Adversarial auto-encoders for speech based emotion recognition

Saurabh Sahu, Rahul Gupta, Ganesh Sivaraman, Wael AbdAlmageed, Carol Espy-Wilson

Abstract:

Recently, generative adversarial networks and adversarial auto-
encoders have gained a lot of attention in machine learning
community due to their exceptional performance in tasks such
as digit classification and face recognition. They map the auto-
encoder’s bottleneck layer output (termed as code vectors) to
different noise Probability Distribution Functions (PDFs), that
can be further regularized to cluster based on class informa-
tion. In addition, they also allow a generation of synthetic sam-
ples by sampling the code vectors from the mapped PDFs. In-
spired by these properties, we investigate the application of ad-
versarial auto-encoders to the domain of emotion recognition.
Specifically, we conduct experiments on the following two as-
pects: (i) their ability to encode high dimensional feature vec-
tor representations for emotional utterances into a compressed
space (with a minimal loss of emotion class discriminability in
the compressed space), and (ii) their ability to regenerate syn-
thetic samples in the original feature space, to be later used for
purposes such as training emotion recognition classifiers. We
demonstrate promise of adversarial auto-encoders with regards
to these aspects on the Interactive Emotional Dyadic Motion
Capture (IEMOCAP) corpus and present our analysis.