Add CUDA-specific compilation of mel spectrograms #2206
Merged
Conversation
This file contains 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
ggerganov
approved these changes
Jun 4, 2024
iThalay
pushed a commit
to iThalay/whisper.cpp
that referenced
this pull request
Sep 23, 2024
) * whisper : use polymorphic class to calculate mel spectrogram * whisper : add cuda-specific mel spectrogram calculation * whisper : conditionally compile cufftGetErrorString to avoid warnings * build : add new files to makefile * ruby : add new files to conf script * build : fix typo in makefile * whisper : suppress cub warning for deprecated C++ std in whisper-mel-cuda
iThalay
pushed a commit
to iThalay/whisper.cpp
that referenced
this pull request
Sep 23, 2024
) * whisper : use polymorphic class to calculate mel spectrogram * whisper : add cuda-specific mel spectrogram calculation * whisper : conditionally compile cufftGetErrorString to avoid warnings * build : add new files to makefile * ruby : add new files to conf script * build : fix typo in makefile * whisper : suppress cub warning for deprecated C++ std in whisper-mel-cuda
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Outdated suggestions cannot be applied.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.
The resulting code is about 5-6 times faster: from a median of 15 ms spent on mel spectrogram calculation on a high-end CPU, this works for about 2.2 ms. Given that the sizes are relatively small, the exact CUDA-capable GPU hardly matters. I see no significant difference between 4090 and 1660
HOWEVER this comes with the cost of adding new files and thus whisper.cpp is no longer a single h/cpp pair:
WHISPER_APIdeclarations. The definition toWHISPER_APImust be moved to yet another file.