{"title":"可扩展文本到语音系统中的语义计算","authors":"Zhang Wei, Pang Min-hui, Dai Li-rong","doi":"10.1109/WSCS.2008.7","DOIUrl":null,"url":null,"abstract":"Because of diversity of hardware environments, building scalable text-to-speech system is an important issue of Corpus-based text-to-speech system. This paper proposes and analyses three semantic computing problems of building scalable text to speech system: similarity calculation, granular computing and automated instances-pruning process framework. According to these, an acoustic clustering algorithm-NuClustering-VPA and a data ranking algorithm-StaRp-VPA are constructed to pruning synthesis instances. In experiments, the naturalness scored by MOS remains almost unchanged when less than 50% instances are pruned off using these two algorithms and the MOS does not severely degrade when reduction rate is above 50% using StaRp-VPA algorithm.","PeriodicalId":378383,"journal":{"name":"IEEE International Workshop on Semantic Computing and Systems","volume":"84 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Semantic Computing in Scalable Text-to-Speech System\",\"authors\":\"Zhang Wei, Pang Min-hui, Dai Li-rong\",\"doi\":\"10.1109/WSCS.2008.7\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Because of diversity of hardware environments, building scalable text-to-speech system is an important issue of Corpus-based text-to-speech system. This paper proposes and analyses three semantic computing problems of building scalable text to speech system: similarity calculation, granular computing and automated instances-pruning process framework. According to these, an acoustic clustering algorithm-NuClustering-VPA and a data ranking algorithm-StaRp-VPA are constructed to pruning synthesis instances. In experiments, the naturalness scored by MOS remains almost unchanged when less than 50% instances are pruned off using these two algorithms and the MOS does not severely degrade when reduction rate is above 50% using StaRp-VPA algorithm.\",\"PeriodicalId\":378383,\"journal\":{\"name\":\"IEEE International Workshop on Semantic Computing and Systems\",\"volume\":\"84 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2008-07-14\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE International Workshop on Semantic Computing and Systems\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/WSCS.2008.7\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE International Workshop on Semantic Computing and Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/WSCS.2008.7","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
由于硬件环境的多样性,构建可扩展的文本到语音系统是基于语料库的文本到语音系统的一个重要问题。本文提出并分析了构建可扩展文本到语音系统的三个语义计算问题:相似度计算、颗粒计算和自动实例修剪过程框架。在此基础上,构造了声学聚类算法-核聚类- vpa和数据排序算法- starp - vpa对合成实例进行剪枝。在实验中,当使用这两种算法修剪少于50%的实例时,MOS的自然度评分基本保持不变,而当使用StaRp-VPA算法修剪率超过50%时,MOS的自然度评分不会严重下降。
Semantic Computing in Scalable Text-to-Speech System
Because of diversity of hardware environments, building scalable text-to-speech system is an important issue of Corpus-based text-to-speech system. This paper proposes and analyses three semantic computing problems of building scalable text to speech system: similarity calculation, granular computing and automated instances-pruning process framework. According to these, an acoustic clustering algorithm-NuClustering-VPA and a data ranking algorithm-StaRp-VPA are constructed to pruning synthesis instances. In experiments, the naturalness scored by MOS remains almost unchanged when less than 50% instances are pruned off using these two algorithms and the MOS does not severely degrade when reduction rate is above 50% using StaRp-VPA algorithm.