TRANSKRIPT 🔗
Mark Mynett
Welcome to the Sound On Sound Recording & Mixing podcast channel. I'm Mark Mynett and in this episode we're looking at recording the world's oldest instrument, which is the human voice.
Let's just start with a few 101s about recording vocals, which sort of like more relate to your skills as a producer, which are incredibly important when recording vocals.
So, the first one is that if you have sufficient freedom from budget and time constraints, it's preferable to have the final bass and guitar and other instrument tracks completed in order to provide a strong context for vocal tracking. However, when studio time is restricted, leaving the vocals to the end of the recording schedule tends to be counterproductive. Most vocal styles involve considerable performance exertion requiring equal measures of vocal strength and stamina and as a result it is rare that vocalists can spend more than three or four hours each day tracking, especially when maintaining the required level of emotional involvement. So in this case it's good practice to try and spread vocal tracking out over as long a time period as possible, preferably interspersing this within maybe tracking the guitars and bass or keyboards or any other instruments and, you know, sometimes you can get creative and schedule this around, sort of restringing the guitars or bass whilst vocals are being recorded, saving time and probably money in the process.
As well as taking care of the schedule when recording vocals, a producer's interpersonal skills are particularly essential when tracking vocals. Don't forget that the vocals are the only embodied instrument, so physical and emotional considerations are really important here. Firstly, this means ensuring that the vocalist is as comfortable as possible, both physically and psychologically, so it's worth checking in with the relevant vocalist to see if they prefer singing in the morning or afternoon. Make sure that the heating and the lighting are given consideration and also check with the vocalist which other band members, if any, they would prefer to be present in the control room when tracking.
And other little Aspects you can consider that if the vocalist is also the bassist or the guitarist in the band, establish whether the familiar feel of playing whilst singing might sort of elicit more vocal energy and emotional conviction and you know, some producers even go so far as to have the live performed element perhaps guitar or bass introduced into the headphone mix. Again, anything you can to get the vocalist into the right frame of mind for committing to the vocal performances for some vocal sessions. Effective coaching is essential for bringing out the right performance, energy, power and conviction and for others, it's a greater focus on metering, clarity and intelligibility.
But whichever your focus is there, or maybe a combination of both, make sure that the lyrics are as absolutely comprehensible as possible and one way, a
Load full transcript
really worthwhile way of ensuring this is to make sure that the vocalist memorises their lyrics so that they can fully concentrate on the emotional delivery and phrasing rather than referencing a lyric sheet. On the subject of lyric sheets though make sure you've got lyric sheets printed out for your own reference so you can make notes as you go along, detailed line by line performance notes and it's also going to speed up your vocal comping process along the way.
And on the subject of vocal comping just before we continue on to the engineering stage, if you do have a session where there is an extensive amount of comping up the vocals take after take you'll often lose the real passion and energy involved in a real vocal performance it can sound very sterilized very quickly. If time is available, a great tactic here is to get the vocalist to listen to the comped up vocal and then try and to get them back into the studio to beat the vocal comp.
And if there's pitching issues with the vocalist, a great way of going a little bit further with this is to use Melodyne, Autotune, whatever you're using to pitch correct the vocal and again have the vocalist listen to the vocal over a couple of weeks to the pitch corrected, comped up vocal take and they'll get used to those intervals and then get them back into the studio to try and beat the comp and it's an often effective tactic.
So onto the engineering side of things, let's firstly consider if you're recording vocals in your home environment. Vocals tend to deliver a preferable impact when recorded in a tight, acoustically neutral environment with a high ceiling. If you're recording in a space with a low, untreated ceiling, it can be result in resonant ringing frequencies emanating from the ceiling, which is often the closest surface to the vocalist.
So you can fix sound absorptive materials to combat this. So thick duvets, blankets, et cetera, to the ceiling or and the walls behind and around the vocalist. And although lower frequencies, so like below approximately 300-400Hz tend to remain unaffected unless purpose built absorption is used, it will still control a lot of those higher frequency reflections and provide a much drier sound. And if you are recording vocals in a commercial recording studio, just don't fall into the trap of having the vocalist sing near the glass pane dividing the control room and live space, you get some really nasty reflections coming off that pane and it can also put vocalists off their pitching, so just be careful about that but as with home environment just make sure that the treatment is above if possible and around the vocalist's head to ensure that the polar pattern of the vocal mic, which is most likely going to be cardioid, is correct, is picking up as minimum resonance and ambience from the recording environment as possible.
