Closed dorpxam closed 10 months ago
hey thanks for reporting this
could you possibly try without stereo and see if the issue is there?
@dorpxam it was indeed a bug with the stereo! should be fixed with 0.3.1
thank you for catching this!
@lucidrains You're right, no problem in mono, the spectrogram show the full bandwidth. Very impressive reactivity, what a skill!
@lucidrains For information, just tested in stereo, works well, with a slightly better initial loss value ;)
tensor(2.1933, device='cuda:0', grad_fn=<AddBackward0>)
@dorpxam May i request for you to share your training code with me? For two weeks i have wanted to and tried to train this model but i have no luck.
@lyndonlauder I'm working on the training code right now. I will share the code ASAP ;)
@lyndonlauder I have open a discussion about training.
Hi, before starting training on a large corpus, I've just test the latest release 3.0 and the new MelBand model. Both models are initialized with the same parameters, no change on others parameters.
For in-situation condition, I load a audio stereo mixture of 8 seconds @ 44100 Hz with a simil target (a drums stem). The batch size is 1, so the tensors are of size :
[1, 2, 352800]
After calling a single training forward step, the backward loss seem coherent:
Unfortunatly, when I save back the audio of both models outputs, I got a strange behavior in the MelBand model. Better than words, the spectrograms:
Mixture [mel scale view]
Target [mel scale view]
BandSplit output [mel scale view]
MelBand output [mel scale view]
MelBand output [linear scale view]
The spectrogram show that the MelBand model output cut the frequencies above 11025 Hz, so half of the Nyquist frequency of 22050 Hz for a 44100 Hz audio.
I don't know if it's normal or a bug, but I prefer to share the information here.
Thank's so much for BS-RoFormer !!!