Can you share your model?

[yds1024]

Do you have plans to open source the model?

[IMLHF]

which one

[BBBoobb]

which one

all

[tianyajun123]

Running python -m PHASEN_009._2_train Will report an error TypeError: forward() missing 1 required positional argument: 'L'，what would cause this？

[wp1120103773]

Running python -m PHASEN_009._2_train Will report an error TypeError: forward() missing 1 required positional argument: 'L'，what would cause this？

i have the same problem.Have you solved it yet？

[wp1120103773]

modifing the line 92 ,let L = 48000, in the conv_stft.py can solve this problem.

[tianyajun123]

modifing the line 92 ,let L = 48000, in the conv_stft.py can solve this problem.

ok,thank you very much！

[wp1120103773]

You're welcome.Are you Chinese?If you like,I think that you can add me to qq：1120103773, so we can talk about some questions.

[wp1120103773]

Sorry，modifing the line 92 ,let L = 48000, in the conv_stft.py can‘t solve this problem，But i get the other way，let L=None，and
insert the line ：”L = outputs.size(-1)“ before line 111 “t = self.window.repeat(1,1,inputs.size(-1))**2”.I think this can solve the problem.

[IMLHF]

Running python -m PHASEN_009._2_train Will report an error TypeError: forward() missing 1 required positional argument: 'L'，what would cause this？

fixed : https://github.com/IMLHF/PHASEN-PyTorch/blob/082444f505e662948d9702f4e844ebd37c453e7e/phasen_torch/models/phasen.py#L505

change line 504 in "phasen_torch/models/phasen.py" to "_mixed_wav_length = mixed_wav_batch.size()[-1]; est_clean_wav_batch = self._istft_fn(est_clean_stft_batch, _mixed_wav_length)".

The parameter "L" is similar with "length" in torch.istft. You could find this function at https://pytorch.org/docs/stable/generated/torch.istft.html?highlight=stft#torch.istft.