Mix Audio Automatic Volume Possible?

[Frank62]

Hello,
we have a problem: They delivered us the normal soundtrack plus an audio description for the visually impaired. But the latter hadn't been mixed to the original track, we have to do it ourselvses, and have zero experience with this.
The volume of the normal soundtrack has to be mixed with the description track in the usual way: Each time when there is spoken in the description the volume of the normal track has to be lowered, otherwise not.
Is there a function in avisynth to solve this? Or can you help with another software hint?
Thanks in advance!

[johnmeyer]

I do this all the time in my NLE (Vegas) and have even written several scripts to automate the process. If you have an NLE (Non-Linear Editor), that would be what I'd use.

Someone else will have to describe how to do this in AVISynth, but to help you figure out what other tools you might already have that can do what you want, the search term you need to use is "audio duck." That is what it is called when you temporarily lower the volume of the main program in order that some other audio, like narration, can be better heard.

[hello_hello]

Under the Effects menu, Audacity has an Auto Duck effect. I hadn't used it before so I gave it a spin. It seems to do what it should. There's instructional videos on YouTube. It has a few settings to play with if need be, but it's pretty easy to use. For the attached sample I peak normalised the music track before ducking, but not the speech track, so the latter was a bit quieter. You can still hear it though.

test.m4a (https://files.videohelp.com/u/210984/test.m4a)

[Frank62]

"audio duck."

Thanks a lot! That term will save us! The master has to be delivered today... :thanks:

[Frank62]

Under the Effects menu, Audacity has an Auto Duck effect.

Works well for us! Thanks! :thanks: