voltage, we reduce the efficiency at which our gear is operating, producing more distortion. I would not recommend imitating Eddie Van Halen's techniques for the safety of you and your gear. Pedals however, are a different story. Danelectro produces a power supply called the Danelectrode which can vary voltage from 3-9 volts. This can yield a wide variety of tones for your recording or live performance. I would not suggest using this device on digital pedals because they will not operate properly, or may not operate at all. Overdrive, Fuzz, Distortion, and analog modulation and delay pedals with dropped voltage may give you the vintage tones which you have been searching for.
Playing with Time
Recording a song faster than performance tempo and slowing it down to drop the pitch used to be quite common in the 1970s. This technique can be heard on many Led Zeppelin recordings, including Hey, Hey, What Can I Do.” The opposite of this was done to Bruce Springsteen's 1980 hit, Hungry Heart.” On Hey Hey What Can I Do,” Plant's vocals were recorded separately from the band, so Plant's voice sounds comparable to his voice on other Led Zeppelin recordings. On Hungry Heart,” however, Springsteen's voice was sped up with the rest of the song, giving it a brighter timbre. Most digital audio workstations will allow you to play with time and pitch independent from one another. When one is dependent on the other, however, new, old, and interesting sounds can be achieved.
At the end of the day, your goal should be creating the sound that's best for you. Given the myriad of recording techniques at your fingertips, this article should only scratch the surface of all possibilities. Be sure to equip yourself with all the proper gear, and you'll be inventing a sound that's unique, and specific to your music style.
Monday, 11 July 2016
Vocal Recording & Production Masterclass
Published June 2006
By Paul White
Everyone knows how much a classy vocal sound can add to any recording, but achieving this in the studio can seem something of a black art, so this month we demystify the recording, processing, and mixing techniques required to produce professional results.
In theory, getting a good vocal sound should now be easier than ever, as studio quality microphones seem now to be available for little more than pocket money. However, we still get lots of letters and emails at the SOS office about problems that musicians are encountering with this task, and we are often called on to advise on this topic in our regular Studio SOS visits. So I felt it would be valuable to put all the pertinent information in one place. Clearly the vocalist will always be the most important factor — a great vocal recording takes a great singer — but if we assume for the moment that the singer is delivering a good performance, what can you do at the engineering end to capture this in its best possible light?
Most pop vocal work is done using some kind of large-diaphragm, cardioid-pattern capacitor microphone, but there are exceptions. For example, some singers like to work with dynamic mics such as Shure SM58s, because the tonality suits their voice and/or the style of music. It's also perfectly possible to make a good vocal recording using a small-diaphragm capacitor microphone and, in the right room, an omnidirectional microphone can give good results, so it's really a case of the end justifying the means.
Perhaps more relevant than the type or make of microphone is its tonal character, as matching the microphone to the singer is more important than many people appreciate — and if you've never had the chance to compare a range of different mics in a recording situation, that's quite understandable. Even now I still get phone calls from people asking which mic is the 'best buy' within a certain price range, but that's really impossible to answer in a meaningful way without knowing something about the voice being recorded. Most studio microphones, especially large-diaphragm models, have a distinct tonal character, so it's best to have a selection of microphones available, allowing you to select the preferred one for each recording task, even if that choice is only between two or three different models.
Although figure-of-eight mics aren't the most common choice for lead vocals, there are occasions where their deep 90-degree rejection nulls are ideal for the task in hand. For example, if you need to record a vocalist while he is playing an acoustic guitar, the nulls can be angled to give much better separation between guitar and vocal than could be achieved with two your budget can reach this far, then the obvious choice is to get one fairly neutral mic, one with a warm, syrupy quality, and another with a bit of sizzle on the top end. You may also wish to consider a tube microphone, though these can sound very different — so different, in fact, that it's hard to say what the tube sound is. To me, the better tube mics sound crisp and clear without being harsh or sibilant, and they also add weight and density to the lower vocal frequencies. Those mics that deliberately overdrive the tube to get a 'warm' sound can end up sounding mushy, dull or just plain distorted. Fortunately most mic manufacturers have now realised that the best-sounding tube mic is the one that doesn't use the tube to create artificial distortion — what it contributes naturally is quite enough.
Now I wouldn't dare suggest that women are in any way 'difficult' as a species, but I have found that choosing the best microphone to suit a female singer is generally more difficult than for male singers. Maybe it's because their voices tend, on balance, to be smoother sounding than men's voices, which probably equates to less harmonic complexity — or at least, a lower level of some of those complex harmonics. This leaves the throaty elements of the sound more exposed, so if you pick the wrong microphone you can end up with what sounds like distortion when what you're really hearing are the less desirable throaty artifacts of the sound being emphasised. Close miking is pretty unforgiving of unattractive vocal characteristics anyway, but if you're unfortunate in choosing a mic that exaggerates these less attractive artifacts, you're in trouble from the start.
As you might expect, microphones with a strong presence peak in the upper mid-range are the most likely to give rise to problems of this kind (presence peaks up above 8kHz are less likely to aggravate this type of problem), but then that same mic might rescue a session where you're dealing with a singer who has a voice that lacks definition. It's not usually a case of 'good mic versus bad mic', but rather 'suitable mic versus unsuitable mic'. Of the microphones I've reviewed over the past few years, very few of them have actually been poor from a technical standpoint, and even then, it's usually been from the perspective of noise rather than tonality.
The same tends also to be true for microphone preamplifiers. It's undeniably true that expensive boutique models can sound noticeably better than budget models, but I've used preamps and mixers costing under £100 in combination with budget Chinese capacitor mics and had great results, both artistically and technically. That's not to say that the very best gear wouldn't sound even better, but in many of the situations I've encountered, the equipment has not been the limiting factor. More often than not, it's what the microphone picks up that's the main problem with the recorded sound.
