{"title":"利用声音的声学-语音相似性进行多语言音素识别","authors":"J. Köhler","doi":"10.21437/ICSLP.1996-556","DOIUrl":null,"url":null,"abstract":"The aim of the work is to exploit the acoustic-phonetic similarities between several languages. In recent work cross-language HMM-based phoneme models have been used only for bootstrapping the language-dependent models and the multi-lingual approach has been investigated only on very small speech corpora. The author introduces a statistical distance measure to determine the similarities of sounds. Further, he presents a new technique to model multi-lingual phonemes. The experiments are conducted with the OGI Multi-Language Telephone Speech Corpus for the languages American English, German and Spanish. In the first experiment phoneme recognition rates between 39.0% and 53.9% are achieved using language-dependent models. Using cross-language models yields improvement for some phonemes, but on average a degradation of recognition performance is observed. However, cross-language models speeds up the cross-language transfer and reduce the size of the phoneme inventory of multi-lingual speech recognition systems. Finally, a new method of modelling multi-lingual phonemes, which can be used for a variety of languages, is presented. This technique reduces the number of phoneme-based units in a multi-lingual speech recognition system.","PeriodicalId":90685,"journal":{"name":"Proceedings : ICSLP. International Conference on Spoken Language Processing","volume":"3 1","pages":"2195-2198"},"PeriodicalIF":0.0000,"publicationDate":"1996-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"101","resultStr":"{\"title\":\"Multi-lingual phoneme recognition exploiting acoustic-phonetic similarities of sounds\",\"authors\":\"J. Köhler\",\"doi\":\"10.21437/ICSLP.1996-556\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The aim of the work is to exploit the acoustic-phonetic similarities between several languages. In recent work cross-language HMM-based phoneme models have been used only for bootstrapping the language-dependent models and the multi-lingual approach has been investigated only on very small speech corpora. The author introduces a statistical distance measure to determine the similarities of sounds. Further, he presents a new technique to model multi-lingual phonemes. The experiments are conducted with the OGI Multi-Language Telephone Speech Corpus for the languages American English, German and Spanish. In the first experiment phoneme recognition rates between 39.0% and 53.9% are achieved using language-dependent models. Using cross-language models yields improvement for some phonemes, but on average a degradation of recognition performance is observed. However, cross-language models speeds up the cross-language transfer and reduce the size of the phoneme inventory of multi-lingual speech recognition systems. Finally, a new method of modelling multi-lingual phonemes, which can be used for a variety of languages, is presented. This technique reduces the number of phoneme-based units in a multi-lingual speech recognition system.\",\"PeriodicalId\":90685,\"journal\":{\"name\":\"Proceedings : ICSLP. International Conference on Spoken Language Processing\",\"volume\":\"3 1\",\"pages\":\"2195-2198\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1996-10-03\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"101\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings : ICSLP. International Conference on Spoken Language Processing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.21437/ICSLP.1996-556\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings : ICSLP. International Conference on Spoken Language Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.21437/ICSLP.1996-556","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Multi-lingual phoneme recognition exploiting acoustic-phonetic similarities of sounds
The aim of the work is to exploit the acoustic-phonetic similarities between several languages. In recent work cross-language HMM-based phoneme models have been used only for bootstrapping the language-dependent models and the multi-lingual approach has been investigated only on very small speech corpora. The author introduces a statistical distance measure to determine the similarities of sounds. Further, he presents a new technique to model multi-lingual phonemes. The experiments are conducted with the OGI Multi-Language Telephone Speech Corpus for the languages American English, German and Spanish. In the first experiment phoneme recognition rates between 39.0% and 53.9% are achieved using language-dependent models. Using cross-language models yields improvement for some phonemes, but on average a degradation of recognition performance is observed. However, cross-language models speeds up the cross-language transfer and reduce the size of the phoneme inventory of multi-lingual speech recognition systems. Finally, a new method of modelling multi-lingual phonemes, which can be used for a variety of languages, is presented. This technique reduces the number of phoneme-based units in a multi-lingual speech recognition system.