Explainable Deep Clustering for Monaural Speech Separation

The proposed X-DC model uses a dataset of mixed speech signals of two, four, or eight speakers.

Data and Resources

Cite this as

Chihiro Watanabe, Hirokazu Kameoka (2024). Dataset: Explainable Deep Clustering for Monaural Speech Separation. https://doi.org/10.57702/907sumtn

DOI retrieved: December 16, 2024

Additional Info

Field Value
Created December 16, 2024
Last update December 16, 2024
Defined In https://doi.org/10.48550/arXiv.2009.08661
Author Chihiro Watanabe
More Authors
Hirokazu Kameoka