A Novel Method for Recognizing Vietnamese Voice Commands on Smartphones with Support Vector Machine and Convolutional Neural Networks

Quang H Nguyen,Tuan-Dung Cao

doi:10.1155/2020/2312908

Abstract

This paper will present a new method of identifying Vietnamese voice commands using Google speech recognition (GSR) service results. The problem is that the percentage of correct identifications of Vietnamese voice commands in the Google system is not high. We propose a supervised machine-learning approach to address cases in which Google incorrectly identifies voice commands. First, we build a voice command dataset that includes hypotheses of GSR for each corresponding voice command. Next, we propose a correction system using support vector machine (SVM) and convolutional neural network (CNN) models. The results show that the correction system reduces errors in recognizing Vietnamese voice commands from 35.06% to 7.08% using the SVM model and 5.15% using the CNN model.

Highlights

Users’ need for devices that support voice communication is increasing rapidly
We built the dataset based on Google speech recognition (GSR) service results, evaluating the rate of incorrect results from GSR and when applying the machine-learning algorithms
We have proposed a novel method to identify Vietnamese voice commands using GSR service results

Summary

Introduction

Users’ need for devices that support voice communication is increasing rapidly. Voice is the natural approach to humanmachine communication, so voice control support systems must meet the needs of a variety of users. Studies on device control via voice are increasingly diverse [2]; Google speech recognition (GSR) [3] is a commonly used system for identifying voice commands. Aripin and Othman [4] researched a voice control system for indoor devices, such as lamps and fans, to support the elderly and the disabled. They suggested GSR to recognize voice commands that would be transmitted via Bluetooth to control the devices. Iskender et al [5] proposed the same method to operate vehicles, and Guo et al [6] used this method to build a spelling correction model

Methods

Results

Conclusion