import module about whisper-vits-svc HOT 8 CLOSED

whisper is built-in, do not install it additionally, it will conflict and report an error

set working directory:

export PYTHONPATH=$PWD

from whisper-vits-svc.

whisper is built-in, do not install it additionally, it will conflict and report an error

set working directory:

export PYTHONPATH=$PWD

thank you , it works but another errors happened :

Traceback (most recent call last):
File "prepare/preprocess_ppg.py", line 62, in
pred_ppg(whisper, f"{wavPath}/{spks}/{file}.wav", f"{ppgPath}/{spks}/{file}.ppg")
File "prepare/preprocess_ppg.py", line 34, in pred_ppg
ppg = whisper.encoder(mel.unsqueeze(0)).squeeze().data.cpu().float().numpy()
File "/home/parisa/anaconda3/envs/voice/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1194, in call_impl
return forward_call(*input, **kwargs)
File "/home/parisa/so-vits-svc-5.0/whisper/model.py", line 159, in forward
x = block(x)
File "/home/parisa/anaconda3/envs/voice/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1194, in call_impl
return forward_call(*input, **kwargs)
File "/home/parisa/so-vits-svc-5.0/whisper/model.py", line 124, in forward
x = x + self.attn(self.attn_ln(x), mask=mask, kv_cache=kv_cache)[0]
File "/home/parisa/anaconda3/envs/voice/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1194, in call_impl
return forward_call(*input, **kwargs)
File "/home/parisa/so-vits-svc-5.0/whisper/model.py", line 30, in forward
return super().forward(x.float()).type(x.dtype)
File "/home/parisa/anaconda3/envs/voice/lib/python3.7/site-packages/torch/nn/modules/normalization.py", line 191, in forward
input, self.normalized_shape, self.weight, self.bias, self.eps)
File "/home/parisa/anaconda3/envs/voice/lib/python3.7/site-packages/torch/nn/functional.py", line 2515, in layer_norm
return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
RuntimeError: expected scalar type Float but found Half

from whisper-vits-svc.

you need update codes, and need a GPU.

from whisper-vits-svc.

you need update codes, and need a GPU.

i got this error after running the fifth step of inference

Traceback (most recent call last):
File "svc_inference.py", line 169, in
main(args)
File "svc_inference.py", line 51, in main
load_svc_model(args.model, model)
File "svc_inference.py", line 17, in load_svc_model
assert os.path.isfile(checkpoint_path)
AssertionError

from whisper-vits-svc.

do you downward the checkpoint of the model, and set the right path?

from whisper-vits-svc.

do you downward the checkpoint of the model, and set the right path?

Whrere are the checkpoints of the model and how to set them?

from whisper-vits-svc.

in readme

from whisper-vits-svc.

I have downloaded the model and saved it in whisper_pretrain as per readme.

But still getting this error:
#46 (comment)

from whisper-vits-svc.