tailieunhanh - Báo cáo khoa học: "PREDICTING INTONATIONAL PHRASING FROM TEXT"

Determining the relationship between the intonational characteristics of an utterance and other features inferable from its text is important both for speech recognition and for speech synthesis. This work investigates the use of text analysis in predicting the location of intonational phrase boundaries in natural speech, through analyzing 298 utterances from the DARPA Air Travel Information Service database. For statistical modeling, we employ Classification and Regression Tree (CART) techniques. . | PREDICTING INTONATIONAL PHRASING FROM TEXT Michelle Q. Wang Churchill College Cambridge University Cambridge UK Julia Hirschberg AT T Bell Laboratories 600 Mountain Avenue Murray Hill NJ 07974 Abstract Determining the relationship between the intona-tional characteristics of an utterance and other features inferable from its text is important both for speech recognition and for speech synthesis. This work investigates the use of text analysis in predicting the location of intonational phrase boundaries in natural speech through analyzing 298 utterances from the DARPA Air Travel Information Service database. For statistical modeling we employ Classification and Regression Tree CART techniques. We achieve success rates of just over 90 representing a major improvement over other attempts at boundary prediction from unrestricted Introduction The relationship between the intonational phrasing of an utterance and other features which can be inferred from its transcription represents an important source of information for speech synthesis and speech recognition. In synthesis more natural intonational phrasing can be assigned if text analysis can predict human phrasing performance. In recognition better calculation of probable word durations is possible if the phrase-finallengthening that precedes boundary sites can be predicted. Furthermore the association of intona-tional features with syntactic and acoustic information can also be used to reduce the number of sentence hypotheses under consideration. Previous research on the location of intonational boundaries has largely focussed on the relationship between these prosodic boundaries and syntactic constituent boundaries. While current research acknowledges the role that semantic and discourse-level information play in boundary as 1 We thank Michael Riley for helpful discussions. Code implementing the CART techniques employed here was written by Michael Riley and Daryl Pregibon. Part-of-speech tagging employed Ken