Most people know that microphones 'hear stuff', but they're not always aware of what stuff they hear. If you stick a singer a few inches away from the microphone, then obviously you're going to record the singer, but you're also going to record any other sound approaching the microphone. Vocals are almost always recorded using cardioid-pattern microphones, as they have a directional characteristic that excludes more off-axis sound (such as room reflections or spill from other sources) than omnidirectional microphones, which are equally sensitive all the way around. If you have a very dead room or vocal booth that will allow you to use an omni-pattern microphone, you may find that you get a more natural or 'open' sound than you do from a cardioid microphone, with the added bonus that there will be no tonal variation if the singer changes position slightly while singing. Omni mics also don't exhibit a bass boost when used close up (the so-called 'proximity effect'), so the amount of low end they produce doesn't vary as the singer gets closer to the microphone.
Here you can see response diagrams for three polar patterns used in vocal mics — the lobes of each figure show the sensitivity of the mic to sound arriving at different angles relative to the mic's capsule. The most common pattern used in modern vocal recording is the cardioid pattern (left), which picks up best from directly in front of the capsule while rejecting sound from directly behind. The omnidirectional pattern (centre) is more common in mics used for acoustic music, and picks up equally well from all directions. The figure-of-eight polar pattern (right) is fairly uncommon for recording vocals, picking up sound from in front and behind while rejecting sound from the sides. However, if you decide that the sound of a ribbon mic would suit your vocalist, then it will almost certainly use this other time an omni microphone might be preferable is when you're recording in a live environment where you want to capture the natural ambience of the room. An obvious example of this is choral music, but it can also be applied to solo parts where artistically suitable. Where you have a choice, a single-diaphragm, dedicated omni mic generally produces a more accurate off-axis response than you get from a multi-pattern mic that combines the outputs from two capsules to create the different polar patterns.
One further polar pattern to mention is figure of eight, because there's recently been a resurgence of interest in ribbon microphones, which all have a figure-of-eight pattern. Ribbon mics have a distinct tonality that is different to both normal condenser and dynamic mics, and they can sound good on a vocal that needs its high end smoothing out a little. These mics are equally sensitive to the front and to the rear of the capsule, so putting an absorbent screen behind the microphone while recording can help prevent unwanted room reflections from getting onto the recording. Figure-of-eight mics can be useful when recording a singer who plays acoustic guitar at the same time, as you can arrange for the dead zone of the mic (90 degrees off-axis) to be aimed towards the guitar to minimise spill. Note that all figure-of-eight mics have a strong proximity effect, so, again, if you work too close to them, the bass end will increase very noticeably.
Returning to the subject of cardioid microphones, one factor that hasn't attracted the attention it deserves is that single-diaphragm mics often behave differently to dual-diaphragm models. Dual diaphragms are used in switchable-pattern mics, and in cardioid mode they seem to pick up more of the room sound in close-miking situations than single-diaphragm models. The single-diaphragm mics always seem to produce a drier, more focused sound, which helps in situations where the amount of acoustic treatment that can be applied is limited. On the other hand, the dual-diaphragm model may deliver a more open sound, but often needs to be used in a more controlled environment to be effective.
Home-made vocal booths typically have too much acoustic absorption at high frequencies, resulting in a boxy sound. Fixing old CDs to the wall can help reinstate some reflections to remedy you graduate from recording in your bedroom and start thinking about converting your loft or garage into a studio, the idea of a separate vocal booth can seem very appealing.
Done properly, a vocal booth is a great asset, but simply creating a cupboard-sized enclosure with a door and then lining it with a couple of inches of acoustic foam invariably leads to disappointing results — what you end up with is a sound that is dull and lifeless at the high end, but uncontrolled and boxy at the low end. The reason this happens is that any absorbent material such as foam needs to be significantly thicker to be effective at low frequencies than it does at high frequencies. Where the foam is only a couple of inches thick, its effectiveness is much reduced below 400Hz or so, leaving those lower frequencies to behave almost as though the room hadn't been treated at all. What's worse is that by drying up all the high-end and upper mid-range reflections, the tonal balance seems worse than it was before, as the boomy low end now overwhelms the top end.
To create an effective vocal booth, the enclosure needs to be large enough to allow you to lose at least six inches from each of the three non-door walls, which will allow you to use a suitable depth of trapping. Four-inch foam or Rockwool slab on two-inch spacers to keep it off the wall is reasonably effective, but you should also consider leaving some small areas untrapped to add a little high-end reflectivity back into the booth, otherwise it may still end up sounding too dull. Another alternative is to fix small areas of reflective material to the surface of your traps — we used old CDs during one Studio SOS visit and it worked really well. Other approaches include putting the trapping behind perforated metal, plastic, or MDF sheeting.
As all microphone pickup patterns are three-dimensional, 'the sides' also pick up from above and below. Even if your room is quiet and there's no traffic noise, unless the room has been designed by a studio acoustician there will be reflections from walls and other surfaces that bounce 'second-hand' vocals back into the microphone. These reflections will be captured in addition to the direct sound of the voice, so you'll get the sound of the room overlaid on the sound you're actually after.
Here you can see the kind of setup that usually yields good results in the home studio. A cardioid microphone is mounted in a suspension shockmount and protected from breath blasts using a pop shield. The vocalist sings within 12 inches of the microphone and with her back to a suspended duvet to avoid sound reflecting off the wall behind her into the sensitive side of the mic.Photo: Mark EwingThis is often a significant problem in the home studio, as there are reflective walls, ceilings and furniture conspiring to corrupt your vocal sound. While you might add reverb afterwards, you'll find that it doesn't hide the boxy tone caused by room reflections. Even people who've bought seriously professional-grade equipment fall foul of this problem. There's a parallel here with photography: even a budget camera can give you an artistically fantastic picture if what is in front of the lens looks good and is correctly lit. But if the subject is badly lit, then no amount of upgrading will bring about a significant improvement in the end result.
