Software & Data Downloads — TUSS

PyTorch code for task-aware unified source separation (TUSS), a separation model that uses a variable number of learnable prompts to specify which source to separate, and changes its behavior depending on the given prompts, enabling it to handle all the major separation tasks.

• MERL Contacts

  • 

    Jonathan
    Le Roux
     

  • 

    François
    Germain
     

  • 

    Gordon
    Wichern
     

• Related Publications

   Saijo, K., Ebbers, J., Germain, F.G., Wichern, G., Le Roux, J., "Task-Aware Unified Source Separation", IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), DOI: 10.1109/​ICASSP49660.2025.10887819, April 2025.

  BibTeX TR2025-032 PDF Software

  • @inproceedings{Saijo2025mar,
  • author = {Saijo, Kohei and Ebbers, Janek and Germain, François G and Wichern, Gordon and {Le Roux}, Jonathan},
  • title = {{Task-Aware Unified Source Separation}},
  • booktitle = {IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)},
  • year = 2025,
  • month = mar,
  • doi = {10.1109/ICASSP49660.2025.10887819},
  • url = {https://www.merl.com/publications/TR2025-032}
  • }



Software & Data Downloads

Access software at https://github.com/merlresearch/unified-source-separation.