Home

Awesome

Mamba-based Segmentation Model for Speaker Diarization

Alexis Plaquet, Naohiro Tawara, Marc Delcroix, Shota Horiguchi, Atsushi Ando, and Shoko Araki

Mamba is a newly proposed architecture which behaves like a recurrent neural network (RNN) with attention-like capabilities. These properties are promising for speaker diarization, as attention-based models have unsuitable memory requirements for long-form audio, and traditional RNN capabilities are too limited. In this paper, we propose to assess the potential of Mamba for diarization by comparing the state-of-the-art neural segmentation of the pyannote.audio pipeline with our proposed Mamba-based variant. Mamba's stronger processing capabilities allow usage of longer local windows, which significantly improve diarization quality by making the speaker embedding extraction more reliable. We find Mamba to be a superior alternative to both traditional RNN and the tested attention-based model. Our proposed Mamba-based system achieves state-of-the-art performance on three widely used diarization datasets.

📄 Read the paper on arXiv

Citations

@misc{plaquet2024mambabasedsegmentationmodelspeaker,
      title={Mamba-based Segmentation Model for Speaker Diarization}, 
      author={Alexis Plaquet and Naohiro Tawara and Marc Delcroix and Shota Horiguchi and Atsushi Ando and Shoko Araki},
      year={2024},
      eprint={2410.06459},
      archivePrefix={arXiv},
      primaryClass={cs.SD},
      url={https://arxiv.org/abs/2410.06459}, 
}

Repository contents

Installation

You can install the plaqntt package using pip.

  1. Clone this repository and open a terminal in the same folder as this file.
  2. Run pip install -e .

License

Please refer to the LICENSE file for details.