Tencent Releases AuK Unified Speech Generation and Editing Model
September 9, 2026
AuK provides a single natural-language interface for zero-shot TTS, acoustic editing, and paralinguistic manipulation. The AuK-Flash distilled variant achieves 4.5x faster inference speeds compared to the base model.
HOW THIS AFFECTS YOU
●
builderYou can use a single model for both synthesis and granular audio editing via text instructions.
●
designerYou can manipulate vocal emotions and acoustics using natural language prompts.