Regular readers will know that there is a way to improve this situation without resorting acoustic treatment of the whole room. All you need to do is improve the acoustics in the vicinity of the microphone. I think most people appreciate this on an instinctive level and, because I keep banging on about it, some musicians have tried using duvets for acoustic treatment. However, a surprising number simply hang a duvet on the wall and then sing facing the duvet! This may help a little, as the duvet will absorb some of the sound that would otherwise bounce back off the wall, but it's far from the best solution. Think back to the cardioid mic's pickup pattern and you'll see that in the above scenario its least sensitive area (directly behind it) is now facing the duvet, while its sensitive front and sides are still picking up room reflections.
A better option is to hang the duvet behind the singer, as any reflections coming past the singer's head will otherwise be picked up by the extremely sensitive front of the microphone.
To reduce side pickup, you can hang a double-size duvet in a curve or use two to form a 'V' shape and have the singer stand with their back facing into the apex. That still leaves the ceiling and floor as possible reflection sources, but a rug or carpet on the floor will help enormously. If the ceiling is low, a foam panel above the mic and singer will do the trick nicely.
Improvised acoustic treatments made from duvets or blankets are effective at high frequencies and in the mid-range, but their efficiency falls off at low frequencies. In most cases, they work adequately over the vocal range, but doubling up the thickness or using a heavy, winter-grade duvet will improve their performance at the low end — useful when recording baritones. Such 'blanket' absorbers also work best when spaced away from a wall, so the best option, where practical, is to set up the vocal mic and absorbers out in the room, well away from the walls, but at the same time avoiding the very centre of the room, as that attracts its own set of special problems associated with room modes based on the room dimensions. As a final fix, you can shield the rear of the mic (after all, even a cardioid mic isn't totally deaf in the rear) by fixing a piece of acoustic foam close behind the microphone.
Auralex make some neat devices of this type in triangle and butterfly shapes that slot onto the mic stand, but if looks don't concern you, a chunk of furniture foam will serve almost as well.
Even great vocalists sing better on some takes than on others, so it is common practice to record the same vocal part several times and then compile (or 'comp') the best phrases together into the final take. This is easy to do in a typical DAW as you can view the various takes on adjacent tracks, audition them, then separate out the phrases you wish to keep and move them to a new track. If you only cut in between phrases, your edits shouldn't be audible, but if you do have to edit during a continuous phrase, try to edit either just before a hard consonant or during a sustained sound such as 'eeh or 'aah', using a short crossfade to hide the transition.
Breath noise is a natural part of singing, and if you remove it completely, the resulting track can sound unnatural. You can reduce excessive breath noise by manually selecting the offending sections of the audio file and doing a destructive gain change to reduce the level by a few decibels or, if you're a little more cautious, you can use mix automation to do the same thing. It's also often possible to use a gate processor, setting the amount of gate attenuation when closed to around 6dB, rather than having it mute everything completely. The same techniques can be used to reduce headphone spill in sections where there is no singing, but this problem is easily avoided by using closed headphones in the first place. If your singer likes to work with one phone off, make up a special lead so that only one of the phones is working.
If your vocalist prefers to sing wearing only one headphone, the spill from the loose earpiece could be very audible on your mic track. Fortunately, it's easy to make up a special switch to turn off one channel, as shown below.Various commercial nylon and metal mesh pop shields are available, but you can improvise your own from nylon stocking material stretched over a suitable hoop, if you prefer. Note that the foam wind shields that come with some mics are not very effective as pop shields, and they also tend to compromise the tone by filtering out some of the high end. I've covered this point many times before, but I make no excuse for repeating it, because I still come across people trying to use microphones with no pop shield. Even if you're using a stage dynamic vocal mic for recording vocals, a pop shield is still necessary to guarantee freedom from popping problems.
While a microphone shockmount is not mandatory, it really helps cut out low-frequency vibrations coming up from the floor, as they are prone to do if the singer has a habit of tapping their feet! This can be a real problem on wooden floors, so a shockmount is highly recommended. Some mics come with a shockmount, but if not, inexpensive generic mounts are available for most microphones — they not only solve vibration problems, but also look cool.
Now you're ready to record, and given that modern recording systems offer so much headroom, I'd suggest not using compression or EQ at the recording stage, as recording flat leaves you more flexibility for adjusting the sound afterwards. More experienced engineers sometimes process as they record, but as you may have already discovered, what sounds right in isolation may not always sound right once all the other parts have been recorded, so until you have gained the necessary experience, it's safer to leave your options open. Just make sure you leave a few extra decibels of safety margin on your recording system, as it's not only guitarists who get louder when the red light goes on!
You'll need to set up a suitably balanced headphone mix for the singer, and in most cases giving the singer a little reverb in the phones will help their pitching and also make them feel more confident about their performance. Make sure you don't record the reverb, though, as you'll want the freedom to adjust it at the mixing stage. Your 'bedside manner' can also make a big difference to the quality of the performance so try to be encouraging rather than critical, but at the same time make sure you capture enough takes to make up one good one.
If you intend to use a pitch-correction program or device, such as Auto-Tune, always apply this after recording — if the corrected sound is fed into the singer's headphone monitor mix, it will completely undermine their ability to control their pitching, and will result in a disastrous performance. Subdued lighting can also help get the singer in the right mood for the performance, but make sure you also look after the practicalities, such as making sure there is cool drinking water on hand and that the room is a comfortable temperature. Once you have captured that perfect vocal performance, you can polish it using EQ, compression, and reverb at the mixing stage, and I have a number of suggestions for tackling this in the most effective way.
