🎙️ Breeze TTS 2
Open-weight bilingual (English/Chinese) text-to-speech with voice design, voice cloning, and voice direction. Supports inline vocal events: (laugh), (sigh), (cough), (clears throat).
Create a distinctive voice from a natural-language description — no reference audio needed. Use cfg_scale = 4 for stronger instruction following.
| Text to speak | Voice description | CFG scale | Seed |
|---|
Clone a speaker from clean reference audio and its exact transcript. Choose a voice from the catalog or upload your own clean reference recording.
Preview clips span many languages; generated Breeze speech remains limited to English and Chinese.
11,430 voices found
Page 1 / 572
No catalog voice selected. You can also upload or record one below.
Clone a reference voice while directing tone, emotion, pace, and delivery. Use cfg_scale = 4 for stronger instruction following.
Model: BreezeBlue/Breeze-TTS-2 · Code: breezeblue-ai/breeze-tts · Weights/outputs are for research and non-commercial use only.