Breeze TTS 2: Text to Speech

Turn a script into speech with Breeze TTS 2: English and Chinese plus a 51-language mode, a voice cloned from a three-second sample, and inline audio tags.

Input

Up to 1000 characters. Audio tags such as (sighs) are performed, not read.

One speaker, at least 3 seconds; only the first 60 seconds are read. The voice is cloned from it.

  • 0:00 / 0:00
Advanced settings

Leave empty for a new result each run. Enter a number to repeat one.

Tempo change without pitch change. 1 is the voice's own pace.

Output
Example
00:00

Bir örnek deneyin

Bu iş akışı hakkında

Breeze TTS 2 is Breeze Blue's open-weight voice model: English and Chinese natively, plus a 51-language multilingual mode. Speak a script in a featured voice, a voice cloned from a three-second recording, or one designed from a written description, and direct the performance in plain language or with inline audio tags such as (sighs) or (laughs), which are performed rather than read.

Graph: LoadAudio → BreezeVoiceClone → PreviewAny (the reusable voice id) → BreezeVoiceSelector (custom, driven by that id) → BreezeTextToSpeech (text up to 1000 characters, language English on Breeze TTS 2, delivery neutral, speed 1, seed randomized) → SaveAudioAdvanced writing FLAC. A SaveText keeps the voice id for reuse.

Two tools ship parked at mode 4: BreezeVoiceRemix (a changed copy of a voice you already have) and BreezeVoiceDesign (a new voice invented from words). Enable with Ctrl+B.