The old pencil trick!Photo: Mark EwingNo recording textbook would be complete without a discussion on de-essers, but in my experience they are very rarely needed. A de-esser is a technological solution to the physical problem where some singers produce very pronounced 'S' and 'T' sounds because of the way air moves around their teeth. In most instances, using a different microphone or moving the mic slightly above or below the singer's mouth is enough to reduce the problem to manageable proportions. You can also try fixing a pencil, vertically, right in front of the mic grille, holding it in place with a couple of elastic bands around the mic body. This shields the centre of the diaphragm, and definitely reduces sibilance, though the exact reasons have probably never been properly researched. Yet another solution is to put an extra layer of stocking material over your pop filter to attenuate the high end slightly.
However, where the problem is serious, electronic de-essing is sometimes necessary. A de-esser is, essentially, a special type of compressor that only reacts to frequencies in the sibilance range (typically 3-6kHz). Simple de-essers will reduce the level of the entire audio signal when the level of the sibilant frequencies rises above a threshold you set. Unless the degree of processing is very light, this can impart a lisping quality to the vocals. Models that attenuate only the upper half of the audio spectrum are better, but the very best ones attenuate only frequencies in the sibilance band, leaving other frequencies unchanged.
Once you have compiled the best vocal take possible and cleaned up any unwanted noise in the pauses, what next? If pitching is still a problem, you can use Antares Auto-Tune or one of its competitors to automatically force the vocal pitch to the nearest note in a scale you set, or to the nearest semitone. That said, in my experience chromatic correction gives problems if the singer uses a lot of vibrato or if they scoop up to notes. I find it's best to take the time to get the scale right, then use the slowest pitch-correction rate that will do the job properly, as this gives the most natural sound. You can always automate the rate of correction if some parts of the song need more help than others. Where more drastic correction is needed, you may be better off using a program such as Celemony Melodyne that allows you to fine-tune individual notes in a graphic environment. Less serious pitching issues can sometimes be resolved by processing individual words using a pitch-shift algorithm to push them up or down by the necessary few cents. You can even automate the amount of shift applied to different words or phrases if you have the patience.
When it comes to EQ, there are no standard settings. In an ideal world you'd have teamed the singer with the mic that best suited their voice, but in less-than-ideal conditions, removing some low mid-range in the 150-400Hz range can help clean up boxiness, while a broad boost at 8kHz and upwards can add 'air' and sizzle to the sound, but use as little as you can get away with. Using a low-cut filter to get rid of subsonic audio and breath rumble can also be useful, though this is best done at source if your mic or preamp has a low-cut filter.
Reverb is also pretty much a question of taste, but for most classic vocal sounds I usually end up going back to a plate emulation, as these don't try to impose any particular room type on the sound. Combining this with a suitable short-ambience treatment (there are some lovely convolution reverbs available) also suits most voices. By balancing the two reverbs, you can get anything from a fairly upfront and focused vocal to a long ballad reverb, and I've had some success making the attack of the plate reverb slower or adding some pre-delay, as this keeps it out of the way of the ambience reverb. Vocal plate reverbs rarely need to be longer than 1.8 to 2.5 seconds, and in many cases rolling off some low end below 200Hz gives a cleaner and more lively sound. Putting a subtle repeating delay before the plate reverb can also produce a nice effect, provided that the music has enough space to allow the effect to breath.
Should you wish to double-track the vocal part for any reason, the best results are obtained by doing it for real, getting the singer to duplicate their part on another track. You may even have enough spare parts left over from compiling your vocal that you can build a second track. Where the singer can't match the first performance closely enough, you have the option of slicing up the phrases and aligning them visually using your DAW's waveform display. Where this fails, you may have to resort to cheating using processing, though in my view none of the fake methods sound as good as the real thing. In the early days of recording, 'fake' double-tracking or Automatic Double-tracking (ADT) was often attempted using a short tape delay using a spare tape recorder. Somebody then tried to vary the delay time by slowing the tape reels by hand and accidentally invented flanging!
Although there are plug-ins that do a passable job of faking double-tracking from a single vocal, introducing pitch and time variation, you can get almost as good results by copying the vocal to a new track, then processing it with Auto-Tune or something similar, as this will create subtle pitch differences between the original and the copy. Adding a delay of 60-100ms to the second part can also enhance the illusion that there are two different parts.
Keeping the vocal level nice and even is generally done with the aid of a compressor, aiming for around 6-8dB of gain reduction (using a ratio of between 2:1 and 6:1) on the loudest points in the track. We've written a lot on this subject, but the optimal attack and release time settings depend on the type of compressor or plug-in you are using. If you're unsure as to how to set these, take a peek at some of the presets and see what values they used, but don't expect a preset to work without further adjustment, as the correct threshold setting always depends on the level and dynamics of the signal being processed. However, compression alone isn't always enough to tame the dynamics of some singers, so use your mix automation to bring up any words or syllables that are getting lost or to drop the level of those that stand out too much.
If you find it hard to judge how loud the vocals should be, try listening to the track from outside the door of your studio, which will make your mix sound more like it would on a domestic radio or hi-fi. From this perspective, the vocal should be clearly audible above the backing, but not so loud that it doesn't sound like part of the same performance. If in doubt, play a commercial record and listen to that from the same position. Ultimately this is a subjective decision and it takes experience to get it dead right every time.
If you're interested in learning even more about the topics covered in this article, then head over to the Sound On Sound web site, where you can find a huge number of related articles from previous issues of the magazine:
Managing any overdubbing session can be difficult, and vocal sessions can be particularly challenging. However, there's help at hand in ' Basic Overdubbing ' (SOS March 2001), and you can also find advice on the important task of setting up the best headphone mix for the talent in ' Cue Monitoring Techniques ' (SOS January 2003).
Equalisation is probably the most common processing tool applied to vocal recordings, so if you want a detailed explanation of what equalisers are, how they work, and how to get the best out of them, try ' Equalisers Explained ' and ' Using Equalisation ' from SOS July and August 2001 respectively.
Compressors are not only extremely common in recording studios, but they are also vital to almost all modern vocal sounds. For an in-depth look at the different types of compression, as well as tips on making the most of your own outboard and plug-ins, have a look at our two-part ' Advanced Compression: Part 1 ' masterclass that appeared in SOS
December 2000 and Advanced Compression: Part 2 in January 2001.
We get asked questions about vocal reverb processing here in the Sound On Sound offices. Our ' Advanced Reverberation: Part 1 ' and ' Advanced Reverberation: Part 2 ' articles discuss all the common reverb types and algorithm parameters you might encounter, as well as going into some of the finer points of how to find the best settings for your particular track.
Mike Senior
SOS
By Paul White
Everyone knows how much a classy vocal sound can add to any recording, but achieving this in the studio can seem something of a black art, so this month we demystify the recording, processing, and mixing techniques required to produce professional results.
In theory, getting a good vocal sound should now be easier than ever, as studio quality microphones seem now to be available for little more than pocket money. However, we still get lots of letters and emails at the SOS office about problems that musicians are encountering with this task, and we are often called on to advise on this topic in our regular Studio SOS visits. So I felt it would be valuable to put all the pertinent information in one place. Clearly the vocalist will always be the most important factor — a great vocal recording takes a great singer — but if we assume for the moment that the singer is delivering a good performance, what can you do at the engineering end to capture this in its best possible light?
Most pop vocal work is done using some kind of large-diaphragm, cardioid-pattern capacitor microphone, but there are exceptions. For example, some singers like to work with dynamic mics such as Shure SM58s, because the tonality suits their voice and/or the style of music. It's also perfectly possible to make a good vocal recording using a small-diaphragm capacitor microphone and, in the right room, an omnidirectional microphone can give good results, so it's really a case of the end justifying the means.
Perhaps more relevant than the type or make of microphone is its tonal character, as matching the microphone to the singer is more important than many people appreciate — and if you've never had the chance to compare a range of different mics in a recording situation, that's quite understandable. Even now I still get phone calls from people asking which mic is the 'best buy' within a certain price range, but that's really impossible to answer in a meaningful way without knowing something about the voice being recorded. Most studio microphones, especially large-diaphragm models, have a distinct tonal character, so it's best to have a selection of microphones available, allowing you to select the preferred one for each recording task, even if that choice is only between two or three different models.
Although figure-of-eight mics aren't the most common choice for lead vocals, there are occasions where their deep 90-degree rejection nulls are ideal for the task in hand. For example, if you need to record a vocalist while he is playing an acoustic guitar, the nulls can be angled to give much better separation between guitar and vocal than could be achieved with two your budget can reach this far, then the obvious choice is to get one fairly neutral mic, one with a warm, syrupy quality, and another with a bit of sizzle on the top end. You may also wish to consider a tube microphone, though these can sound very different — so different, in fact, that it's hard to say what the tube sound is. To me, the better tube mics sound crisp and clear without being harsh or sibilant, and they also add weight and density to the lower vocal frequencies. Those mics that deliberately overdrive the tube to get a 'warm' sound can end up sounding mushy, dull or just plain distorted. Fortunately most mic manufacturers have now realised that the best-sounding tube mic is the one that doesn't use the tube to create artificial distortion — what it contributes naturally is quite enough.
Now I wouldn't dare suggest that women are in any way 'difficult' as a species, but I have found that choosing the best microphone to suit a female singer is generally more difficult than for male singers. Maybe it's because their voices tend, on balance, to be smoother sounding than men's voices, which probably equates to less harmonic complexity — or at least, a lower level of some of those complex harmonics. This leaves the throaty elements of the sound more exposed, so if you pick the wrong microphone you can end up with what sounds like distortion when what you're really hearing are the less desirable throaty artifacts of the sound being emphasised. Close miking is pretty unforgiving of unattractive vocal characteristics anyway, but if you're unfortunate in choosing a mic that exaggerates these less attractive artifacts, you're in trouble from the start.
As you might expect, microphones with a strong presence peak in the upper mid-range are the most likely to give rise to problems of this kind (presence peaks up above 8kHz are less likely to aggravate this type of problem), but then that same mic might rescue a session where you're dealing with a singer who has a voice that lacks definition. It's not usually a case of 'good mic versus bad mic', but rather 'suitable mic versus unsuitable mic'. Of the microphones I've reviewed over the past few years, very few of them have actually been poor from a technical standpoint, and even then, it's usually been from the perspective of noise rather than tonality.
The same tends also to be true for microphone preamplifiers. It's undeniably true that expensive boutique models can sound noticeably better than budget models, but I've used preamps and mixers costing under £100 in combination with budget Chinese capacitor mics and had great results, both artistically and technically. That's not to say that the very best gear wouldn't sound even better, but in many of the situations I've encountered, the equipment has not been the limiting factor. More often than not, it's what the microphone picks up that's the main problem with the recorded sound.
Most people know that microphones 'hear stuff', but they're not always aware of what stuff they hear. If you stick a singer a few inches away from the microphone, then obviously you're going to record the singer, but you're also going to record any other sound approaching the microphone. Vocals are almost always recorded using cardioid-pattern microphones, as they have a directional characteristic that excludes more off-axis sound (such as room reflections or spill from other sources) than omnidirectional microphones, which are equally sensitive all the way around. If you have a very dead room or vocal booth that will allow you to use an omni-pattern microphone, you may find that you get a more natural or 'open' sound than you do from a cardioid microphone, with the added bonus that there will be no tonal variation if the singer changes position slightly while singing. Omni mics also don't exhibit a bass boost when used close up (the so-called 'proximity effect'), so the amount of low end they produce doesn't vary as the singer gets closer to the microphone.
