Chatterbox Turbo – Fast, expressive open-source TTS with native watermarking

Summary: Chatterbox Turbo is a 350M parameter open-source text-to-speech model featuring paralinguistic tags for emotional control, zero-shot voice cloning, and runs six times faster than real-time. It includes built-in PerTh watermarking for safety and traceability.

What it does

It generates expressive speech using paralinguistic tags like [laugh] or [sigh] and supports zero-shot voice cloning while operating at 6x real-time speed. The model integrates PerTh watermarking to ensure content traceability.

Who it's for

Developers and researchers seeking a fast, controllable, and open-source TTS solution with built-in safety features.

Why it matters

It provides a practical, high-speed TTS model with emotional nuance and embedded watermarking to enhance safety in voice synthesis applications.