Yu Tsao

AV-SUPERB evaluates audio and visual models across various tasks for better performance.

Exploring human ability to identify deepfake videos compared to AI detection.

Examining how codecs retain emotional tones in voice data.

A new method improves automatic speech recognition by preserving sound order in knowledge transfer.

Exploring the GenSEC challenge to improve speech transcription accuracy.

MCMamba model improves speech quality in noisy environments using spatial and spectral information.

New method enhances speech clarity by integrating visual information.

As deepfakes rise, the need for effective detection becomes crucial.

Exploring the challenges and implications of deepfake technology in today’s media landscape.

MSEMG efficiently cleans sEMG signals, improving clarity and potential applications.