Chatterbox Turbo is a lightweight text-to-speech model optimized for low-latency voice agents, expressive narration, and efficient speech generation.

Voxtral Mini TTS 2603 is a fast multilingual text-to-speech model with zero-shot voice cloning and ultra-low latency streaming for real-time voice applications.

NVIDIA Riva English is a text-to-speech model that generates natural American English speech with low latency and high audio quality.

NVIDIA Riva German is a low-latency text-to-speech model delivering natural German speech for real-time applications.