how to even out levels in final mix

D

downonthestreet

New member
I'm editing an interview I did recently, which will end up on my blog as a podcast. I used an omnidirectional mike placed in the centre of a table, and there were 4 interviewees sitting around at varying distances from the mike, and speaking at varying volumes. The recording levels were probably too high.

I've ended up with a recording that is very uneven in voice levels. Much of the recording went into the red (minimal distortion - lucky it's only voices) and these parts are peaked right out. The quieter voices are considerably lower in volume. Because of the peaked out sections, normalising isn't effective.

What should I be using to bring the low volume voices up to a level closer to the loudest voices, so that the final result gives an impression of evenness of volume, and so the quietest interviewee voices are as easy to hear as the loudest ones, please?
 
downonthestreet said:
I'm editing an interview I did recently, which will end up on my blog as a podcast. I used an omnidirectional mike placed in the centre of a table, and there were 4 interviewees sitting around at varying distances from the mike, and speaking at varying volumes. The recording levels were probably too high.

I've ended up with a recording that is very uneven in voice levels. Much of the recording went into the red (minimal distortion - lucky it's only voices) and these parts are peaked right out. The quieter voices are considerably lower in volume. Because of the peaked out sections, normalising isn't effective.

What should I be using to bring the low volume voices up to a level closer to the loudest voices, so that the final result gives an impression of evenness of volume, and so the quietest interviewee voices are as easy to hear as the loudest ones, please?
Click to expand...

wow man that is a rough one you can't fix it exactly the way you want.but you can try adding some compression.I can't tell you how much without listen but if you want you could send me the clip and I can see what I can do
 
Hi Jammer429

Thanks for your fast response and generous offer - much appreciated. I was really just wanting to find out what I should be using to try to even up the vols...it seems that compression is the way to go? I was hoping for a quick solution, but I know I can go through the recording and manually bring up the vol on the low voices.

Rather than inflict the task on you, I'll have a go with some compression settings myself and see how it goes, then if the result is not good, I'll have to just bite the bullet and go through it all manually.

Would be most interested if you have any suggestions for compression settings to try initially?
 
sure man no problem you might want to try heavy compression now after you do that you might not have it that loud I don't know how bad the clip is but you should be able to bring to vol up on the track in you editing program if you don''t feel it is loud enough then I will gladly tell you how to get it louder
 
I'd import it into the MultiTrack view and use the volume envelopes to manually bring the louder parts down in volume, or the quieter parts up in volume. Then maybe top it off with compression to make it really evened out.

But a highly dynamic track with lots of quiet parts and loud parts can be too much for a compressor, or it can be very hard to do without having the effects being noticeable.
 
OK, thanks a lot guys. Will try your suggestions. Looks like there's no easy way out. My fault for having the levels too far up at source.

Cheers
Ross
PS: Wonder what would happen if I decreased the amplitude to bring the peaks down, then applied Normalise?
 
Normalize just finds the peaks-> figures out how far they are from 0dbfs-> turns up the volume of everything by that amount. There is no point to it in normal circumstances.

You want to use compression with a relatively fast attack and slow release. The ratio will depend on how much of a difference there is between the loud voices and the quiet ones. Make sure the quiet ones are still into the reduction by about a db, otherwise they will sound weak.
 
Thanks, Farview - I'll try the sort of settings you mention. Not sure what you mean by "make sure the quiet ones are still into the reduction by about a db", but I'll probably get to understand when I try compressing as you recommend.
 
Look at the reduction meter on your compressor, make sure that the quiet voices still make the reduction meter move a bit.
 
Aa?

This usually Works 4 me, Try going into your vocal file, Effects, Dynamics processing and use the compander preset. If not get your hands on Waves Diamond. All there Compressors do an amazing job at this
 
You must log in or register to reply here.
Back
Top