HiFi++: a Unified Framework for Neural Vocoding, Bandwidth Extension and Speech Enhancement
Geek Repo:Geek Repo
Github PK Tool:Github PK Tool
williamsz opened this issue 2 years ago · comments
Hi, I traned a model for the Bandwidth expansion task using the code and VCTK dataset. But I got bad generated wavs, which contain Single frequency lines in Spectrogram , as shown in the figure