Late Fusion of Individual Engines for Improved Recognition of Negative Emotion in Speech - Learning vs. Democratic Vote

Schuller B, Metze F, Steidl S, Batliner A, Eyben F, Polzehl T (2010)

Publication Language: English

Publication Type: Conference contribution, Conference Contribution

Publication year: 2010

Original Authors: Schuller Björn, Metze Florian, Steidl Stefan, Batliner Anton, Eyben Florian, Polzehl Tim

Pages Range: 5230-5233

Conference Proceedings Title: Proceedings of ICASSP

Event location: Dallas US

URI: http://www5.informatik.uni-erlangen.de/Forschung/Publikationen/2010/Schuller10-LFO.pdf

Abstract

The fusion of multiple recognition engines is known to be able to outperform individual ones, given sufficient independence of methods, models, and knowledge sources. We therefore investigate late fusion of different speech-based recognizers of emotion. Two generally different streams of information are considered: acoustics and linguistics fed by state-of-the-art automatic speech recognition. A total of five emotion recognition engines from different sites that provide heterogeneous output information are integrated by either simple democratic vote or learning ‘which predictor to trust when’. We are able to significantly outperform the best individual engine by fusion, and the so far best reported result on the recently introduced Emotion Challenge task.

Authors with CRIS profile

Stefan Steidl Lehrstuhl für Informatik 14 (Bild- und Sprachverarbeitung) Anton Batliner Lehrstuhl für Informatik 14 (Bild- und Sprachverarbeitung)

Involved external institutions

Technische Universität München (TUM) DE Germany (DE) Carnegie Mellon University US United States (USA) (US) Technische Universität Berlin DE Germany (DE)

How to cite

APA:

Schuller, B., Metze, F., Steidl, S., Batliner, A., Eyben, F., & Polzehl, T. (2010). Late Fusion of Individual Engines for Improved Recognition of Negative Emotion in Speech - Learning vs. Democratic Vote. In ICASSP (Eds.), Proceedings of ICASSP (pp. 5230-5233). Dallas, US.

MLA:

Schuller, Björn, et al. "Late Fusion of Individual Engines for Improved Recognition of Negative Emotion in Speech - Learning vs. Democratic Vote." Proceedings of the ICASSP 2010, IEEE International Conference on Acoustics, Speech, and Signal Processing, Dallas Ed. ICASSP, 2010. 5230-5233.

BibTeX: Download