Mic selection is an interesting one because legend has it that Bono from U2 records a lot of his vocal performances with a Shure SM58 and you're probably thinking well why would he do that when he can afford possibly any microphone on the planet. Well arguably, if that is the case, it's because the SM58 matches the required frequency response from Bono's vocals, or whichever vocalist you like and from that perspective, there's no one vocal microphone that will be best for all vocalists. If you're tempted to think that a Neumann U87, a classic vocal recording mic, is the best vocal microphone, then it's not, because, if you've got a very bright vocalist, then a microphone is likely to to suit the recording of that vocal and as a general principle then, it's this idea about using a bright sounding mic for a darker, mellow vocal sound and a flatter, less hyped, more mellow sounding mic for vocalists that is naturally quite sibilant and bright sounding. Of course, compared to most dynamic mics, large diaphragm condenser models tend to pick up more vocal detail with extended, more natural, open sounding highs and of course, this upper frequency detail can help a vocal sound remain intelligible within the production. But all the same, the Shure SM7B, it's a dynamic mic that's regularly considered as the industry standard for recording, you know, rock and indie vocals and this is partly due to its favourable response to proximity effect. You can get very close to an SM7B and the low frequency roll off but bear in mind that due to their weaker comparative signal strength compared to condenser microphones, dynamic microphones are a little bit more reliant on mic pre quality than condensers. So if you are using a dynamic such as an SM7B or a Shure SM58, if you're able to use a much higher standard mic pre, then I would definitely recommend doing so.
Of course, if you do have a number of potential vocal microphones at your dispose, it's a good idea to do a test recording, setting them up equidistant from the vocalist and having the vocalist sing at the center of both or three or four, whichever of the microphones you're doing the test recording with, routing them through the same make and model of mic pre and then after compensating for level differences, assessing which of those vocal sounds works best within the context of the mix. Even if it's just a rough monitor mix, that should give you an idea of which of those vocal colourations and qualities will be most suitable for the production in hand.
So let's provide some context here by listening to a microphone test recording featuring an SM7B, a Neumann U87 and an AKG 414. With the diaphragms of the microphone placed very close together and the same distance from the performer and this is recorded without any printed compression or EQ, all through the same mic pre with very similar gain structure to allow equal signal level, but also the files have been loudness normalised with an LUFS meter to make sure that they're the same perceptual loudness.
Let's listen to the SM7B first of all. And now the Neuman U87. Although it might not be
immediately apparent, the clear differences between those vocal recordings, what you'll hear is that the SM7B has more of a sort of focus around the mid range, sort of like the 1-3, 3.5k range and the U87, the Neumann U87 has more detail, it has more lows and more highs, which doesn't necessarily mean that it would be the preferable mic choice because if a vocalist had a particular timbre that didn't have a particularly strong mid range then the SM7B might be the preferred vocal and might just respond to that vocal in a preferable manner.
Let's now have a listen to a comparison between two condenser microphones. Let's listen to the same recording of the U87 and let's listen to how that compares to an AKG414. So here's the U87 first. And here's the AKG414. The differences here won't be anywhere near as marked as between a condenser and a dynamic. In this instance we were comparing the Neumann U87 and the SM7B, the Shure SM7B. But there are clear differences nonetheless. There's a different sense of brightness, the low frequencies are different, and there's a little, there is difference in the detail in the mid range of the sound. And of course these qualities tend to get accentuated through compression and through EQ use at the mix stage.
So at this stage you might not hear any significant differences between the two signals, but as I say, with compression EQ and certainly effects use, those differences tend to get accentuated. But don't forget that different vocal mics can be suited to different parts of the arrangement, so you can actually change mics for different textures. In other words, if there's a slower section within the song and maybe it moves from electric guitars to acoustic guitars, you might want to highlight that dynamic change with a different vocal microphone and likewise, if you're tracking up backing vocals or harmonies, it can be very beneficial to change the microphone.
