Latif, Siddique and Rana, Rajib 
ORCID: https://orcid.org/0000-0002-0506-2409 and Khalifa, Sara and Jurdak, Raja and Schuller, Bjorn W.
(2020)
Deep architecture enhancing robustness to noise, adversarial attacks, and cross-corpus setting for speech emotion recognition.
In: 21st Annual Conference of the International Speech Communication Association: Cognitive Intelligence for Speech
Processing (INTERSPEECH 2020), 25–29 Oct 2020, Shanghai, China.
![[img]](http://eprints.usq.edu.au/style/images/fileicons/text.png) |
Text (Published Version)
3190.pdf Restricted |
Abstract
Speech emotion recognition systems (SER) can achieve high accuracy when the training and test data are identically distributed, but this assumption is frequently violated in practice and the performance of SER systems plummet against unforeseen data shifts. The design of robust models for accurate SER is challenging, which limits its use in practical applications. In this paper we propose a deeper neural network architecture wherein we fuse Dense Convolutional Network (DenseNet), Long short-term memory (LSTM) and Highway Network to learn powerful discriminative features which are robust to noise. We also propose data augmentation with our network architecture to further improve the robustness. We comprehensively evaluate the architecture coupled with data augmentation against (1) noise, (2) adversarial attacks and (3) cross-corpus settings. Our evaluations on the widely used IEMOCAP and MSP-IMPROV datasets show promising results when compared with existing studies and state-of-the-art models.
|
Statistics for this ePrint Item |
Actions (login required)
 |
Archive Repository Staff Only |