STFT Spectral Loss for Training a Neural Speech Waveform Model.

Shinji Takaki, Toru Nakashika, Xin Wang, Junichi Yamagishi

Browse the full ICASSP paper archive.