BLSP Emo

BLSP-Emo research focuses on developing large speech-language models capable of understanding and generating emotionally nuanced speech. Current efforts concentrate on improving emotion preservation in speech codecs, creating datasets for emotion-cause analysis, and developing novel model architectures, including those based on LLMs and diffusion models, to achieve more accurate and controllable emotional speech synthesis and empathetic responses. This work is significant for advancing human-computer interaction, improving accessibility for individuals with communication challenges, and furthering our understanding of emotion in language and speech.