Terms of sale and use

Vopus is operated by WalkerNash Development LLC, a Wyoming limited liability company. Effective date: 28 March 2026. Last revised 10 September 2026.

These terms govern the Vopus website at vopus.io, the accounts on it, the free builds, and every voice file bought through it. Creating an account or buying a voice means you accept them. If you are acting for a company, you confirm you may bind it.

1. What Vopus sells
2. Accounts and free builds
3. Buying a voice
4. What you get with a voice
5. Origin of the voice
6. Papers and the Vopus mark
7. Acceptable use
8. Other people’s engines
9. Refunds and delivery
10. No warranty
11. Limits on liability
12. Your data
13. Ending an account
14. Changes to these terms
15. Law and disputes
16. Contact

1. What Vopus sells

Vopus sells synthetic voice files. You describe a voice, the service generates it and reads an audition line in it, and if you buy it you download the audio file (a 24 kHz mono WAV) together with its Voice Authorship Papers. Vopus does not sell speech, narration or finished recordings, and does not run the voice-cloning engines that make a voice say new things. Those are yours to choose and are covered by their own terms (section 8).

2. Accounts and free builds

  1. You need an account to build voices. You must be at least 18 and give a working email address. Keep your password to yourself; what happens under your sign-in is your responsibility until you tell us it was not you.
  2. A new account receives ten free builds. A build is one generated audition. The free builds carry no obligation to buy and no card is asked for.
  3. When the ten are used, further builds require a purchase. Each purchase adds ten builds. Builds are not money and cannot be cashed out, transferred or sold.
  4. Audition audio for a voice you have not bought stays on our servers and may be removed at any time. An unbought voice is not held for you.

3. Buying a voice

  1. The price is $50 per voice, in US dollars, shown before you pay. Payment is taken by Stripe; we never see your card details.
  2. When payment settles, the file and its papers are released for download from your library and ten builds are added to your account. A receipt goes to your email address.
  3. The file you buy is the audition you approved. Nothing is re-generated after you approve it, so what you heard is what you get.
  4. Prices may change. A change never applies to a purchase already made.

4. What you get with a voice

  1. The file is yours to keep. On payment WalkerNash Development LLC grants you a perpetual, irrevocable, worldwide, royalty-free licence to use, copy, store, modify, convert and distribute the voice file and any speech produced from it, for any purpose, commercial or not, in any medium, with no further payment to us.
  2. To the extent WalkerNash Development LLC holds any copyright or other right in the file itself, it assigns that right to you on payment. Where the law does not allow assignment, the licence in the previous paragraph applies instead.
  3. You may use the file as a reference voice in any engine that accepts one, make derivative voices from it, and let people who work for or with you use it. You may sell or give away the file to someone else, in which case these terms follow it and your own rights in it end.
  4. We do not promise exclusivity. Voices are generated from descriptions and from a large catalogue of synthetic speakers, and another customer’s voice may sound similar to yours. What is unique to your purchase is the file itself, its serial number and its papers.
  5. Nothing you buy needs our servers to keep working. Your voice file does not contact us, expire, or depend on your account remaining open.
  6. The Vopus name and mark, the website and the service remain ours. Buying a voice does not let you present your work as made or endorsed by Vopus.

5. Origin of the voice

Every Vopus voice is synthesised from numerical parameters and a written description. No recording of any person is used as a sample, donor or reference at any stage of making the file. The voice is not a clone of anyone and any resemblance to a real person is coincidental. The underlying speech models were, like all such models, trained on human speech; that training does not make any individual’s voice part of your file, and we make no claim about the training data of third-party engines you use afterwards.

