PACT: A Pragmatic Prompting Framework for Expressive and Voice-Preserving Speech-to-Speech Translation
PACT is proposed, a target-space pragmatic prompting framework for expressive and voice-preserving S2ST, which achieves the best overall performance among evaluated systems across content fidelity, voice preservation, and expression preservation metrics.
Yu-Chen Song, Satoshi Nakamura
· 0 citations