Advancing Audio Emotion and Intent Recognition with Large Pre-Trained Models and Bayesian Inference

Abstract

Large pre-trained models are essential in paralinguistic systems, demonstrating effectiveness in tasks like emotion recognition and stuttering detection. In this paper, we employ large pre-trained models for the ACM Multimedia Computational Paralinguistics Challenge, addressing the Requests and Emotion Share tasks. We explore audio-only and hybrid solutions leveraging audio and text modalities. Our empirical results consistently show the superiority of the hybrid approaches over the audio-only models. Moreover, we introduce a Bayesian layer as an alternative to the standard linear output layer. The multimodal fusion approach achieves an 85.4% UAR on HC-Requests and 60.2% on HC-Complaints. The ensemble model for the Emotion Share task yields the best rho value of .614. The Bayesian wav2vec2 approach, explored in this study, allows us to easily build ensembles, at the cost of fine-tuning only one model. Moreover, we can have usable confidence values instead of the usual overconfident posterior probabilities.

arXiv commentAccepted at ACMM 2023

References in corpus (7)

• Robust Speech Recognition via Large-Scale Weak Supervision
• SpeechBrain: A General-Purpose Speech Toolkit
• Bayesian Neural Networks: An Introduction and Survey
• Functional Variational Bayesian Neural Networks
• Fine-tuning wav2vec2 for speaker recognition
• TRILLsson: Distilled Universal Paralinguistic Speech Representations
• GradDiv: Adversarial Robustness of Randomized Neural Networks via Gradient Diversity Regularization