deviandice/tortoise-tts
1
0
Fork 0
forked from mrq/tortoise-tts
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fixed regression where the auto_conds do not move to the GPU and causes a problem during CVVP compare pass

Browse Source
This commit is contained in:
mrq 2023-02-11 20:34:12 +00:00
parent 3d69274a46
commit 1b55730e67
2 changed files with 6 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
tortoise/api.py
View File

@ -483,8 +483,11 @@ class TextToSpeech:
auto_conditioning, diffusion_conditioning, auto_conds, _ = conditioning_latents
else:
auto_conditioning, diffusion_conditioning = self.get_random_conditioning_latents()
auto_conditioning = auto_conditioning.to(self.device)
diffusion_conditioning = diffusion_conditioning.to(self.device)
if auto_conds is not None:
auto_conds = auto_conds.to(self.device)
diffuser = load_discrete_vocoder_diffuser(desired_diffusion_steps=diffusion_iterations, cond_free=cond_free, cond_free_k=cond_free_k)
@ -539,8 +542,10 @@ class TextToSpeech:
for batch in tqdm_override(samples, verbose=verbose, progress=progress, desc=desc):
for i in range(batch.shape[0]):
batch[i] = fix_autoregressive_output(batch[i], stop_mel_token)
if cvvp_amount != 1:
clvp = self.clvp(text_tokens.repeat(batch.shape[0], 1), batch, return_loss=False)
if auto_conds is not None and cvvp_amount > 0:
cvvp_accumulator = 0
for cl in range(auto_conds.shape[1]):

2
webui.py
View File

@ -265,7 +265,7 @@ def generate(
with open(f'{get_voice_dir()}/{voice}/cond_latents.pth', 'rb') as f:
info['latents'] = base64.b64encode(f.read()).decode("ascii")
if voicefixer:
if args.voice_fixer and voicefixer:
# we could do this on the pieces before they get stiched up anyways to save some compute
# but the stitching would need to read back from disk, defeating the point of caching the waveform
for path in progress.tqdm(audio_cache, desc="Running voicefix..."):