What Siri Can’t Do:? Creating Clarification Questions for Spoken Dialogue Systems
Dr. Julia Hirschberg
Thursday March 14, 2013
11:00 a.m., 1000 SEO Building
Spoken dialogue systems today rely upon simple techniques to determine when they have failed to recognize a user?s utterance and equally simple techniques to try to elicit new input:? “I?’m sorry, I didn?t understand you.? Can you please repeat?”? We are developing new strategies for identifying system errors and producing reprise clarification questions which model human behavior more closely and elicit more informative user responses.? For example, “Which truck entered the compound today”, where the modifier of truck is determined to be misrecognized.? Such clarification questions make use of material which has been correctly recognized to query items which have not been recognized.? We will describe research which makes use of richer syntactic and prosodic information to augment standard speech recognition confidence scores.? This work allows us to target errors more precisely,? and supports the creation of reprise clarification questions.? This work is implemented in a speech-to-speech translation system in which the dialogue manager attempts to identify and clarify recognition errors before translation occurs.
Julia Hirschberg is Professor of Computer Science at Columbia University and chair of the Department. She does research in prosody, spoken dialogue systems, and emotional and deceptive speech (some of this work was described in a New York Times article http://www.nytimes.com/2011/12/04/business/lie-detection-software-parses-the-human-voice.html). She received her PhD in Computer Science from the University of Pennsylvania in 1985.She worked at Bell Laboratories and AT&T Laboratories — Research from 1985-2003 as a Member of Technical Staff and as a Department Head, creating the Human-Computer Interface Research Department at Bell Labs and moving with it to AT&T Labs. She served as editor-in-chief of /Computational Linguistics/ from 1993-2003 and as an editor-in-chief of/Speech Communication/ from 2003-2006. She was on the Executive Board of the Association for Computational Linguistics (ACL) from 1993-2003, has been on the Permanent Council of International Conference on Spoken Language Processing (ICSLP) since 1996, and served on the board of the International Speech Communication Association (ISCA) from 1999-2007 (as President 2005-2007). She is on the IEEE SLTC, the executive board of the North American chapter of the Association for Computational Linguistics, the CRA Board of Directors, and the board of the CRA-W. She has been active in working for diversity at AT&T and at Columbia. She has been a fellow of the American Association for Artificial Intelligence ?since 1994, an ISCA Fellow since 2008, and became an ACL Fellow in the founding group in 2012.