Record about twenty seconds of your voice, saying our consent line, and Timbre reads any script you give it in that voice - WAV and MP3, made on our own GPUs with a model whose licence lets you sell the result. Every clip carries an inaudible watermark.
Nothing opens in this browser — this one takes a job and does the work on our own graphics cards in Oregon. The free allowance below is how you try it: 10 minutes, no card and no payment, on the key you are issued.
Runs on this computer today. It opens to the public when two pieces are in place — see below.
The problem
The hosted voice-cloning services ban adult audio in their terms, so an audio creator who wants to publish more than they can record has nowhere to go - and the free models that will run anywhere carry non-commercial licences.
A voice is only made from a recording where the speaker says our consent line with their own name. A clip lifted from somewhere else is refused.
Every clip carries an inaudible watermark, and anyone can check whether a recording came out of Timbre.
The voice model's licence allows the audio to be sold. No non-commercial weights.
Timbre answers this computer only until it runs on GPU capacity whose terms allow adult content and has a public https address.
How it works
Ten to ninety seconds of your voice, with the consent line and your name in it.
Our own speech recogniser listens for the line. Ready in about a minute.
Pick a delivery - calm, natural or expressive - and download MP3, WAV and a provenance note.
What it will never do
Consent is spoken, recorded and kept with every voice.
Pricing
A unit is one minute of finished audio, rounded up per job.
Hear yourself read a scene.
A month of episodes.
A series, or a roster of performers.
Paid in bitcoin through our own checkout; the minutes are added to your key as soon as the payment confirms. Your licence key is emailed to you within two working days of payment.
Checkout is our own bitcoin server, not a card processor. The units you buy are added to your key: buy from the Buy more box inside the program and they arrive on their own once the payment confirms, or buy here and paste the invoice number from your receipt into that box.
Questions
So nobody can make a voice from a clip of you they found online. The line, with your name, has to be in the recording, and our recogniser checks it is.
Yes, if they record the consent line themselves and you attach their signed written consent. Both are kept with the voice.
Yes. The model is Chatterbox by Resemble AI, released under the MIT licence, which allows commercial use.
An inaudible mark the model puts in everything it makes. It survives MP3. Check a clip on the page tells you whether it is there.
No. A script goes to our GPU for the one job; only its first line is kept so you can recognise the job.
A key. One line of text, issued to you and shown exactly once — only its hash is kept here, so it cannot be looked up later and we could not tell you what yours is. Every key arrives with a free allowance already on it, which is the tier at the top of the pricing above, and the work starts the first time you send a job with that key on it.
Each job is metered in the unit named in the pricing, and a unit that fails — a file that will not read, a row the model held back — is not charged and is named in the job report rather than silently skipped. When an allowance runs out the next job is refused before any work starts, with the number it used printed on the refusal.
More units are bought in bitcoin. The checkout is a bitcoin server we run ourselves. Buy from the Buy more box inside the program and the units are added to your key on their own as soon as the payment confirms; buy from the plan buttons above and paste the invoice number from your receipt into that same box. Each payment adds its units once, to one key.
Make a key, record the consent line, paste a script.