sigsep / open-unmix-pytorch

Open-Unmix - Music Source Separation for PyTorch

https://sigsep.github.io/open-unmix/

sigsep/open-unmix-pytorch Issues

RuntimeError: "reflection_pad1d_out_template" not implemented for 'Short' : when using separate(...) method
Updated 3 months ago7
Training by trackfolder_fix dataset with bad result
Closed 5 months ago7
Validation Error
Closed 5 months ago3
Cannot Figure Out How to Use without CLI, Docs Do Not Help
Closed 5 months ago2
AttributeError: module 'openunmix' has no attribute 'Separator'
Closed 5 months ago2
Cannot download 'umx' or 'umxhq' model
Closed 5 months ago6
Hub loader doesn't use wiener_win_len argument value
Closed 5 months ago1
Zenodo model should have the separator.json file included
Closed 5 months ago3
Train single channel model using left or right channel
Closed 5 months ago2
SDR performance regression since version 1.0.0
Closed 5 months ago6
Use of Transformers/Attention
Closed 5 months ago
--help option seems to be broken in v1.2.1
Closed 5 months ago3
evaluate audacity torch model
Closed 5 months ago
Model download is very slow
Closed 5 months ago2
Gaps in the resulting audio file in the umxse model
Closed 5 months ago20
There was a problem running the openunmix tutorial in colab!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Closed 5 months ago1
How VRAM-hungry is umx supposed to be?
Closed a year ago2
Size difference between umxhq and umxl
Closed a year ago2
TypeError: umx() got an unexpected keyword argument 'target'
Updated 2 years ago
multichannel generalized Wiener filter question
Updated 2 years ago1
Training Hardware and Issue
Updated 2 years ago11
Submit to huggingfaces hub
Closed 2 years ago
Error when training.
Closed 2 years ago1
Running umxhq on a large test track (Georgia Wonder - Siren) blows up memory >64GB
Updated 3 years ago11
Shape broadcasting error in norbert filter _init__.py while referring v1.0.0
Closed 3 years ago9
Missing import of predict when extracting from youtube
Updated 3 years ago1
Can't just evaluate one goal
Closed 3 years ago2
UMX with flac output
Closed 3 years ago8
Add umxl as new default
Closed 3 years ago
Add progress bar for cli inference
Closed 3 years ago
Fix augmentation with multiple workers
Closed 3 years ago1
UMX-L 1.2 installation
Closed 3 years ago4
UMX-L 1.2 installation?
Closed 3 years ago12
Set default augmentations
Closed 3 years ago12
Cannot use double precision with wiener filtering
Closed 3 years ago3
RuntimeError: Backend "sox_io" is not one of available backends: ['soundfile'].
Closed 3 years ago2
Cant load dataset
Closed 3 years ago5
Docker command example not working
Closed 3 years ago3
Typo in README - input tensor shape of OpenUnmix
Closed 3 years ago2
[Question] Ideal/oracle performance of source estimate + mix phase
Updated 3 years ago6
README News links 404
Closed 3 years ago1
Sources with different numbers of channels for sourcefolder dataset
Closed 4 years ago6
Refactor open-unmix as a package
Closed 4 years ago
Does bandwidth extension even exist as a feature?
Closed 4 years ago3
Input-stage standardization
Closed 4 years ago2
A little confused while using istft in test.py
Closed 4 years ago5
Convert training model to pytorch-lightning
Closed 4 years ago1
I get an error What should I do?
Closed 4 years ago5
Availability on Android
Closed 4 years ago7
Not really open, MUSDB18-HQ is not availiable
Closed 4 years ago1