ABHINAYA -- A System for Speech Emotion Recognition In Naturalistic Conditions Challenge

TL;DR

Abhinaya is a multi-modal speech emotion recognition system that combines speech, text, and speech-text models, employing advanced fine-tuning and class imbalance techniques to achieve state-of-the-art results in naturalistic conditions.

Contribution

This work introduces a novel multi-modal SER system integrating speech and text models with tailored training strategies for naturalistic data.

Findings

Ranked 4th among 166 submissions in the challenge

Achieved state-of-the-art performance after full training

Effectively handled class imbalance and variability

Abstract

Speech emotion recognition (SER) in naturalistic settings remains a challenge due to the intrinsic variability, diverse recording conditions, and class imbalance. As participants in the Interspeech Naturalistic SER Challenge which focused on these complexities, we present Abhinaya, a system integrating speech-based, text-based, and speech-text models. Our approach fine-tunes self-supervised and speech large language models (SLLM) for speech representations, leverages large language models (LLM) for textual context, and employs speech-text modeling with an SLLM to capture nuanced emotional cues. To combat class imbalance, we apply tailored loss functions and generate categorical decisions through majority voting. Despite one model not being fully trained, the Abhinaya system ranked 4th among 166 submissions. Upon completion of training, it achieved state-of-the-art performance among published results, demonstrating the effectiveness of our approach for SER in real-world conditions.