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

Init speaker manager for speaker encoder

This commit is contained in:
Eren Gölge 2021-12-22 15:51:53 +00:00
parent 633dcc9c56
commit 5c5ddd2ba7
2 changed files with 10 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
TTS/tts/utils/speakers.py
View File

@ -256,11 +256,11 @@ class SpeakerManager:
self.speaker_encoder.load_checkpoint(config_path, model_path, eval=True, use_cuda=self.use_cuda) self.speaker_encoder.load_checkpoint(config_path, model_path, eval=True, use_cuda=self.use_cuda)
self.speaker_encoder_ap = AudioProcessor(**self.speaker_encoder_config.audio) self.speaker_encoder_ap = AudioProcessor(**self.speaker_encoder_config.audio)
def compute_d_vector_from_clip(self, wav_file: Union[str, list]) -> list: def compute_d_vector_from_clip(self, wav_file: Union[str, List[str]]) -> list:
"""Compute a d_vector from a given audio file. """Compute a d_vector from a given audio file.
Args: Args:
wav_file (Union[str, list]): Target file path. wav_file (Union[str, List[str]]): Target file path.
Returns: Returns:
list: Computed d_vector. list: Computed d_vector.

7
TTS/utils/synthesizer.py
View File

@ -171,6 +171,11 @@ class Synthesizer(object):
def _init_speaker_encoder(self, speaker_manager): def _init_speaker_encoder(self, speaker_manager):
"""Initialize the SpeakerEncoder""" """Initialize the SpeakerEncoder"""
if self.encoder_checkpoint: if self.encoder_checkpoint:
if speaker_manager is None:
speaker_manager = SpeakerManager(
encoder_model_path=self.encoder_checkpoint, encoder_config_path=self.encoder_config
)
else:
speaker_manager.init_speaker_encoder(self.encoder_checkpoint, self.encoder_config) speaker_manager.init_speaker_encoder(self.encoder_checkpoint, self.encoder_config)
return speaker_manager return speaker_manager
@ -183,6 +188,8 @@ class Synthesizer(object):
language_manager = LanguageManager(language_ids_file_path=self.tts_languages_file) language_manager = LanguageManager(language_ids_file_path=self.tts_languages_file)
elif self.tts_config.get("language_ids_file", None): elif self.tts_config.get("language_ids_file", None):
language_manager = LanguageManager(language_ids_file_path=self.tts_config.language_ids_file) language_manager = LanguageManager(language_ids_file_path=self.tts_config.language_ids_file)
else:
language_manager = LanguageManager(config=self.tts_config)
return language_manager return language_manager
def _load_vocoder(self, model_file: str, model_config: str, use_cuda: bool) -> None: def _load_vocoder(self, model_file: str, model_config: str, use_cuda: bool) -> None: