A Satisfaction-based Model for Affect Recognition from Conversational Features in Spoken Dialog Systems

Lebai Lutfi, Syaheerah Binti and Fernández Martínez, Fernando and Lucas Cuesta, Juan Manuel and López Lebón, Lorena and Montero Martínez, Juan Manuel (2013). A Satisfaction-based Model for Affect Recognition from Conversational Features in Spoken Dialog Systems. "Speech Communication", v. 55 (n. 7-8); pp. 825-840. ISSN 0167-6393. https://doi.org/10.1016/j.specom.2013.04.005.

Description

Title: A Satisfaction-based Model for Affect Recognition from Conversational Features in Spoken Dialog Systems
Author/s:
  • Lebai Lutfi, Syaheerah Binti
  • Fernández Martínez, Fernando
  • Lucas Cuesta, Juan Manuel
  • López Lebón, Lorena
  • Montero Martínez, Juan Manuel
Item Type: Article
Título de Revista/Publicación: Speech Communication
Date: September 2013
Volume: 55
Subjects:
Freetext Keywords: Automatic affect detection, Affective spoken dialog system, Domestic environment, HiFi agent, Social intelligence, Dialog features, Conversational cues, User bias, Predicting user satisfaction
Faculty: E.T.S.I. Telecomunicación (UPM)
Department: Ingeniería Electrónica
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img]
Preview
PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (5MB) | Preview

Abstract

Detecting user affect automatically during real-time conversation is the main challenge towards our greater aim of infusing social intelligence into a natural-language mixed-initiative High-Fidelity (Hi-Fi) audio control spoken dialog agent. In recent years, studies on affect detection from voice have moved on to using realistic, non-acted data, which is subtler. However, it is more challenging to perceive subtler emotions and this is demonstrated in tasks such as labelling and machine prediction. This paper attempts to address part of this challenge by considering the role of user satisfaction ratings and also conversational/dialog features in discriminating contentment and frustration, two types of emotions that are known to be prevalent within spoken human-computer interaction. However, given the laboratory constraints, users might be positively biased when rating the system, indirectly making the reliability of the satisfaction data questionable. Machine learning experiments were conducted on two datasets, users and annotators, which were then compared in order to assess the reliability of these datasets. Our results indicated that standard classifiers were significantly more successful in discriminating the abovementioned emotions and their intensities (reflected by user satisfaction ratings) from annotator data than from user data. These results corroborated that: first, satisfaction data could be used directly as an alternative target variable to model affect, and that they could be predicted exclusively by dialog features. Second, these were only true when trying to predict the abovementioned emotions using annotator?s data, suggesting that user bias does exist in a laboratory-led evaluation.

More information

Item ID: 15781
DC Identifier: http://oa.upm.es/15781/
OAI Identifier: oai:oa.upm.es:15781
DOI: 10.1016/j.specom.2013.04.005
Official URL: http://www.sciencedirect.com/science/article/pii/S0167639313000472
Deposited by: Memoria Investigacion
Deposited on: 06 Jul 2013 07:29
Last Modified: 01 Oct 2015 22:56
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM