MT-LLM

MT-LLM

<em> in International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2025 <br> Authors: Lingwei Meng, Shujie Hu, Jiawen Kang, Yuejiao Wang, Wenxuan Wu, Xixin Wu, Xunying Liu, Helen Meng</em>

This repository contains the trained MT-LLM model for instruction-based multi-talker overlapped speech recognition.

Please check https://github.com/cuhealthybrains/MT-LLM

If you find our work is useful in your research, please cite the following paper:

@inproceedings{meng2025mtllm,
    title={Large Language Model Can Transcribe Speech in Multi-Talker Scenarios with Versatile Instructions},
    author={Meng, Lingwei and Hu, Shujie and Kang, Jiawen and Li, Zhaoqing and Wang, Yuejiao and Wu, Wenxuan and Wu, Xixin and Liu, Xunying and Meng, Helen},
    booktitle={ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
    year={2025}
}

---

license: mit