What Decisions Have You Made?: Automatic Decision Detection in Meeting Conversations

This study addresses the problem of automatically detecting decisions in conversational speech. We formulate the problem as classifying decision-making units at two levels of granularity: dialogue acts and topic segments. We conduct an empirical analysis to determine the characteristic features of decision-making dialogue acts, and train MaxEnt models using these features for the classification tasks. We find that models that combine lexical, prosodic, contextual and topical features yield the best results on both tasks, achieving 72% and 86% precision, respectively. The study also provides a quantitative analysis of the relative importance of the feature types.

[1]  Hagen Soltau,et al.  Advances in automatic meeting record creation and access , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

[2]  Andreas Stolcke,et al.  Direct Modeling of Prosody: An Overview of Applications in Automatic Speech Processing , 2004 .

[3]  Samy Bengio,et al.  Detecting group interest-level in meetings , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[4]  Mari Ostendorf,et al.  Detection Of Agreement vs. Disagreement In Meetings: Training With Unlabeled Data , 2003, NAACL.

[5]  Christopher Cieri,et al.  Research methodologies, observations and outcomes in (conversational) speech data collection , 2002 .

[6]  Johanna D. Moore,et al.  AUTOMATIC TOPIC SEGMENTATION AND LABELING IN MULTIPARTY DIALOGUE , 2006, 2006 IEEE Spoken Language Technology Workshop.

[7]  Matthew Purver,et al.  Shallow Discourse Structure for Action Item Detection , 2006, HLT-NAACL 2006.

[8]  Stanley Peters,et al.  Ontology-Based Discourse Understanding for a Persistent Meeting Assistant , 2005, AAAI Spring Symposium: Persistent Assistants: Living and Working with AI.

[9]  Elizabeth Shriberg,et al.  Spotting "hot spots" in meetings: human judgments and prosodic cues , 2003, INTERSPEECH.

[10]  Mari Ostendorf,et al.  A Quantitative Analysis of Lexical Differences Between Genders in Telephone Conversations , 2005, ACL.

[11]  Julia Hirschberg,et al.  Identifying Agreement and Disagreement in Conversational Speech: Use of Bayesian Networks to Model Pragmatic Dependencies , 2004, ACL.

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

[13]  Dirk Heylen,et al.  Argument Diagramming of Meeting Conversations , 2005 .

[14]  Carolyn Penstein Rosé,et al.  The Necessity of a Meeting Recording and Playback System, and the Benefit of Topic-Level Annotations to Meeting Browsing , 2005, INTERACT.

[15]  V. Pallotta.  Collaborative and Argumentative Models of Meeting Discussions , 2005 .

[16]  Steve Whittaker,et al.  Analysing Meeting Records: An Ethnographic Study and Technological Implications , 2005, MLMI.

[17]  Martial Michel,et al.  The NIST Meeting Room Pilot Corpus , 2004, LREC.

[18]  Jean Carletta,et al.  The AMI Meeting Corpus: A Pre-announcement , 2005, MLMI.