Multi-level information and automatic dialog act detection in human-human spoken dialogs

This paper reports studies on annotating and automatically detecting dialog acts in human-human spoken dialogs. The work reposes on three hypotheses: first, the succession of dialog acts is strongly constrained; second, the initial word and semantic class of word are more important for identifying dialog acts than the complete exact word sequence of an utterance; third, most of the important information is encoded in specific entities. A memory based learning approach is used to detect dialog acts. For each utterance unit, eight dialog acts are systematically annotated. Experiments have been conducted using different levels of information, with and without the use of dialog history information. In order to assess the generality of the method, the specific entity tag based model trained on a French corpus was tested on an English corpus for a similar task and on a French corpus from a different domain. A correct dialog act detection rate of about 86% is obtained for the same domain/language condition and 77% for the cross-language or cross-domain conditions.

[1]  Jennifer Chu-Carroll,et al.  A Statistical Model for Discourse Act Recognition in Dialogue Interactions , 1998 .

[2]  Amy Isard,et al.  Transaction and Action Coding in the Map Task Corpus , 1995 .

[3]  Yorick Wilks,et al.  Machine Learning Approaches to Human Dialogue Modelling , 2005 .

[4]  Yorick Wilks,et al.  Error Analysis of Dialogue Act Classification , 2005, TSD.

[5]  David R. Traum,et al.  20 Questions on Dialogue Act Taxonomies , 2000, J. Semant..

[6]  Ken Samuel,et al.  Dialogue Act Tagging with Transformation-Based Learning , 1998, ACL.

[7]  Helen Wright,et al.  Automatic utterance type detection using suprasegmental features , 1998, ICSLP.

[8]  Walter Daelemans,et al.  TiMBL: Tilburg Memory-Based Learner, version 2.0, Reference guide , 1998 .

[9]  Julia Hirschberg,et al.  Empirical Studies on the Disambiguation of Cue Phrases , 1993, Comput. Linguistics.

[10]  Detlef Koll,et al.  Probabilistic dialogue act extraction for concept based multilingual translation systems , 1998, ICSLP.

[11]  Yorick Wilks,et al.  Dialogue Act Classification Based on Intra-Utterance Features∗ , 2005 .

[12]  Anne H. Anderson,et al.  The Hcrc Map Task Corpus , 1991 .

[13]  James F. Allen,et al.  Draft of DAMSL Dialog Act Markup in Several Layers , 2007 .

[14]  Johanna D. Moore,et al.  An Empirical Investigation of Proposals in Collaborative Dialogues , 1998, ACL.

[15]  Elizabeth Shriberg,et al.  The ICSI Meeting Recorder Dialog Act (MRDA) Corpus , 2004, SIGDIAL Workshop.

[16]  Kentaro Ogura,et al.  ATR dialogue database , 1990, ICSLP.

[17]  Jean Carletta,et al.  Assessing Agreement on Classification Tasks: The Kappa Statistic , 1996, CL.

[18]  Mark Liberman,et al.  Transcriber: Development and use of a tool for assisting speech corpora production , 2001, Speech Commun..

[19]  Marilyn A. Walker,et al.  Training a Dialogue Act Tagger for Human-human and Human-computer Travel dialogues , 2002, SIGDIAL Workshop.

[20]  Walter Daelemans,et al.  TiMBL: Tilburg Memory-Based Learner , 2007 .

[21]  Andreas Stolcke,et al.  The ICSI Meeting Corpus , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..

[22]  Walter Daelemans,et al.  Forgetting Exceptions is Harmful in Language Learning , 1998, Machine Learning.

[23]  Elizabeth Shriberg,et al.  Automatic dialog act segmentation and classification in multiparty meetings , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[24]  Tomek Strzalkowski,et al.  Semantic and dialogic annotation for automated multilingual customer service , 2003, INTERSPEECH.

[25]  Emiel Krahmer,et al.  Detecting Problematic Turns in Human-Machine Interactions: Rule-induction Versus Memory-based Learning Approaches , 2001, ACL.

[26]  Sophie Rosset,et al.  A semantic representation for spoken dialogs , 2003, INTERSPEECH.

[27]  Jeff A. Bilmes,et al.  Dialog act tagging using graphical models , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[28]  Norbert Reithinger,et al.  Predicting dialogue acts for a speech-to-speech translation system , 1996 .

[29]  Norbert Reithinger,et al.  Dialogue act classification using language models , 1997, EUROSPEECH.

[30]  Lori Lamel,et al.  Automatic detection of dialog acts based on multilevel information , 2004, INTERSPEECH.

[31]  JurafskyDaniel,et al.  Dialogue act modeling for automatic tagging and recognition of conversational speech , 2000 .

[32]  R. Cattoni,et al.  Building a corpus of annotated dialogues : the ADAM experience , .

[33]  Masaaki Nagata,et al.  Using pragmatics to rule out recognition errors in cooperative task-oriented dialogues , 1992, ICSLP.

[34]  P Taylor,et al.  Intonation and dialogue context as constraints for speech recognition , 1998 .

[35]  Andreas Stolcke,et al.  Can Prosody Aid the Automatic Classification of Dialog Acts in Conversational Speech? , 1998, Language and speech.

[36]  Eric Brill,et al.  Transformation-Based Error-Driven Learning and Natural Language Processing: A Case Study in Part-of-Speech Tagging , 1995, CL.

[37]  Gina-Anne Levow,et al.  Dialog act tagging with support vector machines and hidden Markov models , 2006, INTERSPEECH.

[38]  Elmar Nöth,et al.  Dialog act classification with the help of prosody , 1996, Proceeding of Fourth International Conference on Spoken Language Processing. ICSLP '96.

[39]  Ken Samuel,et al.  Automatically Selecting Useful Phrases for Dialogue Act Tagging , 1999, ArXiv.