Fully automatic audiovisual emotion recognition: Voice, words, and the face

Martin Wöllmer, Moritz Kaiser, Florian Eyben, Felix Weninger, Björn Schuller, Gerhard Rigoll

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

The recognition of human emotions from spontaneous and non-prototypical real-life data is currently one of the most challenging tasks in the field of affective computing. This contribution presents our recent advances in assessing dimensional representations of emotion, such as arousal, expectation, power, and valence, in an audiovisual humancomputer interaction scenario. We propose a fully automatic multimodal recognition approach based on context-sensitive modeling of audio and video features. Evaluations on the Audiovisual Sub-Challenge of the 2011 Audio/Visual Emotion Challenge show how accurately different affective dimensions can be recognized. Our experiments reveal that the proposed multimodal recognition system outperforms previously introduced techniques evaluated on the same task.

Original languageEnglish
Title of host publicationProceedings of 10th ITG Symposium on Speech Communication
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9783800734559
StatePublished - 2012
Event10th ITG Symposium on Speech Communication, ITGspeech 2012 - Braunschweig, Germany
Duration: 26 Sep 201228 Sep 2012

Publication series

NameProceedings of 10th ITG Symposium on Speech Communication

Conference

Conference10th ITG Symposium on Speech Communication, ITGspeech 2012
Country/TerritoryGermany
CityBraunschweig
Period26/09/1228/09/12

Fingerprint

Dive into the research topics of 'Fully automatic audiovisual emotion recognition: Voice, words, and the face'. Together they form a unique fingerprint.

Cite this