It is worth noting that the DNN, LSTM, and BLSTM model structures used for the VoiceBank-DEMAND dataset were not optimized. Instead, the FCN model is reimplemented based on research: S.-W. Fu, Y. Tsao ...
Floder's name means the generation of Synthesizer V and the language of your voicebank. For example, english-arpabet means English voicebank. 文件夹的名称代表 ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback