Open Access

ARTICLE

Voice Response Questionnaire System for Speaker Recognition Using Biometric Authentication Interface

Chang-Yi Kao¹, Hao-En Chueh^2,*

1 Soochow University, Taipei City, 100, Taiwan
2 Chung Yuan Christian University, Taoyuan City, 32023, Taiwan

* Corresponding Author: Hao-En Chueh. Email:

Intelligent Automation & Soft Computing 2023, 35(1), 913-924. https://doi.org/10.32604/iasc.2023.024734

Received 29 October 2021; Accepted 02 December 2021; Issue published 06 June 2022

Abstract

The use of voice to perform biometric authentication is an important technological development, because it is a non-invasive identification method and does not require special hardware, so it is less likely to arouse user disgust. This study tries to apply the voice recognition technology to the speech-driven interactive voice response questionnaire system aiming to upgrade the traditional speech system to an intelligent voice response questionnaire network so that the new device may offer enterprises more precise data for customer relationship management (CRM). The intelligence-type voice response gadget is becoming a new mobile channel at the current time, with functions of the questionnaire to be built in for the convenience of collecting information on local preferences that can be used for localized promotion and publicity. Authors of this study propose a framework using voice recognition and intelligent analysis models to identify target customers through voice messages gathered in the voice response questionnaire system; that is, transforming the traditional speech system to an intelligent voice complex. The speaker recognition system discussed here employs volume as the acoustic feature in endpoint detection as the computation load is usually low in this method. To correct two types of errors found in the endpoint detection practice because of ambient noise, this study suggests ways to improve the situation. First, to reach high accuracy, this study follows a dynamic time warping (DTW) based method to gain speaker identification. Second, it is devoted to avoiding any errors in endpoint detection by filtering noise from voice signals before getting recognition and deleting any test utterances that might negatively affect the results of recognition. It is hoped that by so doing the recognition rate is improved. According to the experimental results, the method proposed in this research has a high recognition rate, whether it is on personal-level or industrial-level computers, and can reach the practical application standard. Therefore, the voice management system in this research can be regarded as Virtual customer service staff to use.

---

Keywords

---