Language models beyond word strings
Tóm tắt
In this paper we want to show how n-gram language models can be used to provide additional information in automatic speech understanding systems beyond the pure word chain. This becomes important in the context of conversational dialogue systems that have to recognize and interpret spontaneous speech. We show how n-grams can: (1) help to classify prosodic events like boundaries and accents; (2) be extended to directly provide boundary information in the speech recognition phase; (3) help to process speech repairs; and (4) detect and semantically classify out-of-vocabulary words. The approaches can work on the best word chain or a word hypotheses graph. Examples and experimental results are provided from our own research within the EVAR information retrieval system and the VERBMOBIL speech-to-speech translation system.
Từ khóa
#Speech recognition #Speech processing #Speech analysis #Databases #Natural languages #Event detection #Phase detection #Stochastic systems #Automatic speech recognition #Virtual manufacturingTài liệu tham khảo
batliner, 0, The Prosody Module, Wahlster [9], 106
10.3115/992730.992816
10.1016/S0167-6393(98)00037-5
10.1109/89.861370
10.1017/CBO9781139173438
jekat, 1995, Dialogue Acts in Verbmobil, Verbmobil Report 65
mast, 1995, Criteria for the Segmentation of Spoken Input into Individual Utterances, Verbmobil Report
batliner, 1999, Automatic Annotation and Classification of Phrase Accents in Spontaneous Speech, Proc European Conf on Speech Communication and Technology, 1, 519, 10.21437/Eurospeech.1999-134
10.1109/TASSP.1987.1165125
10.1016/B978-0-08-051584-7.50045-0
10.1109/ICSLP.1996.607083
gallwitz, 1998, The Erlangen Spoken Dialogue System EVAR: A State-of-the-art Information Retrieval System, Proc of the 1998 Int Symposium on Spoken Dialogue (ISSD 98), 19
spilker, 2001, Behandlung spontansprachlicher Reparaturen in einem Sprachverarbeitungssystem
10.3115/1075527.1075614
hetherington, 1995, A characterization of the problem of new out-of-vocabulary words in continuous-speech recognition and understanding
10.1016/S0167-6393(02)00079-1
10.1007/3-540-63580-7
fetter, 1998, Detection and Transcription of Out-of-Vocabulary Words in Continuous-Speech Recognition
pallet, 1995, 1994 Benchmark Tests for the ARPA Spoken Language Program, Proc ARPA Spoken Language Systems Technology Workshop
jelinek, 1997, Statistical Methods for Speech Recognition
wahlster, 2000, Verbmobil Foundations of speech-to-speech translations, 10.1007/978-3-662-04230-4
10.1109/PROC.1976.10159
gallwitz, 0, Integrated Stochastic Models for Spontaneous Speech Recognition Studien zur Mustererkennung
jelinek, 1999, Putting Language Into Language Modeling, Proc European Conf on Speech Communication and Technology Budapest, 1, kn-1
10.1006/csla.1993.1010
spilker, 2001, sHow to Repair Speech Repairs in an End-to-End System, Proc ISCA Workshop on Disflueny in Spontaneous Speechs, 73
10.1016/S0167-6393(01)00027-9
brown, 1990, A Statistical Approach to Machine Translation, Computational Linguistics, 16, 79
10.1016/0010-0277(83)90026-4