formant-analyzer icon indicating copy to clipboard operation
formant-analyzer copied to clipboard

Published 20 hours ago • verified publisher icon fulldecent

Try smart vowel extraction

Open fulldecent opened this issue 9 years ago • 0 comments

Some original research to do.

Make a model for vowel utterances which expects:

  • Three formants in 500 Hz - 3000 Hz range
  • Little change in formants for a minimum of (1/10 sec?)
  • LPC model size ~10
  • Model error (consonants, other stuff) below a certain threshold

Then check which parts of the signal meet this model requirements. That is the vowel.

fulldecent avatar Nov 25 '15 03:11 fulldecent