By Topic

Inducing Genuine Emotions in Simulated Speech-Based Human-Machine Interaction: The NIMITEK Corpus

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)

Emotional corpora provide an important empirical foundation for investigation when researchers aim at implementing emotion-aware spoken dialog systems. One of the fundamental research questions is how to acquire an appropriate, realistic emotion corpus. The primary aim of this paper is to address the methodological desiderata in producing emotion corpora in human-machine interaction (HMI). It proposes a substantial refinement of the Wizard-of-Oz (WOZ) technique in order that a scenario designed to elicit affected speech in HMI could result in realistic and useful data. In addition, the paper reports about the NIMITEK corpus of affected behavior in HMI produced during a refined WOZ simulation. The evaluation of the corpus with respect to the perception of its emotional content demonstrated that the corpus contains recordings of emotions that were overtly signaled. The range of emotional reactions is indicative of the kind of emotional reactions than can be expected to occur in the interaction with the sort of spoken dialog systems considered in this study. Since the subjects were not restricted by given predetermined linguistic constraints on the language to use, their utterances are indicative of the way in which nontrained, nontechnical users probably like to converse with conversational agents as well.

Published in:

Affective Computing, IEEE Transactions on  (Volume:1 ,  Issue: 2 )