Steering Language Model to Stable Speech Emotion Recognition via Contextual Perception and Chain of Thought

Large-scale audio language models (ALMs), such as Qwen2-Audio, are capable of comprehending diverse audio signal, performing audio analysis and generating textual responses. However, in speech emotion recognition (SER), ALMs often suffer from hallucinations, resulting in misclassifications or irrelevant outputs. To address these challenges, we propose <inline-formula><tex-math notation="LaTeX">$\mathbf {C^{2}SER}$</tex-math></inline-formula>, a novel ALM designed to enhance the stability and accuracy of SER through <bold>C</bold>ontextual perception and <bold>C</bold>hain of Thought (CoT). C <inline-formula><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula> SER integrates the Whisper encoder for semantic perception and Emotion2Vec-S for acoustic perception, where Emotion2Vec-S extends Emotion2Vec with semi-supervised learning to enhance emotional discrimination. Additionally, C <inline-formula><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula> SER employs a CoT approach, processing SER in a step-by-step manner while leveraging speech content and speaking styles to improve recognition. To further enhance stability, C <inline-formula><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula> SER introduces self-distillation from explicit CoT to implicit CoT, mitigating error accumulation and boosting recognition accuracy. Extensive experiments show that C <inline-formula><tex-math notation="LaTeX">$^{2}$</tex-math></inline-formula> SER outperforms existing popular ALMs, such as Qwen2-Audio and SECap, delivering more stable and precise emotion recognition. We release the training code, checkpoints, and test sets to facilitate further research.