2004 International Symposium on Chinese Spoken Language Processing最新文献_第9页

Double Gaussian based feature normalization for robust speech recognition 基于双高斯特征归一化的鲁棒语音识别

2004 International Symposium on Chinese Spoken Language Processing Pub Date : 2004-12-15 DOI: 10.1109/CHINSL.2004.1409634

Bo Liu, Lirong Dai, Jinyu Li, Ren-Hua Wang

引用次数: 7

Bilingual response generation using semi-automatically-induced templates for a mixed-initiative dialog system 混合主动对话系统中使用半自动诱导模板的双语响应生成

2004 International Symposium on Chinese Spoken Language Processing Pub Date : 2004-12-15 DOI: 10.1109/CHINSL.2004.1409619

Wing Lin Yip, H. Meng

{"title":"Bilingual response generation using semi-automatically-induced templates for a mixed-initiative dialog system","authors":"Wing Lin Yip, H. Meng","doi":"10.1109/CHINSL.2004.1409619","DOIUrl":"https://doi.org/10.1109/CHINSL.2004.1409619","url":null,"abstract":"We have previously developed a framework for natural language response generation for mixed-initiative dialogs in the CUHK Restaurants domain. This paper investigates the use of semi-automatic technique for response templates generation. We adopt a semi-automatic approach for grammar induction to capture the language structures of responses from unannotated corpora. We wish to use this approach to induce a set of grammars from our response data. The induced grammars are coupled with a parser to produce response templates in a semi-automatic way. Our response data consists of 2349 waiter responses. It is used as the training corpus for grammar induction. Unsupervised grammar induction is first performed, followed by using the learned grammars as prior knowledge for seeding the clustering process. Results show that the semi-automatically-induced response templates cover more than 50% of the hand-designed templates in template coverage and provide more realization options. Performance evaluation indicates that the task completion rate is at least 90%, and most of the Grice's maxims as well as the overall user satisfaction scored at 3.5 points or above.","PeriodicalId":212562,"journal":{"name":"2004 International Symposium on Chinese Spoken Language Processing","volume":"85 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126086014","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Detection of language boundary in code-switching utterances by bi-phone probabilities 用双电话概率检测语码转换话语的语言边界

2004 International Symposium on Chinese Spoken Language Processing Pub Date : 2004-12-15 DOI: 10.1109/CHINSL.2004.1409644

Joyce Y. C. Chan, P. Ching, Tan Lee, H. Meng

引用次数: 50

應用語料庫和語意相依法則於中文語音文件之摘要 (Spoken Document Summarization Using Topic-Related Corpus and Semantic Dependency Grammar) [In Chinese] 应用语料库和语意相依法则于中文语音文件之摘要 (Spoken Document Summarization Using Topic-Related Corpus and Semantic Dependency Grammar) [In Chinese]

2004 International Symposium on Chinese Spoken Language Processing Pub Date : 2004-12-15 DOI: 10.1109/CHINSL.2004.1409654

Chia-Hsin Hsieh, Chien-Lin Huang, Chung-Hsien Wu

引用次数: 10

Statistical language model adaptation for Mandarin broadcast news transcription 统计语言模型在普通话广播新闻抄写中的适配

2004 International Symposium on Chinese Spoken Language Processing Pub Date : 2004-12-01 DOI: 10.1109/CHINSL.2004.1409649

Berlin Chen, Wen-Hung Tsai, Jen-wei Kuo

引用次数: 7

A study on Mandarin broadcast news speech recognition 普通话广播新闻语音识别研究

2004 International Symposium on Chinese Spoken Language Processing Pub Date : 2004-12-01 DOI: 10.1109/CHINSL.2004.1409635

C. L. Chen, Yih-Ru Wang, Sin-Horng Chen

引用次数: 0

A maximum entropy approach for integrating semantic information in statistical language models 统计语言模型中语义信息集成的最大熵方法

2004 International Symposium on Chinese Spoken Language Processing Pub Date : 2004-12-01 DOI: 10.1109/CHINSL.2004.1409648

C. Chueh, Jen-Tzung Chien, H. Wang

引用次数: 6

Investigation and modeling of coarticulation in speech production 语音生成中协同发音的研究与建模

2004 International Symposium on Chinese Spoken Language Processing Pub Date : 2004-12-01 DOI: 10.1109/CHINSL.2004.1409577

J. Dang, Jianguo Wei, T. Suzuki, K. Honda, P. Perrier, M. Honda

引用次数: 3

Contributions of periodicity fluctuation cues in individual frequency channels to Chinese speech recognition 各频率通道周期性波动线索对汉语语音识别的贡献

2004 International Symposium on Chinese Spoken Language Processing Pub Date : 2004-12-01 DOI: 10.1109/CHINSL.2004.1409604

Xin Luo, Q. Fu

{"title":"Contributions of periodicity fluctuation cues in individual frequency channels to Chinese speech recognition","authors":"Xin Luo, Q. Fu","doi":"10.1109/CHINSL.2004.1409604","DOIUrl":"https://doi.org/10.1109/CHINSL.2004.1409604","url":null,"abstract":"Studies have revealed near perfect speech recognition with primarily temporal envelope cues and severely degraded spectral cues. Among different types of temporal envelope cues, periodicity fluctuation cues have been found to significantly improve Chinese tone and sentence recognition, while the contributions of periodicity fluctuation cues in individual frequency channels to Chinese speech recognition have not been clearly stated. In order to make periodicity fluctuation cues available in different frequency regions, the present study employed different low-pass cutoff frequencies for the temporal envelope detectors in different channels of a four-channel noise-band cochlear implant simulation. Chinese tone and vowel recognition scores were measured for six native Chinese normal hearing listeners under six low-pass cutoff frequency combinations: all 50 Hz in four channels (all-50 Hz), all 500 Hz in four channels (all-500 Hz), and 500 Hz in one of the four channels while 50 Hz in the other three channels (ch1-500 Hz, ch2-500 Hz, et al.). The results showed that the ch4-500 Hz condition produced the highest Chinese tone recognition among the four single-channel-500 Hz conditions, and was the only condition whose tone recognition was similar to that of the all-500 Hz condition and was significantly higher than that of the all-50 Hz condition. Chinese vowel recognition was not significantly affected by different cutoff frequency combinations. These results suggest that delivering periodicity fluctuation cues in higher frequency channels might be more important and efficient in enhancing Chinese tone recognition for cochlear implant patients.","PeriodicalId":212562,"journal":{"name":"2004 International Symposium on Chinese Spoken Language Processing","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131366669","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1