minor fix

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

View File

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

View File

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