1 paper
Yeongtae Hwang, Hyemin Cho, Hongsun Yang +3
For training the sequence-to-sequence voice conversion model, we need to handle an issue of insufficient data about the number of speech pairs which consist of the same utterance.…