Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fixed bug related to yourtts speaker embeddings issue #2234

Merged
merged 12 commits into from
Jan 2, 2023
Merged

Fixed bug related to yourtts speaker embeddings issue #2234

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
6bae35c
Fixed bug related to yourtts speaker embeddings issue
iamkhalidbashir Dec 22, 2022
ec22bb2
Reverted code for base_tts
iamkhalidbashir Dec 22, 2022
242b593
Bug fix on VITS d_vector_file type
Edresson Dec 22, 2022
d154c22
Ignore the test speakers on YourTTS recipe
Edresson Dec 22, 2022
69f3b9a
Add speaker encoder model and config on YourTTS recipe to easily do z…
Edresson Dec 22, 2022
8298c2c
Update YourTTS config file
Edresson Dec 22, 2022
bd01d85
Update ModelManager._update_path to deal with list attributes
Edresson Dec 22, 2022
1c92d81
Fix lint checks
Edresson Dec 22, 2022
6c3b1ad
Remove unused code
Edresson Dec 22, 2022
ed60197
Fix unit tests
Edresson Dec 22, 2022
c8245cd
Reset name_to_id to get the right speaker ids on load_embeddings_from…
Edresson Dec 22, 2022
0a9d28d
Set weighted_sampler_multipliers as an empty dict to prevent users' m…
Edresson Dec 26, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
5 changes: 4 additions & 1 deletion TTS/tts/models/base_tts.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -416,7 +416,10 @@ def on_init_start(self, trainer):
"""Save the speaker.pth and language_ids.json at the beginning of the training. Also update both paths."""
if self.speaker_manager is not None:
output_path = os.path.join(trainer.output_path, "speakers.pth")
self.speaker_manager.save_ids_to_file(output_path)
if trainer.config.use_d_vector_file and trainer.config.d_vector_file:
self.speaker_manager.save_embeddings_to_file(output_path)
else:
self.speaker_manager.save_ids_to_file(output_path)
Edresson marked this conversation as resolved.
Show resolved Hide resolved
trainer.config.speakers_file = output_path
# some models don't have `model_args` set
if hasattr(trainer.config, "model_args"):
Expand Down
2 changes: 1 addition & 1 deletion TTS/tts/utils/speakers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -111,7 +111,7 @@ def init_from_config(config: "Coqpit", samples: Union[List[List], List[Dict]] =
speaker_manager = SpeakerManager()
if get_from_config_or_model_args_with_default(config, "speakers_file", None):
speaker_manager = SpeakerManager(
d_vectors_file_path=get_from_config_or_model_args_with_default(config, "speaker_file", None)
d_vectors_file_path=get_from_config_or_model_args_with_default(config, "speakers_file", None)
Edresson marked this conversation as resolved.
Show resolved Hide resolved
)
if get_from_config_or_model_args_with_default(config, "d_vector_file", None):
speaker_manager = SpeakerManager(
Expand Down