Here you can see response diagrams for three polar patterns used in vocal mics — the lobes of each figure show the sensitivity of the mic to sound arriving at different angles relative to the mic's capsule. The most common pattern used in modern vocal recording is the cardioid pattern (left), which picks up best from directly in front of the capsule while rejecting sound from directly behind. The omnidirectional pattern (centre) is more common in mics used for acoustic music, and picks up equally well from all directions. The figure-of-eight polar pattern (right) is fairly uncommon for recording vocals, picking up sound from in front and behind while rejecting sound from the sides. However, if you decide that the sound of a ribbon mic would suit your vocalist, then it will almost certainly use this other time an omni microphone might be preferable is when you're recording in a live environment where you want to capture the natural ambience of the room. An obvious example of this is choral music, but it can also be applied to solo parts where artistically suitable. Where you have a choice, a single-diaphragm, dedicated omni mic generally produces a more accurate off-axis response than you get from a multi-pattern mic that combines the outputs from two capsules to create the different polar patterns.
One further polar pattern to mention is figure of eight, because there's recently been a resurgence of interest in ribbon microphones, which all have a figure-of-eight pattern. Ribbon mics have a distinct tonality that is different to both normal condenser and dynamic mics, and they can sound good on a vocal that needs its high end smoothing out a little. These mics are equally sensitive to the front and to the rear of the capsule, so putting an absorbent screen behind the microphone while recording can help prevent unwanted room reflections from getting onto the recording. Figure-of-eight mics can be useful when recording a singer who plays acoustic guitar at the same time, as you can arrange for the dead zone of the mic (90 degrees off-axis) to be aimed towards the guitar to minimise spill. Note that all figure-of-eight mics have a strong proximity effect, so, again, if you work too close to them, the bass end will increase very noticeably.
Returning to the subject of cardioid microphones, one factor that hasn't attracted the attention it deserves is that single-diaphragm mics often behave differently to dual-diaphragm models. Dual diaphragms are used in switchable-pattern mics, and in cardioid mode they seem to pick up more of the room sound in close-miking situations than single-diaphragm models. The single-diaphragm mics always seem to produce a drier, more focused sound, which helps in situations where the amount of acoustic treatment that can be applied is limited. On the other hand, the dual-diaphragm model may deliver a more open sound, but often needs to be used in a more controlled environment to be effective.
Home-made vocal booths typically have too much acoustic absorption at high frequencies, resulting in a boxy sound. Fixing old CDs to the wall can help reinstate some reflections to remedy you graduate from recording in your bedroom and start thinking about converting your loft or garage into a studio, the idea of a separate vocal booth can seem very appealing.
Done properly, a vocal booth is a great asset, but simply creating a cupboard-sized enclosure with a door and then lining it with a couple of inches of acoustic foam invariably leads to disappointing results — what you end up with is a sound that is dull and lifeless at the high end, but uncontrolled and boxy at the low end. The reason this happens is that any absorbent material such as foam needs to be significantly thicker to be effective at low frequencies than it does at high frequencies. Where the foam is only a couple of inches thick, its effectiveness is much reduced below 400Hz or so, leaving those lower frequencies to behave almost as though the room hadn't been treated at all. What's worse is that by drying up all the high-end and upper mid-range reflections, the tonal balance seems worse than it was before, as the boomy low end now overwhelms the top end.
To create an effective vocal booth, the enclosure needs to be large enough to allow you to lose at least six inches from each of the three non-door walls, which will allow you to use a suitable depth of trapping. Four-inch foam or Rockwool slab on two-inch spacers to keep it off the wall is reasonably effective, but you should also consider leaving some small areas untrapped to add a little high-end reflectivity back into the booth, otherwise it may still end up sounding too dull. Another alternative is to fix small areas of reflective material to the surface of your traps — we used old CDs during one Studio SOS visit and it worked really well. Other approaches include putting the trapping behind perforated metal, plastic, or MDF sheeting.
As all microphone pickup patterns are three-dimensional, 'the sides' also pick up from above and below. Even if your room is quiet and there's no traffic noise, unless the room has been designed by a studio acoustician there will be reflections from walls and other surfaces that bounce 'second-hand' vocals back into the microphone. These reflections will be captured in addition to the direct sound of the voice, so you'll get the sound of the room overlaid on the sound you're actually after.
Here you can see the kind of setup that usually yields good results in the home studio. A cardioid microphone is mounted in a suspension shockmount and protected from breath blasts using a pop shield. The vocalist sings within 12 inches of the microphone and with her back to a suspended duvet to avoid sound reflecting off the wall behind her into the sensitive side of the mic.Photo: Mark EwingThis is often a significant problem in the home studio, as there are reflective walls, ceilings and furniture conspiring to corrupt your vocal sound. While you might add reverb afterwards, you'll find that it doesn't hide the boxy tone caused by room reflections. Even people who've bought seriously professional-grade equipment fall foul of this problem. There's a parallel here with photography: even a budget camera can give you an artistically fantastic picture if what is in front of the lens looks good and is correctly lit. But if the subject is badly lit, then no amount of upgrading will bring about a significant improvement in the end result.
Regular readers will know that there is a way to improve this situation without resorting acoustic treatment of the whole room. All you need to do is improve the acoustics in the vicinity of the microphone. I think most people appreciate this on an instinctive level and, because I keep banging on about it, some musicians have tried using duvets for acoustic treatment. However, a surprising number simply hang a duvet on the wall and then sing facing the duvet! This may help a little, as the duvet will absorb some of the sound that would otherwise bounce back off the wall, but it's far from the best solution. Think back to the cardioid mic's pickup pattern and you'll see that in the above scenario its least sensitive area (directly behind it) is now facing the duvet, while its sensitive front and sides are still picking up room reflections.
