Technical Program

Paper Detail

Paper:	SLP-P6.1
Session:	Speech Understanding, Translation, Applications and Systems
Time:	Tuesday, May 16, 16:30 - 18:30
Presentation:	Poster
Topic:	Speech and Spoken Language Processing: Spoken Language Applications and Systems
Title:	Reranking for Sentence Boundary Detection in Conversational Speech
Authors:	Brian Roark, Oregon Health & Science University, United States; Yang Liu, University of Texas, Dallas, United States; Mary Harper, Purdue University, United States; Robin Stewart, Williams College, United States; Matthew Lease, Brown University, United States; Matthew Snover, University of Maryland, College Park, United States; Izhak Shafran, Johns Hopkins University, United States; Bonnie Dorr, University of Maryland, College Park, United States; John Hale, Michigan State University, United States; Anna Krasnyanskaya, University of California, Los Angeles, United States; Lisa Yung, Johns Hopkins University, United States
Abstract:	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.