Improving Fundamental Frequency Generation in EMG-to-Speech Conversion Using a Quantization Approach

Abstract

We present a novel approach to generating fundamental frequency (intonation and voicing) trajectories in an EMG-to-Speech conversion Silent Speech Interface, based on quantizing the EMG-to-F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> mappings target values and thus turning a regression problem into a recognition problem. We present this method and evaluate its performance with regard to the accuracy of the voicing information obtained as well as the performance in generating plausible intonation trajectories within voiced sections of the signal. To this end, we also present a new measure for overall F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> trajectory plausibility, the trajectory-label accuracy (TLAcc), and compare it with human evaluations. Our new F <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> generation method achieves a significantly better performance than a baseline approach in terms of voicing accuracy, correlation of voiced sections, trajectory-label accuracy and, most importantly, human evaluations.