Detail publikace

Investigation of Bottle-Neck Features for Emotion Recognition

POPKOVÁ, A. POVOLNÝ, F. MATĚJKA, P. GLEMBEK, O. GRÉZL, F. ČERNOCKÝ, J.

Originální název

Investigation of Bottle-Neck Features for Emotion Recognition

Anglický název

Investigation of Bottle-Neck Features for Emotion Recognition

Jazyk

en

Originální abstrakt

This paper describes several systems for emotion recognition developed for the AV+EC 2015 Emotion Recognition Challenge. A complete system, making use of all three modalities (audio, video, and physiological data), was submitted to the evaluation. The focus of our work was, however, on the so called Bottle-Neck features used to complement the audio features. For the recognition of arousal, we improved the results of the delivered audio features and combined them favorably with the Bottle-Neck features. For valence, the best results were obtained with video, but a two-output Bottle-Neck structure is not far behind, which is especially appealing for applications where only audio is available.

Anglický abstrakt

This paper describes several systems for emotion recognition developed for the AV+EC 2015 Emotion Recognition Challenge. A complete system, making use of all three modalities (audio, video, and physiological data), was submitted to the evaluation. The focus of our work was, however, on the so called Bottle-Neck features used to complement the audio features. For the recognition of arousal, we improved the results of the delivered audio features and combined them favorably with the Bottle-Neck features. For valence, the best results were obtained with video, but a two-output Bottle-Neck structure is not far behind, which is especially appealing for applications where only audio is available.

Dokumenty

BibTex


@inproceedings{BUT163393,
  author="Anna {Popková} and Filip {Povolný} and Pavel {Matějka} and Ondřej {Glembek} and František {Grézl} and Jan {Černocký}",
  title="Investigation of Bottle-Neck Features for Emotion Recognition",
  annote="This paper describes several systems for emotion recognition developed for the
AV+EC 2015 Emotion Recognition Challenge. A complete system, making use of all
three modalities (audio, video, and physiological data), was submitted to the
evaluation. The focus of our work was, however, on the so called Bottle-Neck
features used to complement the audio features. For the recognition of arousal,
we improved the results of the delivered audio features and combined them
favorably with the Bottle-Neck features. For valence, the best results were
obtained with video, but a two-output Bottle-Neck structure is not far behind,
which is especially appealing for applications where only audio is available.",
  address="International Speech Communication Association",
  booktitle="19th International Conference, TSD 2016, Brno , Czech Republic, September 12-16, 2016, Proceedings",
  chapter="163393",
  doi="10.1007/978-3-319-45510-5_49",
  edition="Lecture Notes in Computer Science, Lecture Notes in Artificial Intelligence",
  howpublished="print",
  institution="International Speech Communication Association",
  number="9",
  year="2016",
  month="september",
  pages="426--434",
  publisher="International Speech Communication Association",
  type="conference paper"
}