You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When I convert n texts in a row to audio, the io of my machine's disk increases a lot and the code freezes, it doesn't continue to work, what could I be doing wrong?
#644
When I convert n texts in a row to audio, the io of my machine's disk increases a lot and the code freezes, it doesn't continue to work, what could I be doing wrong?
its my processing fucntion
`
ım ınıt tts and load vıice here
CUSTOM_VOICE_NAME = "sallybath2"
This discussion was converted from issue #593 on October 18, 2023 18:06.
Heading
Bold
Italic
Quote
Code
Link
Numbered list
Unordered list
Task list
Attach files
Mention
Reference
Menu
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
When I convert n texts in a row to audio, the io of my machine's disk increases a lot and the code freezes, it doesn't continue to work, what could I be doing wrong?
its my processing fucntion
`
ım ınıt tts and load vıice here
CUSTOM_VOICE_NAME = "sallybath2"
voice_samples, conditioning_latents = load_voice(CUSTOM_VOICE_NAME)
tts = TextToSpeech(use_deepspeed=True, kv_cache=True, half=True)
def generate_audio(text,work_dir,pp_dir,sira):
CUSTOM_VOICE_NAME = "sallybath2"
preset="ultra_fast"
gen = tts.tts_with_preset(text, voice_samples=voice_samples, conditioning_latents=conditioning_latents,
preset=preset)
torchaudio.save(f'{pp_dir}{work_dir}/{CUSTOM_VOICE_NAME}_{sira}.wav', gen.squeeze(0).cpu(), 24000)
torch.cuda.empty_cache()`
ım calling this fucntion each time with different text
for sira in range(len(parts)): res = generate_audio(parts[sira],work_dir,pp_dir,sira )
and after some times after disk io is flying
my code stop like this
out log
ransforming autoregressive outputs into audio.. 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 30/30 [00:00<00:00, 46.72it/s] /home/mehmet/tortoise-tts-original/process_dir/205VP050S1F5D2AMH128KUI7DS12Y9ZMPGMA3A7YMN53B0D650ACW0KK16NHYQN83L5DXQUZAWWYB8Y6/63.wav Generating autoregressive samples.. 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:05<00:00, 2.59s/it] Computing best candidates using CLVP 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:00<00:00, 11.87it/s] Transforming autoregressive outputs into audio.. 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 30/30 [00:03<00:00, 8.59it/s] /home/mehmet/tortoise-tts-original/process_dir/205VP050S1F5D2AMH128KUI7DS12Y9ZMPGMA3A7YMN53B0D650ACW0KK16NHYQN83L5DXQUZAWWYB8Y6/64.wav Generating autoregressive samples.. 0%| | 0/2 [00:00<?, ?it/s]Killed
what is problem i cant find ım working on windows 11 wsl2(ubuntu )
my system info (install 1.18 becaouse of ım facıng some problems usıng 1.17)
nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2022 NVIDIA Corporation
Built on Wed_Sep_21_10:33:58_PDT_2022
Cuda compilation tools, release 11.8, V11.8.89
Build cuda_11.8.r11.8/compiler.31833905_0
Beta Was this translation helpful? Give feedback.
All reactions