Hardware-Accelerated, Short-Term Processing Voice and Nonvoice Sound Recognitions for Electric Equipment Control

Wen‐Chung Tsai ,You-Jyun Shih,Nien-Ting Huang

doi:10.3390/electronics8090924

Wen‐Chung Tsai , You-Jyun Shih + Show 1 more

Open Access

https://doi.org/10.3390/electronics8090924

Copy DOI

Journal: Electronics	Publication Date: Aug 23, 2019
Citations: 3	License type: CC BY 4.0

Affiliation: Chaoyang University of Technology

Abstract

We proposed and implemented a sound recognition system for electric equipment control. In recent years, industry 4.0 has propelled a rapid growth in intelligent human–machine interactions. User acoustic voice commands for machine control have been examined the most by researchers. The targeted machine can be controlled through voice without the use of any hand-held device. However, compared with human voice recognition, limited research has been conducted on nonhuman voice (e.g., mewing sounds) or nonvoice sound recognition (e.g., clapping). Processing of such short-term, biometric nonvoice sounds for electric equipment control requires a rapid response with correct recognition. In practice, this could lead to a trade-off between recognition accuracy and processing performance for conventional software-based implementations. Therefore, we realized a field-programmable gate array-based embedded system, such a hardware-accelerated platform, can enhance information processing performance using a dynamic time warping accelerator. Furthermore, information processing was refined for two specific applications (i.e., mewing sounds and clapping) to enhance system performance including recognition accuracy and execution speed. Performance analyses and demonstrations on real products were conducted to validate the proposed system.

Highlights

Technological advances in industry 4.0 have propelled rapid growth in intelligent human–machine interactions [1]
In addition to common voice recognition, we further proposed and implemented both enhancements of function and performance, for a nonhuman voice and a biometric nonvoice sound, as described : 4.1.1
Provides the field-programmable gate array (FPGA)-based system platform to validate that the proposed hardware and software co-design framework can increase the recognition accuracy by comparing additional samples in a co-design framework can increase the recognition accuracy by comparing additional samples in relatively short time period using the designed Dynamic time warping (DTW) accelerator; a relatively short time period using the designed DTW accelerator; 3

Summary

Introduction

Technological advances in industry 4.0 have propelled rapid growth in intelligent human–machine interactions [1]. Voice recognition-controlled machines have generated considerable interest Such human–machine interactions mainly include system parameter adjustments or remote functional operations [3,4,5,6,7,8]; we implemented a sound recognition and control system. Sensing and control are crucial components of industry 4.0, which has propelled rapid growth in intelligent human–machine interactions [1]. Such human–machine interactions can be achieved by constructing appropriate sensors and recognizing user biometric sounds. Park et al [6] improved voice recognition performance by suppressing acoustic interferences for smart television controls

Methods

Results

Discussion

Conclusion