Audio xLSTMs: Learning Self-supervised audio representations with xLSTMs

This is a community based approach to an implementation mostly for practice. I will implement the model architecture as defined in the paper but will leave someone else to implement the training script! So please create a training script if you have the time and energy

Install

$ pip3 install -U audio-xlstm

License

MIT

Todo

• Implement the flip module
• Correctly leverage msltm module
• Ensure model architecture is correct
• Implement training script on whisper like data
• Implement speech and audio recognition datasets

Citation

@article{xlstm,
  title={xLSTM: Extended Long Short-Term Memory},
  author={Beck, Maximilian and P{\"o}ppel, Korbinian and Spanring, Markus and Auer, Andreas and Prudnikova, Oleksandra and Kopp, Michael and Klambauer, G{\"u}nter and Brandstetter, Johannes and Hochreiter, Sepp},
  journal={arXiv preprint arXiv:2405.04517},
  year={2024}
}