Using Audio Characteristics for another recording?

Special Interest Groups Audio Processing
1

I have two audio files, both are a voice of the same person.

I want to use both in the same video I’m creating, but one of the voices
sounds somewhat different then the other.

Is there a way to save all of the characteristics of the preferred voice and apply it to the other file
to get them to sound the same?

Thanks,

Ray

2

There are a number of reasons why the voices may sound different, and several of them you can do nothing about

  1. The person was talking / singing differently
  2. The place where the recording took place had different reverberation characteristics
  3. The microphone placement was different
  4. The person was in a different emotional condition
    5)… .

You will not be able to make them sound exactly the same, but one thing that you can do (and I’m guessing that this is the sort of thing that you had in mind) is to adjust the equalisation (the tonal balance) to make one more similar to the other.

Unfortunately I know of no automatic way of doing this, but it can be done manually.

You can get some visual clues by looking at the spectrum (Analyze menu) of similar parts in the two recordings. You will probably notice that one has more/less bass than the other, more/less treble, peaks/troughs in certain places. By using the Equalizer effect (Effects menu) you can adjust the frequency balance so that they both show similar spectrums.

The two things that make this a bit trick: 1) there is no automatic tool for doing it, and 2) You can not do it in real time (Audacity does not process in real time).

How I would approach it would be to Export the track that I wanted to change as a WAV file and play it in a media player that had a “graphic equaliser” (such as WinAmp). I could then adjust the equaliser settings while it was playing, and compare with the other track. When I had got it as close as possible, I would note down the equaliser settings and set the Audacity Equalizer effect to use those settings.

This would be very much easier if the equaliser could be used in real time, and this is one of the most requested features.

I suggest that the Equalizer can be made adjustable as the sound file is being played .

http://audacityteam.org/wiki/index.php?title=Feature_Requests#Highest-rated
Perhaps you would like to add your vote to this feature request?

3

Of course, it’s a given that all these fancy tools only work if the actor is saying the same words in both tracks.

If you analyze “Lumpy Dump Tom the Luddite” and “Sister Suzy Sells Sea Shells,” you will get very, very different analysis even with with same actor on the same microphone in the same track.

Hollywood maintains very expensive, quiet sound studios and gives awards to people who can do this sort of job.

Koz