HarkonCollider/tortoise-tts
1
0
Fork 0
forked from mrq/tortoise-tts
Code Pull Requests Activity

fixed combining audio, somehow this broke, oops

Browse Source
This commit is contained in:
mrq 2023-02-07 00:26:22 +00:00
parent a3c077ba13
commit 3b56c437aa
1 changed files with 8 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
app.py
View File

@ -125,17 +125,17 @@ def generate(text, delimiter, emotion, prompt, voice, mic_audio, seed, candidate
audio = audio_cache[f'result_{line}.wav']['audio'] audio = audio_cache[f'result_{line}.wav']['audio']
audio_clips.append(audio) audio_clips.append(audio)
audio_clips = torch.cat(audio_clips, dim=-1).squeeze(0).cpu() audio = torch.cat(audio_clips, dim=-1)
torchaudio.save(f'{outdir}/combined_{candidate}.wav', audio_clips, 24000) torchaudio.save(f'{outdir}/combined_{candidate}.wav', audio, 24000)
audio = audio.squeeze(0).cpu()
audio_cache[f'combined_{candidate}.wav'] = { audio_cache[f'combined_{candidate}.wav'] = {
'audio': audio, 'audio': audio,
'text': cut_text, 'text': cut_text,
} }
if output_voice is None: if output_voice is None:
output_voice = audio_clips output_voice = audio
else: else:
if isinstance(gen, list): if isinstance(gen, list):
output_voice = gen[0] output_voice = gen[0]
@ -332,6 +332,9 @@ def export_exec_settings( share, check_for_updates, low_vram, cond_latent_max_ch
def main(): def main():
if not torch.cuda.is_available():
print("CUDA is NOT available for use.")
with gr.Blocks() as webui: with gr.Blocks() as webui:
with gr.Tab("Generate"): with gr.Tab("Generate"):
with gr.Row(): with gr.Row():
@ -384,7 +387,7 @@ def main():
diffusion_iterations = gr.Slider(value=128, minimum=0, maximum=512, step=1, label="Iterations") diffusion_iterations = gr.Slider(value=128, minimum=0, maximum=512, step=1, label="Iterations")
temperature = gr.Slider(value=0.2, minimum=0, maximum=1, step=0.1, label="Temperature") temperature = gr.Slider(value=0.2, minimum=0, maximum=1, step=0.1, label="Temperature")
breathing_room = gr.Slider(value=12, minimum=1, maximum=32, step=1, label="Pause Size") breathing_room = gr.Slider(value=8, minimum=1, maximum=32, step=1, label="Pause Size")
diffusion_sampler = gr.Radio( diffusion_sampler = gr.Radio(
["P", "DDIM"], # + ["K_Euler_A", "DPM++2M"], ["P", "DDIM"], # + ["K_Euler_A", "DPM++2M"],
value="P", value="P",