I'm recording vocals with a SM58 mic through a USB audio interface and getting clean audio with no ambient noise. At this point it sounds pretty flat (no reverb or echo from the room). So in my DAW I've been experimenting with different effects to make it sound more full and natural, like you hear on professional recordings. I've tried adding reverb which helps a little. I've run it through a compressor. I'm using an EQ which works well, but I can't find the optimal settings. After a while of playing around it starts to sound weird and unnatural.

So my question is, which frequencies should I boost to make the best sound? What other things can I do to give my audio recordings more depth, make them stand out and sound great in the mix.

  • 1
    Not a full answer but have you tried doubling? Record the vocals two or three times and mix them together.
    – charlie
    Oct 20, 2014 at 19:44
  • 1
    "more depth, make them stand out and sound great in the mix" - could you explain what you mean by 'more depth' ? I understand it's hard to put into words bit there are lots of answers already listed, it might help to know a bit more about how the sound is lacking. Ta Oct 21, 2014 at 8:59

11 Answers 11


One technique that works like magic is by splitting the vocals channel into multiple channels, panning each channel differently, and then applying a different equalizer to each channel. This creates a very subtle difference between what you're hearing in your left and right ear, which is what depth is all about. The difference can be made even more pronounced by offsetting the timing of the tracks in relation to each other by a very small amount.

  • I have never thought of that but it sounds like an awesome idea! What's your approach to equalize a track two times?
    – muffin
    Oct 21, 2014 at 6:43
  • 3
    I did a preliminary test of this and it sounds promising. I'm doing something similar when recording acoustic guitar. I record both plugged in and through a mic at the same time and pan them to opposite channels. Oct 21, 2014 at 13:42
  • 1
    Although there are a lot of great ideas discussed here, I decided to choose this one as the solution because I think it answers the original intent of the question. Sure, using a different mic might give a better input, but there's still the issue of mixing the vocals into the song and making it stand out. I think that's what this solution accomplishes. However, I suggest that you read all of the solutions offered here because there are a lot of good ideas for making your vocals sound better. Oct 29, 2014 at 18:12
  • 1
    This is a terrible idea. [I only just found this in the sidebar from another question, hence being a decade late] This is a great way to create weird phase-cancellation holes in your spectrum & poor mono summation. Avoid.
    – Tetsujin
    Jan 17 at 17:41

Before trying any effects at all, I should change that mic if I were you. The SM58 is an amazing live mic – not just because it's dead sturdy, but also because it has a very focused, direct, "shallow" sound. Great to make vocals come out in a live mix and to avoid feedback and other common trouble – but none of these benefits have any relevance for a studio recording! If you want a full, deep, clear sound, no dynamic mic can hope to match even cheap large-diaphragm condensers you can get today. One of those1, perhaps through a tube preamp, should give you a much better signal to start with, without the dangers of "fixing it in the mix" with compressors and EQs. Of course you may still want to employ some of those, but with a good signal you can do it much more subtle and to the point.

1You may need a pop screen for a large-diaphragm mic. Also, be sure to record in a room with decent acoustics... parallel-walls-clatter-resoncances are deadly to any recording.

  • 1
    @Almo: yes and no. Yes, tube amps in stereo systems have questionable value, because there's not really a good reason to saturate the signal at that point (the mastering engineer will already have done that as much as sensible). However (unlike $100 power chords etc.), there's no doubt that tubes do alter the sound, and in a way that is quite benefitial in particular as a first stage for vocals (adding smooth even harmonics, and boosting RMS for given peak gain without really committing compression). Sure you can also simulate that later in digital, but that incurs aliasing etc. problems. Oct 21, 2014 at 19:33
  • Annie Lennox never did badly with a 58 - several big international hits, in fact ;) I've heard Bono sometimes records on one, too
    – Tetsujin
    Jan 17 at 17:48
  • @Tetsujin yes. But then, Annie Lennox has a voice that wouldn't sound flat even through a walkie-talkie. Not every singer provides such an advantageous starting point. Jan 17 at 20:09
  • For sure - I'm just not a fan of the 'you cannot possibly do it with a 58' club, because you clearly can. It's not something I do a lot, I've a lovely U87 that's my prime vox mic, but I did this a few years ago, where it had to be live to camera, for a movie audition [you can even see & hear my clap at the start, for sync to the video]. Lousy video & I look like lightly boiled sh*t, full of cold, but all there is on this is a multiband comp & a bit of rev. I really don't think the mic is the weak point. youtube.com/watch?v=jt5grnzXuWg
    – Tetsujin
    Jan 18 at 9:49

Try employing parallel compression. Here is a good article about it. Basically, you make a copy of your vocal track. Leave one copy alone and compress the other. That gives you the original loud transients plus a compressed version where the rest of the material is also louder. Sometimes, you'll see this described where the original has very fast or very subtle compression (or even expansion). Hence "parallel".

