Seanghay Yath

@kekhong95 Also check this sample as well, it might help. https://github.com/google/grafika

@kekhong95 you'll need to use MediaCodec + MediaExtractor to decode video into SurfaceTexture. You'll also need to manage audio as well. MediaCodec is the key.

For merging multiple .mp4 files(video + audio) you'll have to arrange timestamp for reach individual video.

@invisal please help to review it again.

Can you open a pull request?

Sure thanks @cristan!

I don't know about that too. I think you should try using https://pypi.org/project/demucs/ instead it's easier to install. Read more: https://seanghay.com/posts/audio-source-sep

@zakariaelh you can chunk audio into multiple smaller audio and merge it back once it's done.

> Yeah, that's what I ended up doing. Thank you! I haven't used the GUI version as i was looking a function to incorporate in my flow, so what you...

@mediarl I have another project for MDX-Net. https://github.com/seanghay/uvr-mdx-infer/blob/30d837ff95115ce25bf7059b8296bbf37febf7f1/separate.py#L78C26-L78C42 Right now it only uses CPU and if you want it to use GPU you can specify `providers=["CUDAExecutionProvider"]` and install `pip install...