A better option is to hang the duvet behind the singer, as any reflections coming past the singer's head will otherwise be picked up by the extremely sensitive front of the microphone.
To reduce side pickup, you can hang a double-size duvet in a curve or use two to form a 'V' shape and have the singer stand with their back facing into the apex. That still leaves the ceiling and floor as possible reflection sources, but a rug or carpet on the floor will help enormously. If the ceiling is low, a foam panel above the mic and singer will do the trick nicely.
Improvised acoustic treatments made from duvets or blankets are effective at high frequencies and in the mid-range, but their efficiency falls off at low frequencies. In most cases, they work adequately over the vocal range, but doubling up the thickness or using a heavy, winter-grade duvet will improve their performance at the low end — useful when recording baritones. Such 'blanket' absorbers also work best when spaced away from a wall, so the best option, where practical, is to set up the vocal mic and absorbers out in the room, well away from the walls, but at the same time avoiding the very centre of the room, as that attracts its own set of special problems associated with room modes based on the room dimensions. As a final fix, you can shield the rear of the mic (after all, even a cardioid mic isn't totally deaf in the rear) by fixing a piece of acoustic foam close behind the microphone.
Auralex make some neat devices of this type in triangle and butterfly shapes that slot onto the mic stand, but if looks don't concern you, a chunk of furniture foam will serve almost as well.
Even great vocalists sing better on some takes than on others, so it is common practice to record the same vocal part several times and then compile (or 'comp') the best phrases together into the final take. This is easy to do in a typical DAW as you can view the various takes on adjacent tracks, audition them, then separate out the phrases you wish to keep and move them to a new track. If you only cut in between phrases, your edits shouldn't be audible, but if you do have to edit during a continuous phrase, try to edit either just before a hard consonant or during a sustained sound such as 'eeh or 'aah', using a short crossfade to hide the transition.
Breath noise is a natural part of singing, and if you remove it completely, the resulting track can sound unnatural. You can reduce excessive breath noise by manually selecting the offending sections of the audio file and doing a destructive gain change to reduce the level by a few decibels or, if you're a little more cautious, you can use mix automation to do the same thing. It's also often possible to use a gate processor, setting the amount of gate attenuation when closed to around 6dB, rather than having it mute everything completely. The same techniques can be used to reduce headphone spill in sections where there is no singing, but this problem is easily avoided by using closed headphones in the first place. If your singer likes to work with one phone off, make up a special lead so that only one of the phones is working.
If your vocalist prefers to sing wearing only one headphone, the spill from the loose earpiece could be very audible on your mic track. Fortunately, it's easy to make up a special switch to turn off one channel, as shown below.Various commercial nylon and metal mesh pop shields are available, but you can improvise your own from nylon stocking material stretched over a suitable hoop, if you prefer. Note that the foam wind shields that come with some mics are not very effective as pop shields, and they also tend to compromise the tone by filtering out some of the high end. I've covered this point many times before, but I make no excuse for repeating it, because I still come across people trying to use microphones with no pop shield. Even if you're using a stage dynamic vocal mic for recording vocals, a pop shield is still necessary to guarantee freedom from popping problems.
While a microphone shockmount is not mandatory, it really helps cut out low-frequency vibrations coming up from the floor, as they are prone to do if the singer has a habit of tapping their feet! This can be a real problem on wooden floors, so a shockmount is highly recommended. Some mics come with a shockmount, but if not, inexpensive generic mounts are available for most microphones — they not only solve vibration problems, but also look cool.
Now you're ready to record, and given that modern recording systems offer so much headroom, I'd suggest not using compression or EQ at the recording stage, as recording flat leaves you more flexibility for adjusting the sound afterwards. More experienced engineers sometimes process as they record, but as you may have already discovered, what sounds right in isolation may not always sound right once all the other parts have been recorded, so until you have gained the necessary experience, it's safer to leave your options open. Just make sure you leave a few extra decibels of safety margin on your recording system, as it's not only guitarists who get louder when the red light goes on!
You'll need to set up a suitably balanced headphone mix for the singer, and in most cases giving the singer a little reverb in the phones will help their pitching and also make them feel more confident about their performance. Make sure you don't record the reverb, though, as you'll want the freedom to adjust it at the mixing stage. Your 'bedside manner' can also make a big difference to the quality of the performance so try to be encouraging rather than critical, but at the same time make sure you capture enough takes to make up one good one.
If you intend to use a pitch-correction program or device, such as Auto-Tune, always apply this after recording — if the corrected sound is fed into the singer's headphone monitor mix, it will completely undermine their ability to control their pitching, and will result in a disastrous performance. Subdued lighting can also help get the singer in the right mood for the performance, but make sure you also look after the practicalities, such as making sure there is cool drinking water on hand and that the room is a comfortable temperature. Once you have captured that perfect vocal performance, you can polish it using EQ, compression, and reverb at the mixing stage, and I have a number of suggestions for tackling this in the most effective way.
The old pencil trick!Photo: Mark EwingNo recording textbook would be complete without a discussion on de-essers, but in my experience they are very rarely needed. A de-esser is a technological solution to the physical problem where some singers produce very pronounced 'S' and 'T' sounds because of the way air moves around their teeth. In most instances, using a different microphone or moving the mic slightly above or below the singer's mouth is enough to reduce the problem to manageable proportions. You can also try fixing a pencil, vertically, right in front of the mic grille, holding it in place with a couple of elastic bands around the mic body. This shields the centre of the diaphragm, and definitely reduces sibilance, though the exact reasons have probably never been properly researched. Yet another solution is to put an extra layer of stocking material over your pop filter to attenuate the high end slightly.
