FORUMS: list search recent posts

Re: group normalize problem

COW Forums : Adobe Audition

Respond to this post   •   Return to posts index   •   Read entire thread

John Smith
Re: group normalize problem
on Oct 8, 2006 at 10:12:14 pm

The two tracks will appear next to one another on a cdr of customly arranged songs. The two songs won't be mixed together, but because they are side by side in song order, their volume differences are noticeable as one song leds to the next in play mode.

Really what I am doing with these two songs is that I'm putting Adobe Audition's claim about the Group Waveform Normalize effect to the test. See their help file statement below about making "sure that all tracks on the CD have a consistent volume."

Normalizing groups of files
If you're getting ready to master an audio CD, using Group Waveform Normalize is a great way to make sure that all tracks on the CD have a consistent volume.

I challenge this statement and I don't think Adobe Audition can do this consistently. In addition normalizing peak values I think RMS normalization is also important for this process to work properly. Here are more help file tips:

Use the following options in the Normalize tab to specify how you want to normalize the waveforms:

Specifies whether to normalize to an average level or a specific level you enter in decibels.

Note: The Normalization option doesn't use percentages, unlike the Normalize effect, because it is RMS-based rather than peak-based.

Use Equal Loudness Contour

Applies an equal loudness contour, where the middle frequencies are most important. Because the human ear is much more sensitive to frequencies between 2 kHz and 4 kHz, two different pieces of audio with the same RMS amplitude but with different frequencies will have different apparent volumes. Select this option to ensure that audio has the same perceived loudness, regardless of what frequencies are present.

I was hoping the Equal Loudness Contour would correct what I hear as a different "perceived loudness" between both songs. But it did not do anything with the fact that one song is clearly louder than the other one.

As you suggested, I applied a hard limit (of 4 dB in this case) to the quieter song and this effect has helped to make both songs more equal in terms of "perceived loudness." The hard limiting effect is a great tool and I love working with it, but if I had to do this "manual" correction to several songs in a collection of 100 mp3's going on a cdr, for example, it would be quite a laborous task. And even with this I'm guessing at the dB value to enter; in this case, 4 dB was a good choice but that was luck.

So I'm back to wanting Adobe Audition to do the work for me; that is, equalize the RMS perceived loudness values
of two or more songs so that the volume levels from one song to the next will sound consistent or as close as possible.

This sounds like it should be simple to do but my experience has shown that it can sometimes be a challenge, particularly with songs that don't have closely matched RMS loudness values. In the end I'm trying to get Adobe Audition to do this fine tuning automatically, but I don't think it can in all cases. My conclusion is that manual corrections "after" RMS normalization may well be necessary with one or more songs that are being prepared for burning on a cdr. And I'm not being too picky in my judgements, but if one song is considerably louder than the next in a set of songs then I feel RMS normalization has failed me.

Posts IndexRead Thread 

Current Message Thread:

© 2018 All Rights Reserved