Distribution-based Emotion Recognition in Conversation

TL;DR

This paper introduces a distribution-based framework for emotion recognition in conversation that models emotion as a distribution, effectively handling label ambiguity and improving both accuracy and uncertainty estimation.

Contribution

It proposes a novel sequence-to-sequence distribution-based approach with Bayesian training, advancing emotion recognition by capturing uncertainty and handling subjective labels.

Findings

Outperforms single-utterance systems in accuracy

Improves uncertainty estimation in emotion recognition

Demonstrates effectiveness on IEMOCAP dataset

Abstract

Automatic emotion recognition in conversation (ERC) is crucial for emotion-aware conversational artificial intelligence. This paper proposes a distribution-based framework that formulates ERC as a sequence-to-sequence problem for emotion distribution estimation. The inherent ambiguity of emotions and the subjectivity of human perception lead to disagreements in emotion labels, which is handled naturally in our framework from the perspective of uncertainty estimation in emotion distributions. A Bayesian training loss is introduced to improve the uncertainty estimation by conditioning each emotional state on an utterance-specific Dirichlet prior distribution. Experimental results on the IEMOCAP dataset show that ERC outperformed the single-utterance-based system, and the proposed distribution-based ERC methods have not only better classification accuracy, but also show improved uncertainty estimation.