DramaBox v1.0.0
DramaBox is Resemble AI's open-source expressive text-to-speech model with prompt-driven voice cloning. Built on LTX-2.3 (3.3B audio architecture), it lets the prompt control speaker identity, emotion, delivery style, laughs, sighs, and transitions — with optional 10-second voice reference for timbre cloning.
Prompt-driven TTS with fine-grained emotion and delivery control
Voice cloning from a 10-second reference audio clip
Verifiable digital signatures for AI-generated speech (watermarking)
PricingFree (Open Source)
Version1.0.0