English

SeQuiFi: Mitigating Catastrophic Forgetting in Speech Emotion Recognition with Sequential Class-Finetuning

Audio and Speech Processing 2024-10-17 v1 Sound

Abstract

In this work, we introduce SeQuiFi, a novel approach for mitigating catastrophic forgetting (CF) in speech emotion recognition (SER). SeQuiFi adopts a sequential class-finetuning strategy, where the model is fine-tuned incrementally on one emotion class at a time, preserving and enhancing retention for each class. While various state-of-the-art (SOTA) methods, such as regularization-based, memory-based, and weight-averaging techniques, have been proposed to address CF, it still remains a challenge, particularly with diverse and multilingual datasets. Through extensive experiments, we demonstrate that SeQuiFi significantly outperforms both vanilla fine-tuning and SOTA continual learning techniques in terms of accuracy and F1 scores on multiple benchmark SER datasets, including CREMA-D, RAVDESS, Emo-DB, MESD, and SHEMO, covering different languages.

Keywords

Cite

@article{arxiv.2410.12567,
  title  = {SeQuiFi: Mitigating Catastrophic Forgetting in Speech Emotion Recognition with Sequential Class-Finetuning},
  author = {Sarthak Jain and Orchid Chetia Phukan and Swarup Ranjan Behera and Arun Balaji Buduru and Rajesh Sharma},
  journal= {arXiv preprint arXiv:2410.12567},
  year   = {2024}
}
R2 v1 2026-06-28T19:24:14.272Z