mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-09-04 06:34:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

minor fix

Browse Source
This commit is contained in:
yaozengwei 2023-11-21 16:59:56 +08:00
parent 32931b7885
commit 1ed6b4e143
2 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
egs/ljspeech/TTS/vits/infer.py
View File

@ -225,6 +225,7 @@ def main():
tokenizer=tokenizer,
)
logging.info(f"Wav files are saved to {params.save_wav_dir}")
logging.info("Done!")

3
egs/ljspeech/TTS/vits/vits.py
View File

@ -573,6 +573,7 @@ class VITS(nn.Module):
self,
text: torch.Tensor,
text_lengths: torch.Tensor,
sids: Optional[torch.Tensor] = None,
durations: Optional[torch.Tensor] = None,
noise_scale: float = 0.667,
noise_scale_dur: float = 0.8,
@ -585,6 +586,7 @@ class VITS(nn.Module):
Args:
text (Tensor): Input text index tensor (B, T_text).
text_lengths (Tensor): Input text index tensor (B,).
sids (Tensor): Speaker index tensor (B,).
noise_scale (float): Noise scale value for flow.
noise_scale_dur (float): Noise scale value for duration predictor.
alpha (float): Alpha parameter to control the speed of generated speech.
@ -599,6 +601,7 @@ class VITS(nn.Module):
wav, att_w, dur = self.generator.inference(
text=text,
text_lengths=text_lengths,
sids=sids,
noise_scale=noise_scale,
noise_scale_dur=noise_scale_dur,
alpha=alpha,