Normalize All – Must I split audio between originally loud and soft bits?

[John Rofrano]

Yup, that’s how normalize all works. It just raises everything equally to make sure the loudest sound is at 0dBFS. You have several options if both talkers are on the same channel:

1. Add a volume envelope and raise and lower the volume as needed
2. Split the events and adjust their volumes individually
3. Add a compressor to the track and set the threshold very low so that it raises the low audio while compressing the loud audio.

I usually take option 1 and use a volume envelope. Option 3 is the laziest approach but sometimes it produces acceptable audio with minimal effort.

If you have Sound Forge you could open the audio and highlight only the low audio and normalize it separate from the loud audio. If you don’t have Sound Forge you might consider getting it. A lot of audio post work requires this level of precision.

Of course if the talkers are on left and right channels then just split the channels on the event into left and right on separate tracks and adjust the track volume accordingly. (and if they are not, seriously consider recording them this way the next time)

~jr

http://www.johnrofrano.com
http://www.vasst.com

[Stephen Mann]

Download “Levelator”. It’s free and it’s magic. It is made for exactly this kind of problem.

Steve Mann
MannMade Digital Video
http://www.mmdv.com

[Jill Simpson]

Thank you Steve and Thank you John,

If Levelator works well in my case, I’ll use that.

I may use it in combination with volume envelopes for fine tuning.