一种基于TEE的语音活动检测方法

2021 3rd International Conference on Artificial Intelligence and Advanced Manufacture Pub Date : 2021-10-23 DOI:10.1145/3495018.3495154

Lingxiang Xia, Z. Cheng, Changhong Gong, Xiang Zou

{"title":"一种基于TEE的语音活动检测方法","authors":"Lingxiang Xia, Z. Cheng, Changhong Gong, Xiang Zou","doi":"10.1145/3495018.3495154","DOIUrl":null,"url":null,"abstract":"To solve the problem of low accuracy of voice activity detection under low SNR, this paper proposes a voice activity detection method based on TEE. Firstly, TEE feature parameter is proposed and used for speech feature extraction; then, the superiority and real-time performance of TEE algorithm are verified theoretically; Finally, TEE is combined with the double threshod method based on FCMC and BIC for the voice activity detection. The experiment proves that the voice activity detection method has high accuracy.","PeriodicalId":6873,"journal":{"name":"2021 3rd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"25 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2021-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A Voice Activity Detection Method Based on TEE\",\"authors\":\"Lingxiang Xia, Z. Cheng, Changhong Gong, Xiang Zou\",\"doi\":\"10.1145/3495018.3495154\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"To solve the problem of low accuracy of voice activity detection under low SNR, this paper proposes a voice activity detection method based on TEE. Firstly, TEE feature parameter is proposed and used for speech feature extraction; then, the superiority and real-time performance of TEE algorithm are verified theoretically; Finally, TEE is combined with the double threshod method based on FCMC and BIC for the voice activity detection. The experiment proves that the voice activity detection method has high accuracy.\",\"PeriodicalId\":6873,\"journal\":{\"name\":\"2021 3rd International Conference on Artificial Intelligence and Advanced Manufacture\",\"volume\":\"25 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-10-23\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 3rd International Conference on Artificial Intelligence and Advanced Manufacture\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3495018.3495154\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 3rd International Conference on Artificial Intelligence and Advanced Manufacture","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3495018.3495154","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

摘要

为了解决低信噪比下语音活动检测准确率低的问题，本文提出了一种基于TEE的语音活动检测方法。首先，提出TEE特征参数并将其用于语音特征提取;然后，从理论上验证了TEE算法的优越性和实时性;最后，将TEE与基于FCMC和BIC的双阈值方法相结合进行语音活动检测。实验证明，该语音活动检测方法具有较高的准确率。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

A Voice Activity Detection Method Based on TEE

To solve the problem of low accuracy of voice activity detection under low SNR, this paper proposes a voice activity detection method based on TEE. Firstly, TEE feature parameter is proposed and used for speech feature extraction; then, the superiority and real-time performance of TEE algorithm are verified theoretically; Finally, TEE is combined with the double threshod method based on FCMC and BIC for the voice activity detection. The experiment proves that the voice activity detection method has high accuracy.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

2021 3rd International Conference on Artificial Intelligence and Advanced Manufacture

自引率

0.00%

发文量