This research improves emotion recognition in conversations by incorporating speaker characteristics into large language models, enabling more accurate emotion detection in human-computer interactions.

• Introduces LaERC-S framework that enhances LLM-based emotion recognition by considering speaker-specific information
• Achieves state-of-the-art performance on multiple benchmark datasets by modeling speaker patterns and personality traits
• Demonstrates improved accuracy in detecting emotional transitions within conversations compared to traditional methods
• Enables more nuanced understanding of contextual emotions beyond just the words spoken

For support applications, this advancement means more empathetic AI assistants that can better recognize customer frustration, satisfaction, or confusion—ultimately leading to improved customer experience and more effective automated support systems.

LaERC-S: Improving LLM-based Emotion Recognition in Conversation with Speaker Characteristics