I received an email this week asking about how to sing when recording. I’ve re-parsed the sentences in the email so that the questions flow for the purposes of this article.
“Hi Mark,
Do you have any articles [or advice] on how to sing when recording? I feel like my recorded voice sounds both harsh and dull.
I wonder whether I sing too forcefully to try and get emotional intensity. Do I need to improve my loud singing? Or is it about singing differently when recording?
When I listen to my favourite artists’ recordings they sound alive, intense and still have nice higher resonances going on. Would getting a mic that can deal with louder singing help with not losing the higher resonances?”
What can we do?
Sorry to hear you’re having trouble with your vocal recordings. I happen to spend a lot of time recording and working on this stuff, so let’s dive into it…
Quick caveat – without having worked with someone’s voice (let alone considered their recording setup), it’s impossible to say for definite what’s going on. That said, here are my thoughts….
In my opinion, there’s two halves to the answer:
1) How good is what you’re sticking into the mic?
2) How good is what you’re doing with the recorded take?
1) How good is what you’re sticking into the mic?
A microphone does not lie… but albums do
(This is covered in point 5 of this article)
A microphone is literally just an ear without a brain. It captures what it hears. The honest truth is that most artists who are not enjoying their recorded vocals are really just properly hearing how they actually sound. In this singers’ case, they mention a specific singer that leads me to think they are spending a lot of time in chest voice and potentially forcing the top of their voice (as that is how the named singer often sings).
This will generally lead to a lack of connection/bridging that would be sufficient to a) encourage ease into the voice and b) encourage those higher resonances into the voice. These can ONLY occur with good bridging as then you gain access to new registers of resonance that create a thicker fuller and juicier sound. Basic advice to avoid harshness would be to change the key of the song (and/or the melody lines in the song) to be so accessible for your voice that you’re never straining. This will allow better resonance (wherever you are) and also create a smoother less harsh vocal take.
Changing microphones?
This might improve things a little, as some microphones will sound better on some voices than others, but I don’t think it’s the real answer to such problems. Case in point – lot of my best recordings were done with a second hand SM57 that I got for £40, and allegedly much of U2’s material (and other artists) was done with the same mic. It’s a good mic but not the best and not the most expensive. As such, whilst certain mics might flatter your voice more than others (e.g. large diaphragm condensers of varying price), the biggest improvement is not going to be had from changing microphone, but having a great sound source to begin with (i.e. improving your singing, full stop).
Why don’t I sound like my favourite artist?
You’ve got to cut yourself a LOT of slack. As covered in the article linked above, and also in this one on Aguilera taking 100 takes to get this one song right, albums are CONSTANTLY lying to you about how good the singer is. Your favourite artist may well be completely stuck in chest and yelling those top notes, maybe even be horrendously flat, which would be hideous to hear if you were in the room with him, but through studio production such takes can be massaged into something excellent sounding. So don’t believe what you’re hearing or hold yourself to that standard.
2) How good is what you’re doing with the recorded take?
Despite everything I’ve said above, there is a lot you can do to improve your recorded take in your digital audio workstation. You can check out many many Youtube tutorials for explanations on vocal treatment, but I’ll give you a basic rundown of my process here.
a) get your distance to the mic right (and have a pop filter) – play around with it, but anywhere from 4-12 inches is fairly normal. I’m usually 4-6 inches away but I use a dynamic mic for recording a lot at the piano. This enables me to play and sing and get good noise rejection from my vocal take, whereas a condenser can introduce too much bleed.
b) high-pass the vocal – all microphones suffer from the proximity effect, adding a lot of low rumble to the sound. Add a high-pass filter at (say) 100Hz to cut out unnecessary low end rumble. If you’re singing lower you might want the frequency to be a bit lower as well, but 50-100Hz is a good ballpark. If you feel your voice is simultaneously harsh AND dull, high-passing to cut out unnecessary low end will be a good start.
c) subtractive EQ – All microphones and rooms have some nasty resonances floating around in every take, so using a graphic equaliser to scan the frequency range and notch out any nasty sounding bits. Best way I’ve found is to boost a frequency band a LOT, then move it around til I notice something that seems to resonate in a nasty and unpleasing way, then cut it by about 3-5 dB with a very narrow Q width (so it’s only really targeting that frequency rather than a big chunk of your take). Sometimes there’s only 1 or 2 of these frequency spots, but I’ve had as many as 3-5 that bother me in takes.
d) additive/enhancing EQ – by subtracting out nasty frequencies you’ll have made a LOT of headroom for your vocal. This means you’ve got space to add some EQ. Aim to enhance what is already there rather than try to put in something that wasn’t there to begin with. I personally find I only need a very minimal 1dB shelf boost at anywhere around 8-12kHz to add a bit of highs to the sound (bear in mind I’m using a dynamic microphone, not a condenser microphone, and these tend to have less top end response anyway)… that said, often if I’ve done a good take and done subtractive EQ well, I don’t even need that shelf boost depending on the song I’ve recorded.
e) compression – Have a look at Graham Cochrane’s tutorials on his Youtube channel, Recording Revolution. He’s got a great explanation of how to set and use compression to get a clean vocal mix. You’ll find that the better you’ve done the previous steps, the more impressively the vocal take “pops”. He uses two compressors chained in series, with light to medium compression doing 2-3dB of gain reduction. I personally find I want as little as possible, so I use three compressors in series with each doing 1dB of gain reduction or less each (just my current preference for my voice).
Throughout all this make sure your interface isn’t clipping or that you’re recording too hot. There’s obviously a lot to this stage, and many of you reading this may already be doing some of this. However, if you’re not liking your best vocal takes so far, follow the above steps fully and thoroughly, and that should help a lot.
Conclusion: If you follow the above advice…
If you move the song to a key that makes singing the song almost effortless, record at the suggested distance, and follow the above steps to “treat” your vocal take, you’ll have eliminated all the most obvious and easiest to eliminate sources of harshness and dullness that can be addressed in a single attempt (in my opinion).
If you’re still having problems at this stage, I’d suggest investing in your voice to improve your ability to deliver great vocals. The better the sound source is (your voice), the less you really have to do in the mix to fix things. I know that seems like an easy or even trite thing to say, but it really does make all the difference in the world if the voice functions so well that it can just do good to great takes off the bat.
Learn More: Related Articles
If you want to learn more about recording, you can find out more by visiting these related articles:
Recorded vs Live Performance: Which is harder?
Recording Studio on a Budget
Vocal Comping to Get the Best Vocal Take
Live vs Recorded Vocals: Our own double standards
Five Common Vocal Misconceptions (check number 5)