Logmel: use jnp.fft.rfft instead of jnp.fft.fft. #1221
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Audio samples are always real-valued, which is why the more efficient `rfft` can be used. Currently, the logmel frontend uses `fft` and manually slices the output tensor to `fft_size // 2 + 1`. In contrast, `rfft` directly returns a tensor of shape `fft_size // 2 + 1` because, when the input is real, half of the FFT output consists of conjugate complex values, essentially redundant information.
|
@markblee Could you review it? From 1372 |
markblee
approved these changes
May 29, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Audio samples are always real-valued, which is why the more efficient
rfftcan be used.Currently, the logmel frontend uses
fftand manually slices the output tensor tofft_size // 2 + 1. In contrast,rfftdirectly returns a tensor of shapefft_size // 2 + 1because, when the input is real, half of the FFT output consists of conjugate complex values, essentially redundant information.