Spatial HuBERT: Self-supervised Spatial Speech Representation Learning for a Single Talker from Multi-channel Audio

no code implementations • 17 Oct 2023 • Antoni Dimitriadis, Siqi Pan, Vidhyasaharan Sethu, Beena Ahmed

Spatial HuBERT learns representations that outperform state-of-the-art single-channel speech representations on a variety of spatial downstream tasks, particularly in reverberant and noisy environments.

Representation Learning Self-Supervised Learning