In fact, if you're doing vocal harmonies with the same vocalist, you can get a preferable texture by changing the microphones between the lead vocal and the backing vocal and that just helps thicken and track up the sound in different frequency regions. And although the same is true when you're tracking vocal doubles to thicken up the main vocal, the main vocal performance, just be wary that doubling and tracking up the lead vocal can sometimes compromise the emotional delivery, the nuances of the main vocal, which sort of get tends to get homogenized by the inevitable, even slightly different delivery of the second vocal track. So with a really compelling performance and strong vocal tone, there's less to be gained and vocal doubles should really be restricted or avoided altogether. And if you're doing gang vocals a favourable technique here is to set the microphone to figure of eight and either get the two vocalists or perhaps four vocalists the opposite side of the microphone and the eye contact and visual communication cues can really help the tightness of those vocals.
Let's move on to considering microphone placement. A lot of people consider the mouth region is just the sole point source when recording vocals. However, the upper chest, throat and nasal areas are also influential in projecting different tonal characteristics and from this perspective, you can actually consider the human vocal organs almost like a wind instrument and considering the valves and the airways that it travels through. So, from this perspective, chest throat resonance can sometimes contribute to a deeper, richer tonal colour and a more three dimensional sound. And from this perspective, the way to do it is to actually go slightly off axis and have the microphone slightly below the vocalist's mouth region facing up at the mouth region but sort of with the microphone perhaps closer to being horizontal with the the throat region and you'll get just a little bit more resonance and sort of like chest qualities from that vocal mic placement.
Similarly, if you have got quite a nasal vocalist, that's another microphone placement with the microphone slightly lower than the mouth region. pointing at the mouth of regions obviously off axis and it helps sort of slightly attenuate those nasal qualities but of course the opposite might be the case and that the resonance emanating from the nasal region can actually add an edgier colour that contributes to vocal brightness and clarity so perhaps if you've got a slightly darker sounding vocalist then an off axis placement with the microphone slightly above, sort of like towards the nasal region, off axis pointing the voice can be a preferable microphone technique.
If you're using a cardioid microphone, which for most vocal recordings is the preferable polar pattern, then the closer the vocalist is to the microphone, you get less impact from the ambience of the recording space and you get more of a detailed sound and of course you do get the low frequency lift from proximity effect but all the same close up placements can also result in the recording being compromised by plosive consonant breath blasts these are usually Ps and Bs and you can also get distracting clicks from sort of fluid movement in the mouth, they're like referred to as lip separation ticks or lip smacks and you also get whistling airflow qualities and you can also get obvious signal level fluctuations from the vocalist moving between lines or takes if they're, you know, very very close to the vocalist.
So with these qualities and potential aspects in mind it's particularly with the um sibilance and the breath blasts, the plosive consonant breath blasts, it's worth asking the vocalist to sing a line such as "Sat in solemn silence in a blackened prison block" and sing a line like that, that's got the the sibilant vowels the sat in solemn silence and the the breath blasts the blackened prison block. You should be able to establish whether those You aspects of the vocal performance are going to be
problematic or not.
However, with the closer distance from performer to microphone, you do get this enhancement of detail and you get this improved bass response, but great as it distances from performer to microphone, you get an enhanced capture of chest and throat resonance. So if you do get the opportunity, it is time well spent focusing on establishing the optimal mic to vocalist distance, which is typically between roughly five to nine inches. And if time allows, assess vocal to mic distances of five, seven and then nine inches and, you know, establish the best clarity and perceived size and lower frequencies of the vocalist and you will find a difference between, as I say, between five and seven and nine inches from mic to source distance.
So let's have a listen to some recordings of those different microphone to performer distances, namely 5, 7 and 9 inches between the microphone and vocalist. In this instance it's the U87, again without any processing, printed processing in the form of EQ or compression and the resulting files have been loudness normalised so that they have the same inherent loudness. Let's firstly listen to a vocalist to mic distance of five inches. Sounds like this. And now with the vocalist seven inches from the mic diaphragm. And lastly, nine inches between performer and mic diaphragm. What I'm hoping you will have heard there is that as we move from five inches between mic and performer to seven, then nine, that there's a greater capture of ambience and this is despite the fact that it was recorded in a well treated acoustic environment. And for me, for this style of vocal performance and this interaction between the vocalist and microphone, I much preferred the enhanced detail of the closer mic placement, which was five inches from performer to microphone and with a pop shield roughly halfway in between.
