I am working with audio of someone with tremors in their voice- I think they call it vocal paralysis or something. Are there software applications that break sound down into it's individual waveforms using FFT or something similar that would allow me to manipulate these tremors more than I could by automating volume changes?

Anyone know if the Adobe VoCo team has beta users in the wild?

Many thanks!

  • Great and very interesting question! Looking forward to replies on this. Commented Aug 10, 2017 at 17:59
  • Thanks for everyone's suggestions. Currently, the solutions out there don't appear to be able to help in this particular use case. I was able to get in touch with the folks at Adobe about VoCo and the technology is not there yet to create entire sentences (with inflection). I will leave this question open in case something new pops up on the radar. I would leave a sample audio file for folks to play with but this is a VIP type situation and I don't have permission to share the project right now. Thanks for all of your help. Commented Aug 22, 2017 at 4:23

2 Answers 2


You can try Spear http://www.klingbeil.com/spear/ it deconstructs sound into sinusoidal parts. Kind of an old one but it's free and might be helpful. Also Spectral Repair in RX might be able to help.

Curious what the voice sounds like. I'm assuming you are trying to help with legibility. A person's voice is so unique and characteristic that altering it too much should be taken into consideration.

  • Many thanks Diego- I'm going to check it out now. Commented Aug 11, 2017 at 3:39
  • Let us know what are the results :)
    – Diego
    Commented Aug 15, 2017 at 0:11
  • Thanks again Diego. I wasn't able to improve the quality enough to get this to work. I appreciate your suggestion! Commented Aug 22, 2017 at 4:21

There are some general spectral (fft) editors, of which Spear (free) is one of the most powerful (even more than Izotope RX or the Sony Editor), but these should not really be necessary, if it is just a volume modification that you want to do.

That is easily doable with a lot of Wave-Editors (the choice depending on your platform, Audacity is a common cross-platform open-source solution).

A Wave-Editor would also allow you to cut and edit stutters.

Then there is specialized software to edit vocal intonation, pitch as well as vibrato and tremolo, timing and also cutting. Most comprehensive solution here imho is the Melodyne Editor by Celemony.

  • I will take a look at Melodyne. Many thanks! Commented Aug 11, 2017 at 16:01
  • Thanks for your suggestion. I wasn't able to use Spear to improve the quality, but nothing seemed to work. They were too many dropped vowel sounds to repair. Initially, I thought that there was a recurring "tremolo" affect that was cycling fast enough that I would be able to use a delay to smooth it out. Commented Aug 22, 2017 at 4:20
  • 1
    hard to understand what you actually want to do without a sample (which you probably can not give). if the vowels are dropped, they could be copied from somewhere else and spliced together (faded). easiest in melodyne. if no pitch is involved, audacity would be enough. and probably is the easiest to get an understanding what you want anyway..
    – Eike
    Commented Aug 23, 2017 at 10:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.