Medix-AI/coqui-tts
Medix-AI
/
coqui-tts
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

add support for synthesize using variable size external embedding and add bugfix in scipy.io import

This commit is contained in:
Edresson 2020-07-30 03:51:20 -03:00 committed by erogol
parent f91b9eeda1
commit 496a4be3e3
2 changed files with 0 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
mozilla_voice_tts/bin/train_tts.py
View File

@ -523,7 +523,6 @@ def main(args): # pylint: disable=redefined-outer-name
"a previously trained model." "a previously trained model."
elif c.use_external_speaker_embedding_file and c.external_speaker_embedding_file: # if start new train using External Embedding file elif c.use_external_speaker_embedding_file and c.external_speaker_embedding_file: # if start new train using External Embedding file
speaker_mapping = load_speaker_mapping(c.external_speaker_embedding_file) speaker_mapping = load_speaker_mapping(c.external_speaker_embedding_file)
print(speaker_mapping)
speaker_embedding_dim = len(speaker_mapping[list(speaker_mapping.keys())[0]]['embedding']) speaker_embedding_dim = len(speaker_mapping[list(speaker_mapping.keys())[0]]['embedding'])
elif c.use_external_speaker_embedding_file and not c.external_speaker_embedding_file: # if start new train using External Embedding file and don't pass external embedding file elif c.use_external_speaker_embedding_file and not c.external_speaker_embedding_file: # if start new train using External Embedding file and don't pass external embedding file
raise "use_external_speaker_embedding_file is True, so you need pass a external speaker embedding file, run GE2E-Speaker_Encoder-ExtractSpeakerEmbeddings-by-sample.ipynb or AngularPrototypical-Speaker_Encoder-ExtractSpeakerEmbeddings-by-sample.ipynb notebook in notebooks/ folder" raise "use_external_speaker_embedding_file is True, so you need pass a external speaker embedding file, run GE2E-Speaker_Encoder-ExtractSpeakerEmbeddings-by-sample.ipynb or AngularPrototypical-Speaker_Encoder-ExtractSpeakerEmbeddings-by-sample.ipynb notebook in notebooks/ folder"

3
mozilla_voice_tts/tts/utils/synthesis.py
View File

@ -210,13 +210,10 @@ def synthesis(model,
if backend == 'torch': if backend == 'torch':
if speaker_id is not None: if speaker_id is not None:
speaker_id = id_to_torch(speaker_id, cuda=use_cuda) speaker_id = id_to_torch(speaker_id, cuda=use_cuda)
<<<<<<< HEAD:mozilla_voice_tts/tts/utils/synthesis.py
if speaker_embedding is not None: if speaker_embedding is not None:
speaker_embedding = embedding_to_torch(speaker_embedding, cuda=use_cuda) speaker_embedding = embedding_to_torch(speaker_embedding, cuda=use_cuda)
=======
>>>>>>> Added support for Tacotron2 GST + abbility to condition style input with wav or tokens:utils/synthesis.py
if not isinstance(style_mel, dict): if not isinstance(style_mel, dict):
style_mel = numpy_to_torch(style_mel, torch.float, cuda=use_cuda) style_mel = numpy_to_torch(style_mel, torch.float, cuda=use_cuda)
inputs = numpy_to_torch(inputs, torch.long, cuda=use_cuda) inputs = numpy_to_torch(inputs, torch.long, cuda=use_cuda)