Voice Consent Policy

1. Voices on Vocla

• Library voices are synthetic voices licensed to us or produced for Vocla. Their use is covered by our Terms.
• Voice design creates a new synthetic voice from a written description (for example age, tone and dialect). Reference voices you choose from the library can influence style, but a design never copies the identity of a real person.
• Voice changer re-voices your recording into a library voice: your words are transcribed and spoken again by the voice you pick. It does not create a copy of the original speaker's voice.
• Voice cloning creates a private synthetic voice from short reference recordings you upload or record. It is available on paid plans, within your plan's limits, and only under the rules below. Learn more about voice cloning.

2. Consent for recordings you upload

You may upload recordings (for the voice changer, voice cleaner, dubbing, voice cloning or the assistant) only if they contain your own voice, or the voices of people who have given you permission for that use. Tell people when you record them, and do not upload recordings made secretly or in breach of the law. You are responsible for obtaining and keeping evidence of that permission, and we may ask you to provide it.

3. Rules for voice cloning

When you clone a voice on Vocla:

• You may clone only your own voice, or the voice of a person who has given you express, written permission for that use. You are responsible for keeping evidence of that permission, and we may ask you to provide it.
• Cloning or imitating public figures (such as politicians, officials, religious leaders and celebrities), or any other person without their consent, is prohibited.
• Cloned voices are private to the workspace that created them and are not listed in the public voice library.
• You can delete a cloned voice at any time; deleting it also deletes the underlying voice model.
• Reference recordings are processed as described in our Privacy Policy, including by a specialised voice-model provider acting as our processor.

We act on reports of voices cloned without consent, and we may remove cloned voices and suspend accounts that break these rules.

4. No imitation of real people

You may not use any Vocla feature — including voice-design descriptions such as “sounds like [a named person]” — to create a voice that imitates an identifiable real person, unless that person has given you express, written, informed consent for that specific use. Voices imitating public figures (such as politicians, officials, religious leaders and celebrities) are not allowed at all, and we block known attempts.

5. Custom voices under an enterprise agreement

If our team creates a custom voice for you based on a real person's recordings, we will do so only under a separate written agreement and only if all of these conditions are met:

1. the speaker is an adult and gives written, informed and specific consent, including how the voice will be used and for how long;
2. the speaker records a spoken consent statement, which we check against the voice recordings;
3. the speaker's identity is verified; and
4. the consent record is stored with the voice and the speaker can revoke it at any time.

Public figures' voices are never accepted.

6. Withdrawing consent

A person whose voice is used on Vocla with their consent can withdraw it at any time by emailing legal@vocla.ai. We will disable the voice for new generations and remove it from the account within five business days of verifying the request. Audio lawfully generated before withdrawal is governed by the agreement between that person and the user.

7. Provenance and AI labelling

For every file generated on Vocla we keep an internal record — including a cryptographic fingerprint (SHA-256) of the file and a label that it is AI-generated — so we can answer whether a given file was produced on Vocla. We do not currently embed a verifiable watermark or signed content credentials in downloads, and we cannot guarantee that watermarks applied by AI providers survive processing or editing.

When you publish synthetic audio, label it as AI-generated wherever the law or the platform requires, and never present it as a real recording of someone who did not say it.

8. Reporting misuse of a voice

If you believe your voice, or the voice of someone you represent, has been imitated or used without consent on Vocla, email legal@vocla.ai with the audio or link, an explanation, and how we can confirm your identity or authority. You do not need a Vocla account. We acknowledge reports within two business days, can check the file against our provenance records, disable the voice or share link concerned, preserve evidence and cooperate with lawful requests from authorities.

9. Enforcement

Breaking this policy is a serious breach of our Terms and Acceptable Use Policy. We may remove content, disable voices, and suspend or terminate accounts without refund.