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 use_speaker_embedding logic

This commit is contained in:
WeberJulian 2021-11-09 12:20:43 +01:00 committed by Eren Gölge
parent 5f40e96010
commit e04577575e
2 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
TTS/tts/models/base_tts.py
View File

@ -102,7 +102,7 @@ class BaseTTS(BaseModel):
config.d_vector_dim if "d_vector_dim" in config and config.d_vector_dim is not None else 512 config.d_vector_dim if "d_vector_dim" in config and config.d_vector_dim is not None else 512
) )
# init speaker embedding layer # init speaker embedding layer
if config.use_speaker_embedding and not config.use_d_vector_file: if config.use_speaker_embedding:
print(" > Init speaker_embedding layer.") print(" > Init speaker_embedding layer.")
self.speaker_embedding = nn.Embedding(self.num_speakers, self.embedded_speaker_dim) self.speaker_embedding = nn.Embedding(self.num_speakers, self.embedded_speaker_dim)
self.speaker_embedding.weight.data.normal_(0, 0.3) self.speaker_embedding.weight.data.normal_(0, 0.3)

4
TTS/tts/models/vits.py
View File

@ -578,7 +578,7 @@ class Vits(BaseTTS):
outputs = {} outputs = {}
sid, g, lid = self._set_cond_input(aux_input) sid, g, lid = self._set_cond_input(aux_input)
# speaker embedding # speaker embedding
if self.args.use_speaker_embedding and sid is not None and not self.use_d_vector: if self.args.use_speaker_embedding and sid is not None:
g = self.emb_g(sid).unsqueeze(-1) # [b, h, 1] g = self.emb_g(sid).unsqueeze(-1) # [b, h, 1]
# language embedding # language embedding
@ -801,7 +801,7 @@ class Vits(BaseTTS):
x_lengths = torch.tensor(x.shape[1:2]).to(x.device) x_lengths = torch.tensor(x.shape[1:2]).to(x.device)
# speaker embedding # speaker embedding
if self.args.use_speaker_embedding and sid is not None and not self.use_d_vector: if self.args.use_speaker_embedding and sid is not None:
g = self.emb_g(sid).unsqueeze(-1) g = self.emb_g(sid).unsqueeze(-1)
# language embedding # language embedding