AI RESEARCH
On the Emotion Understanding of Synthesized Speech
arXiv CS.CL
•
ArXi:2603.16483v1 Announce Type: new Emotion is a core paralinguistic feature in voice interaction. It is widely believed that emotion understanding models learn fundamental representations that transfer to synthesized speech, making emotion understanding results a plausible reward or evaluation metric for assessing emotional expressiveness in speech synthesis. In this work, we critically examine this assumption by systematically evaluating Speech Emotion Recognition (SER) on synthesized speech across datasets, discriminative and generative SER models, and diverse synthesis models.