Self-EmoQ: Plutchik-Guided Value-based Planning to Drive Streaming Emotional TTS
Researchers have developed a new framework for conversational AI that enables systems to determine and express emotions in a streaming text-to-speech (TTS) manner. This approach uses a plug-and-play LLM module trained with reinforcement learning, incorporating Plutchik's wheel of emotions to guide the emotional output. Experiments show this method surpasses traditional prompting and fine-tuning techniques in both emotion determination and response quality, leading to a more emotionally aligned and fluent user experience. AI
IMPACT Enhances conversational AI by enabling more natural and contextually aware emotional expression in speech synthesis.