Search papers, labs, and topics across Lattice.
2
0
1
2
Whispered speech recognition just got a major upgrade, slashing hallucination rates from over 25% to just 4.5% with a new self-supervised uncertainty learning approach.
Achieving state-of-the-art acoustic echo cancellation and noise suppression with only 480K parameters, LMPAN redefines efficiency in spoken dialogue systems.
