{"title":"基于数据驱动的语音分离方法","authors":"J. Ming, R. Srinivasan, D. Crookes, Ayeh Jafari","doi":"10.1109/TASL.2013.2250959","DOIUrl":null,"url":null,"abstract":"This paper studies single-channel speech separation, assuming unknown, arbitrary temporal dynamics for the speech signals to be separated. A data-driven approach is described, which matches each mixed speech segment against a composite training segment to separate the underlying clean speech segments. To advance the separation accuracy, the new approach seeks and separates the longest mixed speech segments with matching composite training segments. Lengthening the mixed speech segments to match reduces the uncertainty of the constituent training segments, and hence the error of separation. For convenience, we call the new approach Composition of Longest Segments, or CLOSE. The CLOSE method includes a data-driven approach to model long-range temporal dynamics of speech signals, and a statistical approach to identify the longest mixed speech segments with matching composite training segments. Experiments are conducted on the Wall Street Journal database, for separating mixtures of two simultaneous large-vocabulary speech utterances spoken by two different speakers. The results are evaluated using various objective and subjective measures, including the challenge of large-vocabulary continuous speech recognition. It is shown that the new separation approach leads to significant improvement in all these measures.","PeriodicalId":55014,"journal":{"name":"IEEE Transactions on Audio Speech and Language Processing","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2013-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1109/TASL.2013.2250959","citationCount":"25","resultStr":"{\"title\":\"CLOSE—A Data-Driven Approach to Speech Separation\",\"authors\":\"J. Ming, R. Srinivasan, D. Crookes, Ayeh Jafari\",\"doi\":\"10.1109/TASL.2013.2250959\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper studies single-channel speech separation, assuming unknown, arbitrary temporal dynamics for the speech signals to be separated. A data-driven approach is described, which matches each mixed speech segment against a composite training segment to separate the underlying clean speech segments. To advance the separation accuracy, the new approach seeks and separates the longest mixed speech segments with matching composite training segments. Lengthening the mixed speech segments to match reduces the uncertainty of the constituent training segments, and hence the error of separation. For convenience, we call the new approach Composition of Longest Segments, or CLOSE. The CLOSE method includes a data-driven approach to model long-range temporal dynamics of speech signals, and a statistical approach to identify the longest mixed speech segments with matching composite training segments. Experiments are conducted on the Wall Street Journal database, for separating mixtures of two simultaneous large-vocabulary speech utterances spoken by two different speakers. The results are evaluated using various objective and subjective measures, including the challenge of large-vocabulary continuous speech recognition. It is shown that the new separation approach leads to significant improvement in all these measures.\",\"PeriodicalId\":55014,\"journal\":{\"name\":\"IEEE Transactions on Audio Speech and Language Processing\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2013-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://sci-hub-pdf.com/10.1109/TASL.2013.2250959\",\"citationCount\":\"25\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE Transactions on Audio Speech and Language Processing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/TASL.2013.2250959\",\"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 Transactions on Audio Speech and Language Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/TASL.2013.2250959","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 25
摘要
本文研究单通道语音分离,假设待分离语音信号的时间动态未知。描述了一种数据驱动的方法,该方法将每个混合语音段与复合训练语音段进行匹配,以分离底层的干净语音段。为了提高分离精度,该方法通过匹配的复合训练片段寻找最长的混合语音片段并进行分离。通过延长混合语音片段的长度来匹配,减少了组成训练片段的不确定性,从而减少了分离的误差。为方便起见,我们称这种新方法为最长分段组合(Composition of Longest Segments, CLOSE)。CLOSE方法包括一种数据驱动的方法来模拟语音信号的长时间动态,以及一种统计方法来识别具有匹配的复合训练片段的最长混合语音片段。实验在《华尔街日报》数据库上进行,用于分离两个不同说话者同时说的两个大词汇的混合语音。使用各种客观和主观测量来评估结果,包括大词汇量连续语音识别的挑战。结果表明,新的分离方法使这些指标得到了显著改善。
This paper studies single-channel speech separation, assuming unknown, arbitrary temporal dynamics for the speech signals to be separated. A data-driven approach is described, which matches each mixed speech segment against a composite training segment to separate the underlying clean speech segments. To advance the separation accuracy, the new approach seeks and separates the longest mixed speech segments with matching composite training segments. Lengthening the mixed speech segments to match reduces the uncertainty of the constituent training segments, and hence the error of separation. For convenience, we call the new approach Composition of Longest Segments, or CLOSE. The CLOSE method includes a data-driven approach to model long-range temporal dynamics of speech signals, and a statistical approach to identify the longest mixed speech segments with matching composite training segments. Experiments are conducted on the Wall Street Journal database, for separating mixtures of two simultaneous large-vocabulary speech utterances spoken by two different speakers. The results are evaluated using various objective and subjective measures, including the challenge of large-vocabulary continuous speech recognition. It is shown that the new separation approach leads to significant improvement in all these measures.
期刊介绍:
The IEEE Transactions on Audio, Speech and Language Processing covers the sciences, technologies and applications relating to the analysis, coding, enhancement, recognition and synthesis of audio, music, speech and language. In particular, audio processing also covers auditory modeling, acoustic modeling and source separation. Speech processing also covers speech production and perception, adaptation, lexical modeling and speaker recognition. Language processing also covers spoken language understanding, translation, summarization, mining, general language modeling, as well as spoken dialog systems.