Burst2vec: An Adversarial Multi-task Approach For Predicting Emotion, Age, And Origin From Vocal Bursts

Abstract

We present Burst2Vec, our multi-task learning approach to predict emotion, age, and origin (i.e., native country/language) from vocal bursts. Burst2Vec utilises pre-trained speech representations to capture acoustic information from raw waveforms and incorporates the concept of model debiasing via adversarial training. Our models achieve a relative 30 % performance gain over baselines using pre-extracted features and score the highest amongst all participants in the ICML ExVo 2022 Multi-Task Challenge.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyanuchitanukul2022burst2vec

Related papers

• Self-supervision And Learnable Strfs For Age, Emotion, And Country Prediction (2022)0.00
• An Efficient Multitask Learning Architecture For Affective Vocal Burst Analysis (2022)0.00
• Self-supervised Attention Networks And Uncertainty Loss Weighting For Multi-task Emotion Recognition On Vocal Bursts (2022)0.00
• Multitask Vocal Burst Modeling With Resnets And Pre-trained Paralinguistic Conformers (2022)0.00
• Jointly Predicting Emotion, Age, And Country Using Pre-trained Acoustic Embedding (2022)6.77
• Self-relation Attention And Temporal Awareness For Emotion Recognition Via Vocal Burst (2022)4.18
• Dynamic Restrained Uncertainty Weighting Loss For Multitask Learning Of Vocal Expression (2022)0.00
• Classification Of Vocal Bursts For ACII 2022 A-vb-type Competition Using Convolutional Neural Networks And Deep Acoustic Embeddings (2022)0.00