Support batch-dimension in log_mel_spectogram #839
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
log_mel_spectrogram
makes the assumption instft[:, :-1].abs() ** 2
that the input to this function is one dimensional (sequence of audio samples in 16kHz).However, all the tensor operating functions can deal with a leading batch dimensions. PyTorch supports the Ellipsis like NumPy to indicate remaining dimensions. This allows us to directly reference the last dimension, instead of assuming that the
stft
frames end up in the second dimension.By using the ellipsis symbol on the
stft
tensor we can makelog_mel_spectogram
support batched inputs of shape [batch, samples].