Speech Recovery for Real-World Self-powered Intermittent Devices

TL;DR

This paper introduces a novel intermittent speech recovery system designed for real-world self-powered devices, significantly improving speech quality, intelligibility, and recognition accuracy despite intermittent power issues.

Contribution

The paper proposes a new ISR system with three stages—interpolation, enhancement, and combination—that effectively reconstructs speech from intermittent signals in real-world devices.

Findings

Speech quality increased by up to 591.7%.

Speech intelligibility improved by up to 80.5%.

Word error rate reduced by up to 52.6%.

Abstract

The incompleteness of speech inputs severely degrades the performance of all the related speech signal processing applications. Although many researches have been proposed to address this issue, they controlled the data missing conditions by simulation with self-defined masking lengths or sizes. Besides, the masking definitions are different among all these experimental settings. This paper presents a novel intermittent speech recovery (ISR) system for real-world self-powered intermittent devices. Three contributive stages: interpolation, enhancement, and combination are applied to the ISR system for speech reconstruction. The experimental results show that our recovery system increases speech quality by up to 591.7%, while increasing speech intelligibility by up to 80.5%. Most importantly, the proposed ISR system improves the WER scores by up to 52.6%. The promising results not only confirm the effectiveness of the reconstruction but also encourage the utilization of these battery-free wearable/IoT devices.