Reranking for Sentence Boundary Detection in Conversational Speech

TitleReranking for Sentence Boundary Detection in Conversational Speech
Publication TypeConference Papers
Year of Publication2006
AuthorsRoark B, Liu Y, Harper M, Stewart R, Lease M, Snover M, Shafran I, Dorr BJ, Hale J, Krasnyanskaya A, Yung L
Conference Name2006 IEEE International Conference on Acoustics, Speech and Signal Processing, 2006. ICASSP 2006 Proceedings
Date Published2006/05/14/19
ISBN Number1-4244-0469-X
KeywordsAutomatic speech recognition, conversational speech, data mining, Ear, EARS metadata extraction tasks, Feature extraction, hidden Markov models, meta data, Model driven engineering, NIST, NIST RT-04F community evaluation, oracle accuracy, performance evaluation, reranking, sentence-like unit boundary detection, Speech processing, Speech recognition, Telephony

We present a reranking approach to sentence-like unit (SU) boundary detection, one of the EARS metadata extraction tasks. Techniques for generating relatively small n-best lists with high oracle accuracy are presented. For each candidate, features are derived from a range of information sources, including the output of a number of parsers. Our approach yields significant improvements over the best performing system from the NIST RT-04F community evaluation