But that doesn't mean to say it would be the same with other vocalists. This is, you know, a country and western style vocal and if it was more of a powerful vocal performance you might find that the enhanced capture of details from the chest cavity or the throat region of the greater mic distances means that that is a preferable placement, especially with greater mic to vocalist distances. It's sometimes the case that plosive sounds aren't problematic, meaning that a pop shield is unnecessary. In most cases, though and especially with closer mic placements, a pop shield is essential for diffusing plosives before they hit the mic's diaphragm. You should set this up at least three inches from the mic. Never put a pop shield right next to the microphone next to the diaphragm as it doesn't actually give it opportunity to diffuse the those breath blasts so it needs to be at least three inches from the mic in order for the diffusion to be effective and to ensure that the vocalist stays at the same distance from the pop shield from subsequent takes. It's often worth asking them to roughly measure this distance in terms
of finger widths. You know, I'll usually say to a vocalist, ok, stay a three fingered distance between your mouth and the pop shield and that just gives them a reference for when they come back into the control room and go back out to allow them to stay the same distance from the pop shield.
So, having discussed different mic selection, different mic placement and distances from the vocal and pop shields etc, an important caveat is that some vocalists just will not give their best performance whilst having to remain static and the same distance from a pop shield and a microphone that they just cannot provide the necessary emotional delivery and if this is the case, it's actually better to capture an improved vocal performance with a slightly inferior sound than a substandard vocal performance with a comparatively superior sound. So in this instance with a vocalist that is not able to commit and give the best vocal delivery staying very static in the recording space, just give them a handheld dynamic vocal microphone. Just try and coach them to try and adopt good microphone technique. Especially try and get them to avoid adopting microphone techniques that they might use in the live environment, such as cupping the microphone to kind of break it up, get them to avoid that and demonstrate that you can get the same effect with processing in the mix. And again, try and get them to adopt a microphone technique that at least does give you a good vocal recording. But again, it's important to say that a much more impassioned, emotional, energetic vocal delivery with a slightly inferior recording technique is better than the other way around.
So having looked at the microphone considerations and pop shields etc, let's just take in terms of printed processing and especially printed compression. So printed compression, as you know, is recording with the processing directly applied to the captured signal and this tends to be more appropriate to vocal tracking than arguably any other instrument, apart from, you know, perhaps room mics with the drums. The added benefit of printed compression is with the headphone mix of the performer, you'll often find that vocalists get a more sort of proximate in your face vocal sound that they find more inspiring with compression.
Be careful though, unless you really know what you're doing as a general guide, it's best to start with only providing no more than maybe 6-8db of gain reduction at most and only that gain reduction impacting the louder note. So perhaps go with a low ratio, perhaps 4 to 1 and with a fairly fast attack time, you know, under about 3 milliseconds, but avoid going over maybe dB of gain reduction. It's unlikely you're going to get that sufficiently wrong that will cause the vocal to be unusable and you can set the release with a medium to fast setting, maybe in the region of 100 to 150 milliseconds, but if you've got a faster vocal metering, if you're recording a rap vocal that's very fast, set it
with a, you know, faster release setting, perhaps 50 to 100 milliseconds.
