2024/01/23 by Yan Zhao, Zhao, Yan, Jincen Wang +11
Computer Science · Psychology · #Adaptation (eye) #Artificial intelligence #Audio and Speech Processing (eess.AS) #Class (philosophy) #Computer science #Consistency (knowledge bases) #Domain adaptation #Emotion and Mood Recognition #FOS: Computer and information sciences #FOS: Electrical engineering #Natural language processing #Psychology #Sentiment Analysis and Opinion Mining #Sound (cs.SD) #Speech recognition #Task (project management) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2401.12925
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2024/01/23 · openalex created_date 2024/01/25 · openalex updated_date 2026/07/28
Cross-corpus speech emotion recognition (SER) aims to transfer emotional knowledge from a labeled source corpus to an unlabeled corpus. However, prior methods require access to source data during adaptation, which is unattainable in real-life scenarios due to data privacy protection concerns. This paper tackles a more practical task, namely source-free cross-corpus SER, where a pre-trained source model is adapted to the target domain without access to source data. To address the problem, we propose a novel method called emotion-aware contrastive adaptation network (ECAN). The core idea is to capture local neighborhood information between samples while considering the global class-level adaptation. Specifically, we propose a nearest neighbor contrastive learning to promote local emotion consistency among features of highly similar samples. Furthermore, relying solely on nearest neighborhoods may lead to ambiguous boundaries between clusters. Thus, we incorporate supervised contrastive learning to encourage greater separation between clusters representing different emotions, thereby facilitating improved class-level adaptation. Extensive experiments indicate that our proposed ECAN significantly outperforms state-of-the-art methods under the source-free cross-corpus SER setting on several speech emotion corpora.