Support ONNX export for Demucs#10
Open
dhunstack wants to merge 6 commits intoadefossez:mainfrom
Open
Conversation
Add `onnx_exportable` flag to add a parallel computation path with complex number expressed as a 2D tensor with real and imag components Signed-off-by: Anmol Mishra <anmolmishra1997@gmail.com>
Add `STFT_Process` class that uses PyTorch's convolution to compute STFT with real numbers, avoiding the use of complex numbers. Signed-off-by: Anmol Mishra <anmolmishra1997@gmail.com>
Add `ISTFT_Process` class that computes ISTFT with real numbers, avoiding any complex tensors Signed-off-by: Anmol Mishra <anmolmishra1997@gmail.com>
| KAISER = 'kaiser' | ||
|
|
||
| def __call__(self, window_length): | ||
| match self: |
There was a problem hiding this comment.
The conda setup only supports python versions <3.10
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
In this PR, I've modified the STFT and ISTFT layers to be ONNX exportable.
Details of the changes are documented in this blog - Mixxx Blog Link
The exported model measures closely on the MusDB benchmark.
Table: SI-SDR (dB) comparison for each stem and overall, using torchmetrics. Results are shown for the native PyTorch model and the exported ONNX model running in C++.