An auto-encoder for neural pitch transformations

Frederik Bous, Axel Roebel
UMR9912 STMS | IRCAM - CNRS - Sorbonne Université | Paris, France













10-207-4

Model Ground truth PaN Vocoder $s_b=2$ (speech) $s_b=3$ (speech) $s_b=8$ (speech) $s_b=3$ (speech + singing)
2200
1760
1320
880
440
0
-440
-880
-1320
-1760
-2200

The sounds used on this page are under Copyright © 2021 Ircam, Institut de recherche et coordination acoustique/musique.