Spaces:

djkesu

/

tortoise5c

like 4

Running

No wait nvm

hold on it softlocks on the generating candidates thing

what does RuntimeError: Error: You set cuda=True but no cuda device found. even mean

Is your issue resolved? It might be because it's running on free cpus right now

Nope, no fixes. I tried making another voice and I was greeted with
File "/home/user/.local/lib/python3.10/site-packages/streamlit/runtime/scriptrunner/script_runner.py", line 541, in _run_script
exec(code, module.dict)
File "/home/user/app/app.py", line 314, in
main()
File "/home/user/app/app.py", line 283, in main
filepaths = run_and_save_tts(
File "/home/user/app/tortoise/inference.py", line 109, in run_and_save_tts
save_gen_with_voicefix(g, fps[-1], squeeze=False, voicefixer=voicefixer)
File "/home/user/app/tortoise/inference.py", line 173, in save_gen_with_voicefix
vfixer.restore(
File "/home/user/.local/lib/python3.10/site-packages/voicefixer/base.py", line 141, in restore
out_np_wav = self.restore_inmem(
File "/home/user/.local/lib/python3.10/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "/home/user/.local/lib/python3.10/site-packages/voicefixer/base.py", line 106, in restore_inmem
check_cuda_availability(cuda=cuda)
File "/home/user/.local/lib/python3.10/site-packages/voicefixer/tools/pytorch_util.py", line 8, in check_cuda_availability
raise RuntimeError("Error: You set cuda=True but no cuda device found.")

Have you tried running this space on ZERO or using tortoise5x?

Quick update: apparently updating the basic settings brings up "NameError: name 'extra_voices_dir' is not defined"