Normalizing accross waveforms for an Album.

  • Thread starter Thread starter Change of POETS
  • Start date Start date
Change of POETS

Change of POETS

New member
Okay, since this is the first album I've used Audition for, I'm wondering the best method for ensuring proper volumes accross my range of tracks.

Most of the songs have been compressed/limited with the use of T-Racks and hover about the -1db mark in Audition. There seems to be some minor discrepancy between a few of the files, and I'd like to go about fixing this in the best manner, without experiencing any clipping.

My natural instinct tells me to use Hard limiting with some very soft settings to simply increase the volume of the overall file with very little limiting being applied... but, I'm wondering although it may be a longer and more gratuitous process, should I use Normalization, within specific parts of each file that don't meet my -1db requirement? That way I'm not processing the entire file, but merely increasing volume in areas where it dips a bit too low...

Now, I don't want to destroy the dynamics of the files either, and I know I run a risk of doing so... but is it even worth my time to do this? If the discrepancies are so minimal that only my "overly anal" engineering ear-set will notice, should I put in the extra effort?

Any advice is appreciated.
 
CD Architect (if you have it) allows you to adjust levels just prior to burning a CD. I haven't tried it yet, because I tweak everything to death in the mix.;)
Seems like the "batch normalize" function is made for this problem, too.

I also occasionally use T-Racks and when I do I like to boost the input on the compressor & limiter a few dB to bring up the average. I don't like to do this for "finished" mixes but a lot of people want to hear a slammed signal on a preliminary mix CD...how does Bruce do that :rolleyes thing, anyhow?
 
CoP - I think I don't understand your question exactly.

Are the levels of the tracks right relative to each other? In that case, Hard Limit and then Normalize to get a bit of extra volume on the whole song, right? No harm done, right?
 
Change of POETS said:
but, I'm wondering although it may be a longer and more gratuitous process, should I use Normalization, within specific parts of each file that don't meet my -1db requirement?
Click to expand...

Do you mean raising the volume on quiet parts of a song? That would be exactly screwing up the dynamics. What you should be doing is making each song sound like it is the same volume as the rest. The peak volume matters very little, as we don't hear peaks, we hear RMS as volume.

You cant look at a wav file and tell if it sounds good, so stop trying.
 
Dobro,

I'm thinking you're right. You seemed to grasp what I was saying... The problem lies in overall volume of the track. Each track seperately, sounds great... but when burned to CD, Track 1 is a slight bit louder than track 2, etc... I want to make them all near the same volume, to provide congruence to the entire album. I suppose Hard limiting is the best way to go about it.
 
You're doing the right thing...LISTENING instead of just looking at RMS values. Obviously, a song with a ton of low end could have very high "RMS" values, but not SOUND very loud compared to other tracks.

It may be incredibly backwards, but what I do is load all the tracks in the multitrack...maybe 1-12. Play the middle of the first track against the middle of another track. Lightly hard limit to get values up.

Then I usually take the time to line them up end to end and check the fades and intros to make sure nothing is too jarring and that the songs fit together...that's how I sequence a CD. Something can look good on paper and not sound right together.
 
Remember you can also turn down the loud ones to match the quiet ones. Wavelab has a feature called a Meta-normalizer that does what you are talking about.
 
Farview said:
Remember you can also turn down the loud ones
Click to expand...
Okay, out of that sentence, I understood nothing after "remember," lol.
 
chrisharris said:
You're doing the right thing...LISTENING instead of just looking at RMS values. Obviously, a song with a ton of low end could have very high "RMS" values, but not SOUND very loud compared to other tracks.

It may be incredibly backwards, but what I do is load all the tracks in the multitrack...maybe 1-12. Play the middle of the first track against the middle of another track. Lightly hard limit to get values up.

Then I usually take the time to line them up end to end and check the fades and intros to make sure nothing is too jarring and that the songs fit together...that's how I sequence a CD. Something can look good on paper and not sound right together.
Click to expand...
That's a great idea, Chris. I suppose I'll do that. Load all 13 track into one session, and go from each track to the next.

Thanks...

Farview, what is this "turn down" you speak of? :confused:

hehehe
 
I forget which version of Audition / Cool Edit "Group Waveform Normalise" was introduced in, but that sounds appropriate to your problem.
 
Change of Poets:

Okay, now that I understand what you're trying to do, I'll add this. I figured it out on my own too, so I'm really proud of it. You've got an album's worth of songs, so how to set the level on each song so that it sounds right? Right? Right!

Solo the vocal in each song. Make a note of its average level. (In my songs, the vocal's often hitting between -12 and -6 dB.) Then adjust the master volume for each mix so that the LEVEL OF THE VOCAL IS RIGHT, RELATIVE TO THE LEVEL OF THE VOCALS IN THE OTHER SONGS. Quiet songs have quieter vocals, screamers are louder - that's the rule of thumb. When you're done adjusting all the different master volumes in Multitrack, THEN do the mixdowns of all the different mixes.

This works really well if your vocals are the centerpiece of the song.
 
You must log in or register to reply here.
Back
Top