一种改进的开放集文本无关说话人识别方法

2017 Third International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN) Pub Date : 2017-11-01 DOI:10.1109/ICRCICN.2017.8234480

ShrutiSarika Chakraborty, R. Parekh

{"title":"一种改进的开放集文本无关说话人识别方法","authors":"ShrutiSarika Chakraborty, R. Parekh","doi":"10.1109/ICRCICN.2017.8234480","DOIUrl":null,"url":null,"abstract":"This paper focuses on open set text independent speaker identification which is one of the most challenging subclass of Speaker recognition. The initial stage is similar to closed set speaker identification, where the distortion for each test voice against all train voices are determined. The distortions after normalization is set as decision criteria which eases the process of thresholding. The threshold variation which is mostly independent of dataset but dependent on the size of train data set and its values are quite similar for three datasets. The identification rate with balanced False Acceptance Rate(FAR) and False Rejection Rate(FRR) is 73–86%.","PeriodicalId":166298,"journal":{"name":"2017 Third International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":"{\"title\":\"An improved approach to open set text-independent speaker identification (OSTI-SI)\",\"authors\":\"ShrutiSarika Chakraborty, R. Parekh\",\"doi\":\"10.1109/ICRCICN.2017.8234480\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper focuses on open set text independent speaker identification which is one of the most challenging subclass of Speaker recognition. The initial stage is similar to closed set speaker identification, where the distortion for each test voice against all train voices are determined. The distortions after normalization is set as decision criteria which eases the process of thresholding. The threshold variation which is mostly independent of dataset but dependent on the size of train data set and its values are quite similar for three datasets. The identification rate with balanced False Acceptance Rate(FAR) and False Rejection Rate(FRR) is 73–86%.\",\"PeriodicalId\":166298,\"journal\":{\"name\":\"2017 Third International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)\",\"volume\":\"24 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"3\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 Third International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICRCICN.2017.8234480\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 Third International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICRCICN.2017.8234480","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 3

摘要

开放集文本无关说话人识别是说话人识别中最具挑战性的一个子类。初始阶段类似于闭集扬声器识别，其中确定每个测试声音相对于所有列车声音的失真。将归一化后的失真作为判定标准，简化了阈值处理过程。阈值变化主要与数据集无关，但取决于训练数据集的大小，其值在三个数据集上非常相似。虚假接受率(FAR)和虚假拒绝率(FRR)平衡的识别率为73-86%。

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

An improved approach to open set text-independent speaker identification (OSTI-SI)

This paper focuses on open set text independent speaker identification which is one of the most challenging subclass of Speaker recognition. The initial stage is similar to closed set speaker identification, where the distortion for each test voice against all train voices are determined. The distortions after normalization is set as decision criteria which eases the process of thresholding. The threshold variation which is mostly independent of dataset but dependent on the size of train data set and its values are quite similar for three datasets. The identification rate with balanced False Acceptance Rate(FAR) and False Rejection Rate(FRR) is 73–86%.

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

2017 Third International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN)

自引率

0.00%

发文量