Wearable Vital Sign Monitoring System for Reinforced Speech Processing

Abstract

Existing speech technologies largely focus on modeling the oral cavity information from the acquired acoustic signals. However, there are unexploited aspects such as changes in speaker’s behavior or psychological state which could effectively be linked to the variations in the physiological signals. Additionally, it is, according to current trends, foreseen that the future of health care and rehabilitation will be shaped by an amalgamation of novel sensing methods, signal processing, and machine learning. Consequently, there is an increasing interest in reinforcing speech processing by means of multi-modal signal acquisition and this work serves to respond this interest.