Is batching-for-transform done?

[keunwoochoi]

Hi, I assumed it's done by #337 (which is great!), but at least transforms.MFCC is not. What's the current state?
Also related, the docstrings of transforms need to be updated. The current version reads like it only deals with a single audio signal.

[vincentqb]

Thanks for pointing this out. The intention is to have all supported, and so MFCC should indeed be updated. Let me know if you notice another one that doesn't support it.

Which docstring is not updated?

[keunwoochoi]

Sorry I didn't make it clear - The docstrings of functionals are updated for example

waveform (torch.Tensor): Tensor of audio of dimension (..., channel, time)

, but the corresponding transforms are not, for example, here,

waveform (torch.Tensor): Tensor of audio of dimension (channel, time)

[keunwoochoi]

how could you fix everything in one day? 👍👍