what level is normal for music & vocals?

I need help because I need to know what level should my voice tracks be and what level should my music track be ,to create a successful mixdown.The reason for this question is because either the voice is to low or to high or the music track is to low or to high.So what level is normal for music & vocals? (please explain in detail).

You seem to be having trouble getting the vocal to “sit” well in the mix, yes? Try these things:

1-compressor on the vocal track to even the levels.
2-put some reverb on the vocal track. Get it to where you think it sounds good, and then take about 25% of the reverb off. :)
3-EQ! cut everything below 100-300Hz. You don’t need all of that low end on a vocal track, and it will allow the other tracks to have some space.

As for the level itself… It’s a mater of taste, mostly. I usually turn the monitors down until I can barely hear anything… Before the sound disappears completely, the last thing you should be able to hear in the vocal track.

Hope that helps a little.

Right – it’s more of an art than a science. And if it WERE a science, we’d have to play with lots more numbers than you’d probably want to bargain with, as a newbie!

No, it’s something you have to do by ear, and tweak until you get it right. Defintely take John’s suggestions. Compression is almost always used on vocals, to even out the dynamics. That way, it won’t be too loud in one place and too quiet in another.

To get an idea of what a compressor is, see Compressor Basics.

For a lot of good pointers on recording, mixing, etc., see audiominds.com. IN particular, see their article Compression, and at the bottom, see the link to Jezar’s tutorial, a true masterpiece written by a true master.

I’d also suggest you read Limey’s Pyramid HERE. Scroll down to the bottom of the page.


For recording, in DAW work, -12 dB average is usually advised.

Peak levels? And turn your 16-bit soundcard into a 14-bit soundcard? I disagree. That's a fine target for folks with 24-bit inputs, leaves plenty of headroom in case someone lets out a honker.

But if you're recording yourself (and so a do-over is easy), and if you're using a 16-bit soundcard, it's better to shoot for -6dB or even -3dB. Frankly, you want to get as close to zero as possible without any clipping. On the other hand, n-Track can't detect every clip, so you want to leave a little buffer (a dB or two).

In general, shoot for as high as possible without clipping, but be happy if you get anything over -6dB. For 24-bit soundcards, -12 is fine, and really it's the analog gear you have to optimize to get the best results and that's a whole nuther ball of wax.

Even with a 24-bit soundcard, I shoot for -6dB or better, partly so that the waves are easier to look at! But if I was recording live I'd leave quite a bit more headroom to avoid clips. It also depends on the nature of the material being recorded and how well controlled it is. Sometimes 12dB of headroom isn't nearly enough!

But I don't think any of this really applies to the question, which was about mix levels rather than record levels.

Actually, I did read all of your post. “-12 dB average” isn’t a well-defined term. It could mean “typical peak levels” or “average peak levels” or “RMS levels”, and maybe more things. What do you mean by average levels? (I guess that’s the question I should have asked in the first place, sorry!)

I believe n-Track’s meters are peak meters (despite being labeled “VU”). I could be wrong about this. In any event, what we need during recording is peak meters. Also, I always set the recording meters to “very fast” to help avoid missing peaks.

You should generally ignore RMS levels when recording and focus on peak levels. RMS levels are for use primarily during mastering. You should pay attention to peak levels during recording, because that’s the measure of how much “room” is left unused in the recording media.

RMS levels can give an indication of S/N ratio in the recording, but unless the signal is unusually peaky, like vibraphone, there’s little point in worrying about it. Furthermore, unless you’re considering using outboard limiting or compression, there’s nothing to do about it anyway (other than upgrade to 24-bit converters).

the median level at which the needle of a well-slugged VU sits
OK, I understand. I suggest that this is not the best thing to pay attention to when recording. Instead, what you need to watch (or really, anticipate), is the maximum peak level during the performance. The numbers printed above the recording meters are an excellent indication of this (though not perfect). (You can click on them to reset them.)