However, where the problem is serious, electronic de-essing is sometimes necessary. A de-esser is, essentially, a special type of compressor that only reacts to frequencies in the sibilance range (typically 3-6kHz). Simple de-essers will reduce the level of the entire audio signal when the level of the sibilant frequencies rises above a threshold you set. Unless the degree of processing is very light, this can impart a lisping quality to the vocals. Models that attenuate only the upper half of the audio spectrum are better, but the very best ones attenuate only frequencies in the sibilance band, leaving other frequencies unchanged.
Once you have compiled the best vocal take possible and cleaned up any unwanted noise in the pauses, what next? If pitching is still a problem, you can use Antares Auto-Tune or one of its competitors to automatically force the vocal pitch to the nearest note in a scale you set, or to the nearest semitone. That said, in my experience chromatic correction gives problems if the singer uses a lot of vibrato or if they scoop up to notes. I find it's best to take the time to get the scale right, then use the slowest pitch-correction rate that will do the job properly, as this gives the most natural sound. You can always automate the rate of correction if some parts of the song need more help than others. Where more drastic correction is needed, you may be better off using a program such as Celemony Melodyne that allows you to fine-tune individual notes in a graphic environment. Less serious pitching issues can sometimes be resolved by processing individual words using a pitch-shift algorithm to push them up or down by the necessary few cents. You can even automate the amount of shift applied to different words or phrases if you have the patience.
When it comes to EQ, there are no standard settings. In an ideal world you'd have teamed the singer with the mic that best suited their voice, but in less-than-ideal conditions, removing some low mid-range in the 150-400Hz range can help clean up boxiness, while a broad boost at 8kHz and upwards can add 'air' and sizzle to the sound, but use as little as you can get away with. Using a low-cut filter to get rid of subsonic audio and breath rumble can also be useful, though this is best done at source if your mic or preamp has a low-cut filter.
Reverb is also pretty much a question of taste, but for most classic vocal sounds I usually end up going back to a plate emulation, as these don't try to impose any particular room type on the sound. Combining this with a suitable short-ambience treatment (there are some lovely convolution reverbs available) also suits most voices. By balancing the two reverbs, you can get anything from a fairly upfront and focused vocal to a long ballad reverb, and I've had some success making the attack of the plate reverb slower or adding some pre-delay, as this keeps it out of the way of the ambience reverb. Vocal plate reverbs rarely need to be longer than 1.8 to 2.5 seconds, and in many cases rolling off some low end below 200Hz gives a cleaner and more lively sound. Putting a subtle repeating delay before the plate reverb can also produce a nice effect, provided that the music has enough space to allow the effect to breath.
Should you wish to double-track the vocal part for any reason, the best results are obtained by doing it for real, getting the singer to duplicate their part on another track. You may even have enough spare parts left over from compiling your vocal that you can build a second track. Where the singer can't match the first performance closely enough, you have the option of slicing up the phrases and aligning them visually using your DAW's waveform display. Where this fails, you may have to resort to cheating using processing, though in my view none of the fake methods sound as good as the real thing. In the early days of recording, 'fake' double-tracking or Automatic Double-tracking (ADT) was often attempted using a short tape delay using a spare tape recorder. Somebody then tried to vary the delay time by slowing the tape reels by hand and accidentally invented flanging!
Although there are plug-ins that do a passable job of faking double-tracking from a single vocal, introducing pitch and time variation, you can get almost as good results by copying the vocal to a new track, then processing it with Auto-Tune or something similar, as this will create subtle pitch differences between the original and the copy. Adding a delay of 60-100ms to the second part can also enhance the illusion that there are two different parts.
Keeping the vocal level nice and even is generally done with the aid of a compressor, aiming for around 6-8dB of gain reduction (using a ratio of between 2:1 and 6:1) on the loudest points in the track. We've written a lot on this subject, but the optimal attack and release time settings depend on the type of compressor or plug-in you are using. If you're unsure as to how to set these, take a peek at some of the presets and see what values they used, but don't expect a preset to work without further adjustment, as the correct threshold setting always depends on the level and dynamics of the signal being processed. However, compression alone isn't always enough to tame the dynamics of some singers, so use your mix automation to bring up any words or syllables that are getting lost or to drop the level of those that stand out too much.
If you find it hard to judge how loud the vocals should be, try listening to the track from outside the door of your studio, which will make your mix sound more like it would on a domestic radio or hi-fi. From this perspective, the vocal should be clearly audible above the backing, but not so loud that it doesn't sound like part of the same performance. If in doubt, play a commercial record and listen to that from the same position. Ultimately this is a subjective decision and it takes experience to get it dead right every time.
If you're interested in learning even more about the topics covered in this article, then head over to the Sound On Sound web site, where you can find a huge number of related articles from previous issues of the magazine:
Managing any overdubbing session can be difficult, and vocal sessions can be particularly challenging. However, there's help at hand in ' Basic Overdubbing ' (SOS March 2001), and you can also find advice on the important task of setting up the best headphone mix for the talent in ' Cue Monitoring Techniques ' (SOS January 2003).
Equalisation is probably the most common processing tool applied to vocal recordings, so if you want a detailed explanation of what equalisers are, how they work, and how to get the best out of them, try ' Equalisers Explained ' and ' Using Equalisation ' from SOS July and August 2001 respectively.
Compressors are not only extremely common in recording studios, but they are also vital to almost all modern vocal sounds. For an in-depth look at the different types of compression, as well as tips on making the most of your own outboard and plug-ins, have a look at our two-part ' Advanced Compression: Part 1 ' masterclass that appeared in SOS
December 2000 and Advanced Compression: Part 2 in January 2001.
We get asked questions about vocal reverb processing here in the Sound On Sound offices. Our ' Advanced Reverberation: Part 1 ' and ' Advanced Reverberation: Part 2 ' articles discuss all the common reverb types and algorithm parameters you might encounter, as well as going into some of the finer points of how to find the best settings for your particular track.
Mike Senior
SOS
Subscribe to:
Posts (Atom)