Predicting the Attitude Flow in Dialogue Based on Multi-Modal Speech Cues

Peter Juel Henrichsen; Jens Allwood

Predicting the Attitude Flow in Dialogue Based on Multi-Modal Speech Cues

Peter Juel Henrichsen, Jens Allwood

University of Gothenburg

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Abstract

We present our experiments on attitude detection based on annotated multi-modal dialogue data1. Our long-term goal is to establish a computational model able to predict the attitudinal patterns in humanhuman dialogue. We believe, such prediction algorithms are useful tools in the pursuit of realistic discourse behavior in conversational agents and other intelligent man-machine interfaces. The present paper deals with two important subgoals in particular: How to establish a meaningful and consistent set of annotation categories for attitude annotation, and how to relate the annotation data to the
recorded data (audio and video) in computational models of attitude prediction. We present our current results including a recommended set of analytical annotation labels and a recommended setup for extracting linguistically meaningful data even from noisy audio and video signals.

Original language	English
Title of host publication	NEALT2012 : Proceedings of the 4th Nordic Symposium on Multimodal Communication, Nov. 15-16, Gothenburg, Sweden
Editors	Jens Allwood, Elisabeth Ahlsén, Patrizia Paggio, Kristiina Jokinen
Place of Publication	Göteborg
Publisher	Göteborg Universitet
Publication date	2013
Pages	47-53
Publication status	Published - 2013
Event	The 4th Nordic Symposium on Multimodal Communication - University of Gothenburg, Gothenburg, Sweden Duration: 15 Nov 2012 → 16 Nov 2012 Conference number: 4

Conference

Conference	The 4th Nordic Symposium on Multimodal Communication
Number	4
Location	University of Gothenburg
Country/Territory	Sweden
City	Gothenburg
Period	15/11/2012 → 16/11/2012

Series	Linköping Electronic Conference Proceedings
Number	93
ISSN	1650-3686

Access to Document

http://www.ep.liu.se/ecp/093/ecp12093.pdf

Handle.net

Persistent link

Cite this

Juel Henrichsen, Peter ; Allwood, Jens. / Predicting the Attitude Flow in Dialogue Based on Multi-Modal Speech Cues. NEALT2012: Proceedings of the 4th Nordic Symposium on Multimodal Communication, Nov. 15-16, Gothenburg, Sweden. editor / Jens Allwood ; Elisabeth Ahlsén ; Patrizia Paggio ; Kristiina Jokinen. Göteborg : Göteborg Universitet, 2013. pp. 47-53 (Linköping Electronic Conference Proceedings; No. 93).

@inproceedings{ba77a40d28e54ccf8ebecde93c806cb5,

title = "Predicting the Attitude Flow in Dialogue Based on Multi-Modal Speech Cues",

abstract = "We present our experiments on attitude detection based on annotated multi-modal dialogue data1. Our long-term goal is to establish a computational model able to predict the attitudinal patterns in humanhuman dialogue. We believe, such prediction algorithms are useful tools in the pursuit of realistic discourse behavior in conversational agents and other intelligent man-machine interfaces. The present paper deals with two important subgoals in particular: How to establish a meaningful and consistent set of annotation categories for attitude annotation, and how to relate the annotation data to therecorded data (audio and video) in computational models of attitude prediction. We present our current results including a recommended set of analytical annotation labels and a recommended setup for extracting linguistically meaningful data even from noisy audio and video signals.",

keywords = "Attitude detection, Prediction of attitude flow, Attitude annotation, Multimodal speech cues",

author = "{Juel Henrichsen}, Peter and Jens Allwood",

year = "2013",

language = "English",

series = "Link{\"o}ping Electronic Conference Proceedings",

number = "93",

pages = "47--53",

editor = "Jens Allwood and Elisabeth Ahls{\'e}n and Patrizia Paggio and Kristiina Jokinen",

booktitle = "NEALT2012",

publisher = "G{\"o}teborg Universitet",

address = "Sweden",

note = "The 4th Nordic Symposium on Multimodal Communication ; Conference date: 15-11-2012 Through 16-11-2012",

