search

janniss91 / SpeedyPanther

An repo that contains different Speech Processing projects and applications.
0 stars 0 forks source link

Learn Spectral envelope from data for vowels and consonants #2

Open janniss91 opened 1 year ago

janniss91 commented 1 year ago

To generate a general form of a speech sound, an impulse train or noise can be used. In order to do that, a spectral envelope must be learnt for the specific sounds from data.

  1. Find a speech dataset with annotated sounds.
  2. Fit spectral envelopes to the data with the annotated sounds.
  3. Use the spectral envelopes to generate (very simple) speech sounds.