Discriminatively Trained Recurrent Neural Networks for Continuous Dimensional Emotion Recognition from Audio

@inproceedings{Weninger2016DiscriminativelyTR,
  title={Discriminatively Trained Recurrent Neural Networks for Continuous Dimensional Emotion Recognition from Audio},
  author={Felix Weninger and Fabien Ringeval and Erik Marchi and Bj{\"o}rn W. Schuller},
  booktitle={IJCAI},
  year={2016}
}
Continuous dimensional emotion recognition from audio is a sequential regression problem, where the goal is to maximize correlation between sequences of regression outputs and continuous-valued emotion contours, while minimizing the average deviation. As in other domains, deep neural networks trained on simple acoustic features achieve good performance on this task. Yet, the usual squared error objective functions for neural network training do not fully take into account the above-named goal… CONTINUE READING
Highly Cited
This paper has 19 citations. REVIEW CITATIONS
13 Citations
40 References
Similar Papers

Citations

Publications citing this paper.
Showing 1-10 of 13 extracted citations

References

Publications referenced by this paper.
Showing 1-10 of 40 references

6(2):97–108

  • Soroosh Mariooryad, Carlos Busso. Correcting time-continuous emotional labels Computing
  • April-June
  • 2015
1 Excerpt

74(22):9983–10000

  • Fengna Wang, Hichem Sahli, +4 authors Applications
  • November
  • 2015
1 Excerpt

Jürgen Schmidhuber , and Fred Cummins . Learning to forget : Continual prediction with LSTM

  • Felix A. Gers
  • Neural Computation
  • 2015

Similar Papers

Loading similar papers…