Page 1 of 1

How to transform a neutral sound copying another waveform

Permanent link to this postPosted: Wed Mar 08, 2017 3:33 pm
Posted by amy24
Hello!! I would like to know if it is possible to transform a neutral sound(e.g. "aaaaaaaaa", neutral) copying another waveform of another sound, for example, a person making a question. In the end, the result should be "aaaaaaaaa" but with the rhythm and pitch changed as if we were making a question. Is it possible to do this with audacity? How? Should I use another program? Thanks.

Re: How to transform a neutral sound copying another wavefor

Permanent link to this postPosted: Wed Mar 08, 2017 7:18 pm
Posted by kozikowski
While I'm pretty sure one of the other elves can think of a way to do that, I think chances of that working aren't very good because of everything else going on.

I'm guessing your work is not a studio production. Anything you change taken from real life is going to sound funny because not only is the voice going to change, but the dog barking and traffic sounds behind the voice are going to change, too.

Context is terrifically important to pull off a special effect. Not only must a voice change, but everything else has to not change. Audacity is terrible at splitting voices or instruments or sounds from each other in a mixed show.

That and you can't change the pitch of a voice more than about one piano key or so without it sounding funny. We're famously unable to make you sound like the wrong gender or a grownup/kid.

Koz

Re: How to transform a neutral sound copying another wavefor

Permanent link to this postPosted: Wed Mar 08, 2017 8:08 pm
Posted by steve
amy24 wrote: the result should be "aaaaaaaaa" but with the rhythm and pitch changed as if we were making a question.

Rhythm change is fairly straightforward - you need an "envelope follower" so that the amplitude of one track controls the amount of "gain" (amplification) of another track. There's a Nyquist script available to do that: viewtopic.php?p=291614#p291614

Changing the pitch to match is more complicated, and with speech as the "control" track, I doubt that it will work in the way that you envisage because speech is not strongly pitched.