}

Juel Henrichsen, P & Allwood, J 2013, Predicting the Attitude Flow in Dialogue Based on Multi-Modal Speech Cues. in J Allwood, E Ahlsén, P Paggio & K Jokinen (eds), NEALT2012: Proceedings of the 4th Nordic Symposium on Multimodal Communication, Nov. 15-16, Gothenburg, Sweden. Göteborg Universitet, Göteborg, Linköping Electronic Conference Proceedings, no. 93, pp. 47-53, The 4th Nordic Symposium on Multimodal Communication, Gothenburg, Sweden, 15/11/2012. <http://www.ep.liu.se/ecp/093/ecp12093.pdf>

Predicting the Attitude Flow in Dialogue Based on Multi-Modal Speech Cues. / Juel Henrichsen, Peter; Allwood, Jens.
NEALT2012: Proceedings of the 4th Nordic Symposium on Multimodal Communication, Nov. 15-16, Gothenburg, Sweden. ed. / Jens Allwood; Elisabeth Ahlsén; Patrizia Paggio; Kristiina Jokinen. Göteborg: Göteborg Universitet, 2013. p. 47-53 (Linköping Electronic Conference Proceedings; No. 93).

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - Predicting the Attitude Flow in Dialogue Based on Multi-Modal Speech Cues

AU - Juel Henrichsen, Peter

AU - Allwood, Jens

N1 - Conference code: 4

PY - 2013

Y1 - 2013

N2 - We present our experiments on attitude detection based on annotated multi-modal dialogue data1. Our long-term goal is to establish a computational model able to predict the attitudinal patterns in humanhuman dialogue. We believe, such prediction algorithms are useful tools in the pursuit of realistic discourse behavior in conversational agents and other intelligent man-machine interfaces. The present paper deals with two important subgoals in particular: How to establish a meaningful and consistent set of annotation categories for attitude annotation, and how to relate the annotation data to therecorded data (audio and video) in computational models of attitude prediction. We present our current results including a recommended set of analytical annotation labels and a recommended setup for extracting linguistically meaningful data even from noisy audio and video signals.

AB - We present our experiments on attitude detection based on annotated multi-modal dialogue data1. Our long-term goal is to establish a computational model able to predict the attitudinal patterns in humanhuman dialogue. We believe, such prediction algorithms are useful tools in the pursuit of realistic discourse behavior in conversational agents and other intelligent man-machine interfaces. The present paper deals with two important subgoals in particular: How to establish a meaningful and consistent set of annotation categories for attitude annotation, and how to relate the annotation data to therecorded data (audio and video) in computational models of attitude prediction. We present our current results including a recommended set of analytical annotation labels and a recommended setup for extracting linguistically meaningful data even from noisy audio and video signals.

KW - Attitude detection

KW - Prediction of attitude flow

KW - Attitude annotation

KW - Multimodal speech cues

M3 - Article in proceedings

T3 - Linköping Electronic Conference Proceedings

SP - 47

EP - 53

BT - NEALT2012

A2 - Allwood, Jens

A2 - Ahlsén, Elisabeth

A2 - Paggio, Patrizia

A2 - Jokinen, Kristiina

PB - Göteborg Universitet

CY - Göteborg

T2 - The 4th Nordic Symposium on Multimodal Communication

Y2 - 15 November 2012 through 16 November 2012

ER -

Juel Henrichsen P, Allwood J. Predicting the Attitude Flow in Dialogue Based on Multi-Modal Speech Cues. In Allwood J, Ahlsén E, Paggio P, Jokinen K, editors, NEALT2012: Proceedings of the 4th Nordic Symposium on Multimodal Communication, Nov. 15-16, Gothenburg, Sweden. Göteborg: Göteborg Universitet. 2013. p. 47-53. (Linköping Electronic Conference Proceedings; No. 93).