Reusing Neural Speech Representations for Auditory Emotion Recognition

Egor Lakomkin, Cornelius Weber, Sven Magg, Stefan Wermter


Abstract
Acoustic emotion recognition aims to categorize the affective state of the speaker and is still a difficult task for machine learning models. The difficulties come from the scarcity of training data, general subjectivity in emotion perception resulting in low annotator agreement, and the uncertainty about which features are the most relevant and robust ones for classification. In this paper, we will tackle the latter problem. Inspired by the recent success of transfer learning methods we propose a set of architectures which utilize neural representations inferred by training on large speech databases for the acoustic emotion recognition task. Our experiments on the IEMOCAP dataset show ~10% relative improvements in the accuracy and F1-score over the baseline recurrent neural network which is trained end-to-end for emotion recognition.
Anthology ID:
I17-1043
Volume:
Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
Month:
November
Year:
2017
Address:
Taipei, Taiwan
Editors:
Greg Kondrak, Taro Watanabe
Venue:
IJCNLP
SIG:
Publisher:
Asian Federation of Natural Language Processing
Note:
Pages:
423–430
Language:
URL:
https://aclanthology.org/I17-1043
DOI:
Bibkey:
Cite (ACL):
Egor Lakomkin, Cornelius Weber, Sven Magg, and Stefan Wermter. 2017. Reusing Neural Speech Representations for Auditory Emotion Recognition. In Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 423–430, Taipei, Taiwan. Asian Federation of Natural Language Processing.
Cite (Informal):
Reusing Neural Speech Representations for Auditory Emotion Recognition (Lakomkin et al., IJCNLP 2017)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-5/I17-1043.pdf
Data
IEMOCAP