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

Fix the bug in extract tts spectrograms

This commit is contained in:
Edresson 2021-11-22 13:43:44 -03:00 committed by Eren Gölge
parent 2b2cecaea2
commit 85418ffeaa
2 changed files with 6 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
TTS/bin/extract_tts_spectrograms.py
View File

@ -37,8 +37,8 @@ def setup_loader(ap, r, verbose=False):
enable_eos_bos=c.enable_eos_bos_chars, enable_eos_bos=c.enable_eos_bos_chars,
use_noise_augment=False, use_noise_augment=False,
verbose=verbose, verbose=verbose,
speaker_id_mapping=speaker_manager.speaker_ids, speaker_id_mapping=speaker_manager.speaker_ids if c.use_speaker_embedding else None,
d_vector_mapping=speaker_manager.d_vectors if c.use_speaker_embedding and c.use_d_vector_file else None, d_vector_mapping=speaker_manager.d_vectors if c.use_d_vector_file else None,
) )
if c.use_phonemes and c.compute_input_seq_cache: if c.use_phonemes and c.compute_input_seq_cache:
@ -235,13 +235,14 @@ def main(args): # pylint: disable=redefined-outer-name
meta_data = meta_data_train + meta_data_eval meta_data = meta_data_train + meta_data_eval
# init speaker manager # init speaker manager
if config.use_speaker_embedding: if c.use_speaker_embedding:
speaker_manager = SpeakerManager(data_items=meta_data) speaker_manager = SpeakerManager(data_items=meta_data)
elif config.use_d_vector_file: elif c.use_d_vector_file:
speaker_manager = SpeakerManager(d_vectors_file_path=c.d_vector_file) speaker_manager = SpeakerManager(d_vectors_file_path=c.d_vector_file)
else: else:
speaker_manager = None speaker_manager = None
# setup model # setup model
model = setup_model(c) model = setup_model(c)

2
TTS/bin/train_tts.py
View File

@ -64,7 +64,7 @@ def main():
train_samples=train_samples, train_samples=train_samples,
eval_samples=eval_samples, eval_samples=eval_samples,
training_assets={"audio_processor": ap}, training_assets={"audio_processor": ap},
parse_command_line_args=True, parse_command_line_args=False,
) )
trainer.fit() trainer.fit()