Stefano van Gogh

Diff-ETS: Learning a Diffusion Probabilistic Model for Electromyography-to-Speech Conversion

2024

Electromyography-to-Speech (ETS) conversion has demonstrated its potential for silent speech interfaces by generating audible speech from Electromyography (EMG) signals during silent articulations. ETS models usually consist of an EMG encoder which converts EMG signals to …