Abstract

In this paper we develop a maximum-entropy based method for annotating spontaneous conversational speech with punctuation. The goal of this task is to make automatic transcriptions more readable by humans, and to render them into a form that is useful for subsequent natural language processing and discourse analysis. Our basic approach is to view the insertion of punctuation as a form of tagging, in which words are tagged with appropriate punctuation, and to apply a maximum entropy tagger that uses both lexical and prosodic features. We present experimental results on Switchboard data with both reference transcriptions and transcriptions produced by a speech recognition system.

Citations

114	A Maximum Entropy Model for Prepositional Phrase Attachment - Ratnaparkhi, Reynar, et al. - 1994
72	Can prosody aid the automatic classification of dialog acts in conversational speech - Shriberg, Bates, et al. - 1998
47	NYU: Description of the MENE Named Entity System as Used - Borthwick, Sterling, et al. - 1998
35	A Maximum Entropy Part of Speech Tagger - Ratnaparkhi - 1996
28	2001. The Use of Prosody in a Combined System for Punctuation Generation and Speech Recognition - Kim, Woodland
26	Punctuation annotation using statistical prosody models - Christensen, Gotoh, et al. - 2001
21	Cyberpunc: A lightweight punctuation annotation system for speech - Beeferman, Berger, et al. - 1998
18	Combining Words and Prosody for Information Extraction from Speech - hakkani-Tur, Tur, et al. - 1999
10	Speech Recognition with Automatic Punctuation - Chen - 1999
1	Jing Huang and Mukund Padmanabhan. 2001. Extracting Caller Information from Voicemail - Zweig