Markovian combination of language and prosodic models for better speech understanding and recognition
Tóm tắt
Summary form only given. Traditionally, "language" models capture only the word sequences of a language. A crucial component of spoken language, however is its prosody, i.e., rhythmic and melodic properties. This paper summarizes recent work on integrated, computationally efficient modeling of word sequences and prosodic properties of speech, for a variety of speech recognition and understanding tasks, such as dialog act tagging, disfluency detection, and segmentation into sentences and topics. In each case it turns out that hidden Markov representations of the underlying structures and associated observations arise naturally, and allow existing speech recognizers to be combined with separately trained prosodic classifiers. The same HMM-based models can be used in two modes: to recover hidden structure (such as sentence boundaries), or to evaluate speech recognition hypotheses, thereby integrating prosody into the recognition process.
Từ khóa
#Natural languages #Speech recognition #Tagging #Australia #Laboratories #Computational modeling #Hidden Markov models #Automatic speech recognition #Gratings #NASATài liệu tham khảo
stolcke, 1999, Modeling the prosody of hidden events for improved word recognition, Proc 6th European Conf on Speech Communication and Technology, 1, 307
sönmez, 1998, Modeling dynamic prosodic variation for speaker verification, Proceedings of the International Conference on Spoken Language Processing, 7, 3189
10.1162/089120100561737
10.1016/S0167-6393(00)00028-5
shriberg, 2001, Can prosody aid the automatic processing of multi-party meetings? Evidence from predicting punctuation, disfluencies, and overlapping speech, Proceedings ISCA Tutorial and Research Workshop on Prosody in Speech Recognition and Understanding, 139
10.1162/089120101300346796
shriberg, 1998, Can prosody aid the automatic classification of dialog acts in conversational speech?, Language and Speech, 41, 439, 10.1177/002383099804100410
stolcke, 1998, Automatic detection of sentence boundaries and disfluencies based on recognized words, Proceedings of the International Conference on Spoken Language Processing, 5, 2247