Clone a voice from
30 seconds of audio.
Record or upload a short sample. Generate speech in that voice, in ten languages.
Three steps, no setup.
Runs in the browser. Nothing to install.
Add a voice
Record or upload 2–30 seconds. We transcribe the sample for you, so there is nothing to type. Most audio formats work.
Write the script
Up to 50,000 characters at once. Long text is split and crossfaded back together, so a chapter sounds like one take.
Generate and edit
Playback starts on the first chunk, not the last. Download WAV, MP3 or Opus, or drop the clip into a timeline.
A studio, not a text box.
Voice cloning
Three engines to choose between. Voice, engine and language all in one box, and you can repeat a take exactly or ask for a different one.
Stories
A multi-clip timeline. Trim, reorder, layer in music, pick the take you like, export the lot as one file.
Dictation Beta
Record or upload, get a transcript. Optional cleanup drops fillers and restores punctuation, and stays in the language you spoke.
Effects
Eight studio effects, chained in any order. Hear them before you commit - the untouched original is always kept.
Ten languages for cloned voices.
A cloned voice speaks all ten.
One credit, one second.
1 credit per second of audio. You pay only for what is actually produced.
Free
500 credits on signup, one per mailbox
- ≈ 8 minutes of generated audio
- One grant per mailbox - addresses with a plus sign or extra dots count as the same inbox
- The whole app: cloning, stories, effects, dictation
- Best-quality original kept; export WAV, MP3 or Opus
Starter
per month - 900 credits added monthly
- The full app, same as Free - plans only add credits
- Credits accumulate - unused ones roll over
- Balance and per-engine usage itemised in the app
- Cancel any time from the app
Pro
Recommendedper month - 2,900 credits added monthly
- ≈ 48 minutes of generated audio every month
- Credits accumulate - unused ones roll over
- Change plans any time - upgrades are prorated immediately
- One-off top-ups in the app - €5 for 500 credits, €20 for 2,000
Scale
per month - 7,900 credits added monthly
- ≈ 132 minutes of generated audio every month
- The best rate per credit
- Credits accumulate - unused ones roll over
- One-off top-ups in the app - €5 for 500 credits, €20 for 2,000
Straight answers.
How much audio do you need to clone a voice?
Two to thirty seconds of real speech - near-silent clips are rejected rather than turned into a bad voice. We transcribe the sample, so there is nothing to type.
What does a credit buy?
One second of generated audio. You start with 500, about eight minutes.
What happens if a generation fails?
Nothing - you are billed on audio actually delivered. A failure costs nothing, asking for exactly the same thing twice costs nothing the second time, and cancelling bills only what was already made. If an engine is unreachable the job falls back to another voice, and the app says so, because it will not be your cloned voice.
Where does my audio go?
Your account, voices and transcripts live on our server in France. The paid engines run in the United States, so text and reference audio leave the EU. Dictation stays in France. Full detail in the privacy policy.
Can I delete my account?
Yes, from inside the app, any time. It erases your voices, generations, stories, credit history and login. The copy held in the United States is queued and gone within 24 hours; off-site backups age out over about 30 days.
Can I use it commercially?
Yes, for voices you have the right to use. Cloning someone’s voice without their permission is not something we will help with. See the terms.
Try it with 500 free credits.
No card, no trial timer. Clone a voice and make eight minutes of audio.