Syntactically-Informed Models for . . .

by Benoit Favre , Dilek Hakkani-Tür , Elizabeth Shriberg

User correction supplied by mph

Datum	Value	Source
TITLE	Syntactically-Informed Models for . . .	user correction
AUTHOR NAME	Benoit Favre	user correction
AUTHOR AFFIL	International Computer Science Institute, Berkeley, USA,	user correction
AUTHOR NAME	Dilek Hakkani-Tür	user correction
AUTHOR AFFIL	International Computer Science Institute, Berkeley, USA,	user correction
AUTHOR NAME	Elizabeth Shriberg	user correction
AUTHOR AFFIL	SRI International, Menlo Park, USA,	user correction
ABSTRACT	Providing punctuation in speech transcripts not only improves readability, but it also helps downstream text processing such as information extraction or machine translation. In this paper, we improve by 7 % the accuracy of comma prediction in English broadcast news by introducing syntactic features inspired by the role of commas as described in linguistics studies. We conduct an analysis of the impact of those features on other subsets of features (prosody, words...) when combined through CRFs. The syntactic cues can help characterizing large syntactic patterns such as appositions and lists which are not necessarily marked by prosody.	SVM HeaderParse 0.2
CITATIONS	11 found	ParsCit 1.0