Let's have a listen to some printed compression audio examples. This is a Neumann U87 with the vocalist around 5 inches from the mic diaphragm using a pop shield and the compressions provided by an Empirical Labs Distressor using a low threshold and a low ratio. So, the low threshold means that the compressor is providing gain reduction the vast majority of the time, but often not in a significant way due to the low ratio. In this instance, a four to one relatively low ratio with a relatively fast attack and a medium to slower release due to the vocal performance having a slower metering, sounds like this. And this time, a high threshold, or an equivalent of a high threshold, obviously you drive the input of the distressor, but it's a high threshold equivalent and a high ratio. 10-1 sounds like this. And if you weren't hearing that greater difference between those two audio examples, you wouldn't be wrong there. Certainly, hard to discern the differences in the impact of compression there. What I would say is that if you're recording a vocal with a greater dynamic range and you're trying to really level out those differences, I would opt for a high ratio and a high threshold, whereas if it was a vocal performance with a more consistent dynamic range, more consistent power, more consistent emotion, I would opt for a low ratio and a low threshold, so that it's more consistently massaging the dynamic range of that vocal. Of course, if you're more experienced with compression, I've interviewed some producers, for example, Fredrik Nordstrom, a Swedish rock and metal producer, who prints vocals with between 20dB of gain reduction from three different compressors. Just be very careful if you're adopting this technique, that you're definitely pushing it towards your desired target sound and you're not kind of like, chewing up the vocal inappropriately because you're not going to be able to salvage that at the mix stage if you are.
Let's have a brief listen to a more radical approach to printed compression, keeping the high threshold, high ratio settings on the Distressor from earlier, but then taking that signal into a Warm Audio WA2A. With the WA2A providing around 8-10dB of gain reduction, meaning that collectively between the Distressor and the WA2A, there's around 18dB of gain reduction during peaks. And you've got that faster attack setting from the Distressor, but the WA2A with an attack time that is program dependent, but varies around 10 milliseconds. Hopefully you will have heard the compression lifting up the lower level detail of that vocal performance. If you just listen to the, I swear, at the beginning of the line, and there's that little bit of chest resonance and warmth and sort of like almost grit that it's bringing out of the vocal performance there. Importantly though, If that vocal performance was to get any louder and more dynamic, suddenly you could be into the range of 25-30dB of gain reduction and could be chewing that vocal up inappropriately, meaning that unless I've got a microphone splitter, so I've got one uncompressed and one with printed compression, I'm going to have real problems at the mix stage or I'm going to have to re record that vocal, because once you've got printed compression significantly wrong, there's no way of undoing it. So, if you are adopting a more radical approach to compression, really make sure you understand the dynamics of the whole performance that the vocalist is going to go through and really establish that the compression isn't chewing up the vocal inappropriately.
Lastly then let's consider effective headphone cue mix which is vital for eliciting and inspiring vocal performance of any style. What's very difficult though is it tends to be the case that different vocalists require different qualities in the headphone mix. Of course, there can be a whole host of reasons why a vocalist might be singing with pitch issues, but a couple of aspects worth highlighting here is that If the vocalist is tending to sing sharp, it might be due to an excessively loud monitoring level in their headphones, which means they're pushing too hard. Conversely, it can be the case that with a slightly quiet headphone level, that they're not exerting themselves enough and that they're singing slightly flat. Alternative approaches to pitch issues is that sometimes vocalists are obscured by being able to perceive their own natural vocal sound and their own sort of like innate sense of pitch by the headphone mix and it's worth establishing whether the vocalist is feeling a little bit closed in and actually getting them to lift off one of the earphones so they can hear their own natural voice unhindered, but just make sure that the removed headphone side is not bleeding onto the mic. One way of doing so is just to hard pan the backing track over to the relevant side.
Other considerations, if you're really trying to get a vocalist immersed into the mix and you're trying to really get them a vibe that will help them with their performance, it can be valuable to put more bass guitar into the headphone mix and it sort of like provides a slightly more three dimensional sound for them. And of course if it's different challenges you're facing like vocal timing then perhaps stating the obvious but just turn the drums up and give them some sort of like more subdivisions with the metal work in the headphones will often help with vocal timing.
So we've considered scheduling when recording vocals, we've looked at coaching considerations, communication skills, et cetera and we've considered vocal comping. We've also looked at engineering considerations, including acoustic treatment, mic selection and mic placement, use of pop shields, then we looked at handheld vocal recording and then printed compression and headphone monitoring.
And that brings us to the end of this Sound On Sound Recording & Mixing podcast. I hope you found these podcasts on recording drums, then bass, then guitars, then vocals, beneficial and I hope you tune in for next month's podcast, which is on the mix stage.
Thank-you for listening and be sure to check out the show notes page for this episode, where you'll find further information along with web links and details of all the other episodes. Oh, and just before you go, let me point you to the soundonsound.com/podcasts website page where you can explore what's playing on our other channels.