Tomiinek / Multilingual_Text_to_Speech

An implementation of Tacotron 2 that supports multilingual experiments with parameter-sharing, code-switching, and voice cloning.
MIT License
826 stars 157 forks source link

Params.py issue #75

Closed XSXIEYCH closed 2 years ago

XSXIEYCH commented 2 years ago

Hello I found the code miss some parameter when I run the project, File name: Params.py issue mel_normalize_mean mel_normalize_variance But I don't know how I can assign values to them in addition, I can not run here,(file: tacotron2.py, line:145) The system prompts that the parameter encoded must be of type Int or long def _add_conditional_embedding(self, encoded, layer, condition): """Compute speaker (lang.) embedding and concat it to the encoder output.""" embedded = layer(encoded if condition is None else condition) return torch.cat((encoded, embedded), dim=-1) so I do some change encoded=encoded.long() But I got other error "index out of range is self"