基于多风格训练的双峰生成语音识别

2020 Zooming Innovation in Consumer Technologies Conference (ZINC) Pub Date : 2020-05-01 DOI:10.1109/ZINC50678.2020.9161815

J. Galic, B. Markovic

{"title":"基于多风格训练的双峰生成语音识别","authors":"J. Galic, B. Markovic","doi":"10.1109/ZINC50678.2020.9161815","DOIUrl":null,"url":null,"abstract":"In this paper an analysis on the recognition of words from speech database Whi-Spe in normal and whispered phonation, based on the conventional HMM/GMM framework, is presented. The analysis based on multi-style training is performed in the speaker dependent (SD) and speaker independent mode (SI). The analysis showed that a small portion of whisper data in training (10%) is required for the recognition of whisper higher than 90%, for both the SD and SI recognition.","PeriodicalId":6731,"journal":{"name":"2020 Zooming Innovation in Consumer Technologies Conference (ZINC)","volume":"86 1","pages":"11-14"},"PeriodicalIF":0.0000,"publicationDate":"2020-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"The Recognition of Bimodal Produced Speech based on Multi-style Training\",\"authors\":\"J. Galic, B. Markovic\",\"doi\":\"10.1109/ZINC50678.2020.9161815\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this paper an analysis on the recognition of words from speech database Whi-Spe in normal and whispered phonation, based on the conventional HMM/GMM framework, is presented. The analysis based on multi-style training is performed in the speaker dependent (SD) and speaker independent mode (SI). The analysis showed that a small portion of whisper data in training (10%) is required for the recognition of whisper higher than 90%, for both the SD and SI recognition.\",\"PeriodicalId\":6731,\"journal\":{\"name\":\"2020 Zooming Innovation in Consumer Technologies Conference (ZINC)\",\"volume\":\"86 1\",\"pages\":\"11-14\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-05-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2020 Zooming Innovation in Consumer Technologies Conference (ZINC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ZINC50678.2020.9161815\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 Zooming Innovation in Consumer Technologies Conference (ZINC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ZINC50678.2020.9161815","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

本文在传统HMM/GMM框架的基础上，对语音数据库中正常发音和低声发音单词的识别进行了分析。在说话人依赖模式(SD)和说话人独立模式(SI)下进行基于多风格训练的分析。分析表明，对于SD和SI识别，耳语的识别率高于90%，需要训练中的一小部分耳语数据(10%)。

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

The Recognition of Bimodal Produced Speech based on Multi-style Training

In this paper an analysis on the recognition of words from speech database Whi-Spe in normal and whispered phonation, based on the conventional HMM/GMM framework, is presented. The analysis based on multi-style training is performed in the speaker dependent (SD) and speaker independent mode (SI). The analysis showed that a small portion of whisper data in training (10%) is required for the recognition of whisper higher than 90%, for both the SD and SI recognition.

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

2020 Zooming Innovation in Consumer Technologies Conference (ZINC)

自引率

0.00%

发文量