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 compute embeddings issue

This commit is contained in:
Edresson Casanova 2022-05-25 10:05:28 -03:00
parent d699416735
commit 5b641ff0e3
2 changed files with 3 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
TTS/bin/compute_embeddings.py
View File

@ -31,11 +31,7 @@ parser.add_argument(
) )
parser.add_argument("output_path", type=str, help="path for output .json file.") parser.add_argument("output_path", type=str, help="path for output .json file.")
parser.add_argument("--old_file", type=str, help="Previous .json file, only compute for new audios.", default=None) parser.add_argument("--old_file", type=str, help="Previous .json file, only compute for new audios.", default=None)
parser.add_argument("--use_cuda", type=bool, help="flag to set cuda.", default=True) parser.add_argument("--use_cuda", type=bool, help="flag to set cuda.", default=False)
parser.add_argument(
"--use_predicted_label", type=bool, help="If True and predicted label is available with will use it.", default=False
)
parser.add_argument("--use_cuda", type=bool, help="flag to set cuda. Default False", default=False)
parser.add_argument("--no_eval", type=bool, help="Do not compute eval?. Default False", default=False) parser.add_argument("--no_eval", type=bool, help="Do not compute eval?. Default False", default=False)
parser.add_argument( parser.add_argument(
"--use_predicted_label", type=bool, help="If True and predicted label is available with will use it.", default=False "--use_predicted_label", type=bool, help="If True and predicted label is available with will use it.", default=False
@ -59,6 +55,7 @@ encoder_manager = EmbeddingManager(
use_cuda=args.use_cuda, use_cuda=args.use_cuda,
) )
print("Using CUDA?", args.use_cuda)
class_name_key = encoder_manager.encoder_config.class_name_key class_name_key = encoder_manager.encoder_config.class_name_key
# compute speaker embeddings # compute speaker embeddings

2
TTS/tts/utils/managers.py
View File

@ -110,7 +110,7 @@ class EmbeddingManager(BaseIDManager):
self.load_embeddings_from_file(embedding_file_path) self.load_embeddings_from_file(embedding_file_path)
if encoder_model_path and encoder_config_path: if encoder_model_path and encoder_config_path:
self.init_encoder(encoder_model_path, encoder_config_path) self.init_encoder(encoder_model_path, encoder_config_path, use_cuda=use_cuda)
@property @property
def embedding_dim(self): def embedding_dim(self):