6. Papers and the Vopus mark

  1. The Voice Authorship Papers record how the file was made, what it measures and its SHA-256 digest. They describe the file; they are not a licence and do not grant rights on their own. Section 4 does that.
  2. Every voice carries the Vopus mark, an inaudible signal that encodes a Vopus identifier and the voice’s serial number. It carries no name or account detail, does not report back to us, and does not restrict where the file can be used. Anyone may check a file at vopus.io/verify.
  3. You must not apply a Vopus mark, or anything meant to read as one, to audio that did not come from Vopus, and you must not alter the papers so that they describe a different file. Presenting audio as a Vopus voice when it is not one is a breach of these terms.

7. Acceptable use

You may not use the service or a voice bought from it to:

  1. impersonate a real person, or present synthetic speech as a real person’s own words or recording;
  2. deceive anyone in a way that causes them harm, including fraud, scams, fake endorsements and fabricated evidence;
  3. harass, threaten or defame anyone, or produce content that is illegal where you are or where it is published;
  4. evade a platform’s consent or disclosure rules by misrepresenting what the voice is. Where you are asked, the true answer is that it is a synthetic voice, generated from a description, with no human speaker;
  5. attack, overload, scrape or reverse-engineer the service, or share an account to get around the build limits.

Where a law requires synthetic speech to be labelled as such, that is your responsibility as the publisher. We may close an account used in breach of this section. A voice already bought stays yours; the licence in section 4 survives.

8. Other people’s engines

To make a Vopus voice say new lines you load the file into a voice-cloning engine. Vopus does not operate, control or endorse any of them, whether run on your own machine or hosted by someone else. Each has its own licence and terms, some restrict commercial use, and some will not accept a synthetic reference at all. Checking that an engine’s terms fit your use is up to you. Speech that an engine produces is governed by that engine’s terms as well as by section 4 of these.

9. Refunds and delivery

  1. A voice file is delivered the moment your payment settles and cannot be returned, so a purchase is final once the file has been released for download. Before buying you can hear the whole audition as many times as you like; buy only a voice you have listened to.
  2. If a payment settles and the file is not released within one day, or the file released is not the audition you approved, tell us and we will release the right file.

10. No warranty

The service and every voice file are provided as they are. We do not promise that a voice will suit a particular purpose, sound a particular way in a particular engine, be accepted by any platform, or be free of artefacts. We do not promise the service will always be available; builds depend on a shared GPU and may queue or fail, and a failed build is not charged. Nothing in this section limits what section 9 gives you.

11. Limits on liability

To the fullest extent the law allows, WalkerNash Development LLC is not liable for indirect, incidental, special or consequential loss, or for lost profit, data or goodwill, arising from the service or a voice file. Our total liability to you for any claim connected with a purchase is limited to the amount you paid for that purchase. Some jurisdictions do not allow these limits; where that is so they apply as far as permitted.

12. Your data

  1. We keep your email address, a salted hash of your password (never the password), your builds, purchases and the voices in your library. Card details go to Stripe and are never stored by us.
  2. Your email is used for receipts, password resets and notices about your account. It is not sold and is not used for marketing unless you ask.
  3. The site sits behind Cloudflare, which sees connection data and applies its own privacy policy. A signed-in session is a cookie set by us for that purpose only.
  4. You may ask us to delete your account and its data at any time (section 16). Purchase records needed for tax and accounting are kept as the law requires.

13. Ending an account

You may close your account whenever you like. We may suspend or close an account that breaches section 7 or that is used to attack the service. Closing an account ends unused builds. It never revokes a voice you have bought; download your files first, and if you cannot, ask us and we will send them.

14. Changes to these terms

We may revise these terms. The revised version is posted here with a new date, and applies to purchases and builds made after that date. A purchase is always governed by the terms in force when it was made.

15. Law and disputes

These terms are governed by the laws of the State of Wyoming, United States, without regard to conflict-of-law rules. Any dispute that cannot be settled by talking to us first will be brought in the state or federal courts located in Wyoming, and you consent to their jurisdiction. If a court finds part of these terms unenforceable, the rest stands.

16. Contact

WalkerNash Development LLC
Email: [email protected]
For a support question, a delivery problem under section 9, or a data request under section 12, write to that address with the email you signed up with.