Julien Hauret

`torch.compile` is not working with `torchaudio.functional.lfilter`

2

### 🐛 Describe the bug The new `torch.compile` feature does not work with `torchaudio.functional.lfilter`. My understanding is that `torch.compile` needs to know the shapes of the tensor, but these shapes...

Repeated calls to generator and discriminator's forward in GAN tutorial

1

### 📚 Documentation In the [training_step of the GAN(L.LightningModule)](https://lightning.ai/docs/pytorch/2.1.0/notebooks/lightning_examples/basic-gan.html#:~:text=def%20training_step(self%2C%20batch)%3A) the generator and discriminator forward are called several times on the same input. Obviously this slows down the training because more...

`perceptual_weighting` was passed in the constructor args, but forgotten in the docstring.

Dear Authors, Thank you for publishing your work and making your code available online. It is of great value to the audio community. I was curious about how using more...

`remove_columns` method used with a streaming enable dataset mode produces a LibsndfileError on multichannel audio

2

### Describe the bug When loading a HF dataset in streaming mode and removing some columns, it is impossible to load a sample if the audio contains more than one...

Dear authors, Have you done any experiments where, instead of learning new embedding tables of the code indices produced by DAC/Encodec, you directly used their raw token embeddings as input...