  • 2
    Parallel compression is a useful technique in a number of situations. But so are a hundred of other particular FX tricks. I don't think it's very helpful to list all those! And FWIW, most transients are actually preserved quite well in regular a non-limiting compressor with attack > 3 ms. Such a rather slow compressor is a much better point to start learning dynamics processing, instead of going parallel right away. But before even doing the most basic EQ, get the mic right! Oct 20, 2014 at 23:27
  • 1
    As your own answer says, @leftaroundabout, a large diaphragm condenser mic and a decent preamp is certainly the ideal setup and it's what I use, too. And I agree that it's not useful to list every FX trick. But for the original poster's situation, this technique is apropos, particularly if running out to buy more equipment isn't an option. There's hardly any recording problem you can't solve with enough money, but many of them are also solvable with a bit of ingenuity.
    – trw
    Oct 21, 2014 at 7:19
  • 1
    I very much agree on that last sentence. It's just that such ingenuity comes only through long experience, experimentation with all kinds of effects – and, IMO, to start you should get a proper understanding of the basics, so you can even appreciate how such and such effect behaves in a more complex setup like parallel-comp. These days, a lot of people think they know all about mixing because they have read so much about a bunch of "miracle effects". Well, to me the records such guys produce usually sound less than ingenious! Oct 21, 2014 at 8:17
  • Or, to turn your argument around: there's hardly any recording problem you can't solve with enough FX-foo, but many of them wouldn't even have surfaced if you had used a better mic-setup in a room with great acoustics! Oct 21, 2014 at 8:20

The only answer possible is "it depends". It depends on what sound you're trying to achieve, it depends on the recording itself. In my experience, I drop everything below 600 Hz and above 5KHz. You may need to give a little boost at 2 KHz, which is the frequency of speech.

Compression helps smooth the sound out in terms of volume but shouldn't affect the tone. Reverb adds depth; there are different types of reverb - the depth of the reverb (how many millisecs, obviously), but also how long before the reverb kicks in.


For lead vocals it is very standard practice for the singer to sing the song twice onto two parallel tracks. That alone adds significant richness the the vocal.

Another thing I'll do requires a real-time compressor: set the threshold relatively low and the ratio fairly low. If you feed that to the vocalist monitor, he will have to put more power out to hear it "correctly". This forces both better tone, and can cover a little for bad mic technique.


I like to use an analog pre-amp which adds harmonic distortion to an otherwise "cold" sounding digital recording. It adds body but could also sound muddy. You could eq out some of the tones giving it a muddy sound. Run an out into the pre and run it back into your recording setup. You could use 2 audio tracks - clean one and pre-amped one - and adjust the volume/eq of each until you get a sound you like.


Former sound-guy here. Here's how it was done ca. 20 years ago:

  • Quality condenser mic in neutral / near dead room. Neuman preferred
  • Quality mic pre-amp
  • Parametric EQ. add splash of treble, cut down bass, control clarity with mid-tones
  • Aphex Multiband compressor with de-ess engaged to boost dynamics
  • BSS 822 or similar sonic maximiser, make it 'deeper'
  • Lexington reverb, make it 'wider'
  • Eventide Ultra Harmoniser for delay / effects

Today, outboard gear is obsolete but you get the idea what is needed to achieve really 'rich' and 'crispy' vocals. it's easy to add similar plug-inns to todays DAWs


  • you could just hire a really good singer :)
  • Yes, deeper and wider, those are what I'm looking for. And I wish I had a really good singer. I like to write music but unfortunately I don't have anyone to sing it but myself. I'm not too bad but I'm no professional. If I was I'd be in a real studio right now letting someone else worry about the mixing :) Oct 21, 2014 at 14:00

The SM-58 is a more-than-capable microphone which has been used on countless studio recordings. The key issues preventing you from getting a lively sound are probably as follows:

You are using a USB audio interface - these range GREATLY in quality. Most A/D converters in them are relatively poor from an audiophile standpoint. If you get a higher end one with good reviews (do your research) you should get much better results (it also makes multitracking easier if you get a device with low latency - look for a newer usb 3.0 implementation or even go back to Firewire... I did and it's the best decision I could have made at the time). It seems like this would be a minor issue, but it honestly makes all the difference in the world.

You may be recording too close to the mic. As the SM-58 documentation so prominently indicates, the 'proximity effect' occurring from being within about 6-10 inches away from the mic will start increasing bass frequencies immensely, vastly altering the timbre. Usually you want to be 1-2 feet away, and use your preamp gain to boost the signal to an acceptable level.

Finally, find a good compressor and learn how to use it. In software there are several free options, one from fxpansion called DCAM which I recommend if you can find where to get a copy (I think you just have to sign up at their website). Another good one (Evaluation version only without paying) is Bombardier Bus Compressor.

EDIT: One more thing I forgot helps a ton, especially when you add the vocals to a mix, but it is the most expensive and difficult part of the whole process. Getting a low pass and ideally also a high pass or even bandpass filter (or two, or three...) and using them to tweak the frequency response can give you any sound you want. EDIT 2: Especially if you dynamically modulate the control paramters, i.e. with a triggered envelope.


This article about the "Exciting Compressor" explains a technique that really pumps up any vocal part. If you use a SM57 or SM58 specifically, here is a little non-invasive modification you can do to your mic that supposedly does miracles. You could also achieve the effect the variable impedence on a PreSonus Eureka preamp.


A little bit of chorus or double-tracking the vocals (singing the same line two or more times and then mixing the takes) can do wonders for a vocal track.


Pro old school engineer here! Still recording many hit artists

Some nice comments. SM58 is a great mic, did you know almost ALL the vocals on all of SHED 7's albums was an SM58! It is how it is mixed. I would favour a high pass filter and lift the bottom end until it sits in the mix. And as suggested! the better the reverb the better the spacing and image. Obviously it does not help with a USB interface as vocals (as with any other instrument) is really all about dynamics, I tend to make my EQ by the type of mic used! Old school, but there are so many hits done that way. Lets face it, EQ like anything else in the chain ads noise and colour. Great, but it also can be a hinderance!

Not the answer you're looking for? Browse other questions tagged or ask your own question.