Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

RE-LLM: Refining Empathetic Speech-LLM Responses by Integrating Emotion Nuance

About

With generative AI advancing, empathy in human-AI interaction is essential. While prior work focuses on emotional reflection, emotional exploration, key to deeper engagement, remains overlooked. Existing LLMs rely on text which captures limited emotion nuances. To address this, we propose RE-LLM, a speech-LLM integrating dimensional emotion embeddings and auxiliary learning. Experiments show statistically significant gains in empathy metrics across three datasets. RE-LLM relatively improves the Emotional Reaction score by 14.79% and 6.76% compared to text-only and speech-LLM baselines on ESD. Notably, it raises the Exploration score by 35.42% and 3.91% on IEMOCAP, 139.28% and 9.83% on ESD, and 60.95% and 22.64% on MSP-PODCAST. It also boosts unweighted accuracy by 5.4% on IEMOCAP, 2.3% on ESD, and 6.9% on MSP-PODCAST in speech emotion recognition. These results highlight the enriched emotional understanding and improved empathetic response generation of RE-LLM.

Jing-Han Chen, Bo-Hao Su, Ya-Tse Wu, Chi-Chun Lee• 2026

Related benchmarks

TaskDatasetResultRank
Speech Emotion RecognitionIEMOCAP
UA76.6
14
Empathetic Response GenerationIEMOCAP
Emotional Reaction1.856
8
Empathetic Response GenerationESD
Emotional Reaction1.851
8
Empathetic Response GenerationMSP-Podcast
Emotional Reaction Score1.889
8
Speech Emotion RecognitionESD
UA98.9
5
Speech Emotion RecognitionMSP-Podcast
UA66.4
5
Showing 6 of 6 rows

Other info

Follow for update