An Articulatory Speech-Prosthesis System

摘要

We investigate speech-coding strategies for brain-machine-interface (BMI) based speech prostheses. We present an articulatory speech-synthesis system using an experimental integrated-circuit vocal tract that models the human vocal tract. Our articulatory silicon vocal tract makes feasible the transmission of low bit-rate speech-coding parameters over a bandwidth-constrained body sensor network (BSN). To the best of our knowledge, this is the first articulatory speech-prosthesis system reported to date. We also present a speech-prosthesis simulator (SPS) as a means to generate realistic articulatory parameter sequences.