---
title: "An Emotion-Preserving Conditional Information Bottleneck for Domain-Generalizable Speech Emotion Recognition"
authors:
  - "Zhichen Yuan"
  - "C. L. Philip Chen"
  - "Shuzhen Li"
  - "Tong Zhang"
doi: "10.24963/ijcai.2026/414"
url: "https://ijcai.org/proceedings/2026/414"
pdf: "https://ijcai.org/proceedings/2026/0414.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "3719-3727"
track: "Main Track"
keywords:
  - "Humans and AI: Human-computer interaction"
  - "Natural Language Processing: Speech"
  - "Natural Language Processing: Sentiment analysis, stylistic analysis, and argument mining"
  - "Machine Learning: Robustness"
---

# An Emotion-Preserving Conditional Information Bottleneck for Domain-Generalizable Speech Emotion Recognition

Zhichen Yuan, C. L. Philip Chen, Shuzhen Li, Tong Zhang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 3719-3727. DOI: [10.24963/ijcai.2026/414](https://doi.org/10.24963/ijcai.2026/414)

## Abstract

Domain-generalizable speech emotion recognition (DG-SER) aims to ensure the robustness of SER models across unknown domains, which is essential for real-world human-machine interaction systems. Most DG-SER approaches employ alignment or adversarial strategies with domain labels to promote generalization. However, these strategies often confine generalization to predefined domains, limiting robustness under diverse real-world speech variations. To address these challenges, this paper proposes an emotion-preserving conditional information bottleneck framework (EP-CIB) for domain-free DG-SER. Specifically, EP-CIB introduces a nuisance proxy representation learning module to learn a nuisance proxy as the broad non-emotional variability without requiring any domain annotations, covering both defined and previously unseen domains. It then extracts coarse-grained emotion features via the consistency-aware emotion representation learning module. EP-CIB further introduces an emotion-preserving conditional information bottleneck that, conditioned on the emotion label, disentangles the nuisance proxy from the coarse-grained emotion representation, improving domain-free generalization under open-ended domain shifts. EP-CIB departs from implicit domain surrogates in prior domain-free methods by explicitly learning a proxy for nuisance domains and disentangling it from emotion features, enabling domain-free emotion representation learning. The state-of-the-art performance in both speaker-independent and cross-corpus settings, including an 18% improvement on EmoDB-to-CASIA transfer, demonstrates the effectiveness of EP-CIB for DG-SER.