search

GANtastic3 / MaskCycleGAN-VC

Implementation of Kaneko et al.'s MaskCycleGAN-VC model for non-parallel voice conversion.
MIT License
112 stars 31 forks source link

Wrong spectrogram scaling in inference script #7

Closed terbed closed 3 years ago

terbed commented 3 years ago

In the inference script the scaling back with speaker statistics is wrong:

            wav_fake_B = decode_melspectrogram(self.vocoder, fake_B[0].detach(
            ).cpu(), self.dataset_A_mean, self.dataset_A_std).cpu()

This should be scaled with B speaker statistics.

This correction largely improves the performance of the inference.

Solves issue #3 .

terbed commented 3 years ago

I noticed that the file bash_scripts/mask_cyclegan_test.sh was deleted. Was there a reason for this? Otherwise, I will add it back.

No, it was accidental, thanks!

Thank you for the constructive comments, I will include them!