Privacy & your voice: how TutDub protects your cloned voice

Giving an AI tool your voice is a bigger act of trust than handing over your credit card. Your voice is how people recognize you. So before you upload anything, you deserve a straight answer about what happens next.

Here's exactly what happens to your voice when you dub a video with TutDub: what's processed, what's kept, what's deleted, and the controls you hold the whole way through.

What happens to your voice after you upload

1. TutDub takes the audio from your video and picks a short reference slice — the longest continuous stretch of your speech, capped at about ten seconds. That slice is where the voice for your dubbed output comes from.
2. The slice and all intermediate files exist only for the duration of the job, in a per-job working directory on the processing machine.
3. When the job finishes and the dubbed video is saved, that working directory — reference slice included — is deleted automatically. No persistent "voice model" or voiceprint is created. Nothing reusable is left behind.
4. The finished videos are the only thing kept, and even they follow the standard 96-hour retention window. After that, they're permanently deleted from storage. The only thing that remains is job metadata (job ID, timestamps, the languages you picked) used for billing and analytics.

What we never do

• We don't create or keep a reusable voice model, voiceprint, or biometric profile of any speaker.
• We don't maintain a catalog of voices, and we don't offer celebrity or third-party voices. The clone is always derived from the video you upload.
• We don't use your uploads to train models or improve the service. Your voice is not the product.
• We don't share voice data with advertisers — or with anyone beyond the storage provider hosting your files.

Consent is on the upload screen, not in fine print

Before any job runs, TutDub asks you to tick a checkbox on the upload screen confirming two things: that you agree to clone the voice from the footage, and that you have the rights to use the content and the voices in it. The button doesn't light up until both are checked. That's not buried in a terms page — it's right where you click start.

Because every dubbed output is produced in your own voice by design, cloning can't be switched off for a job; the confirmation is simply part of the flow.

Your voice, your controls

• Delete a video or output from your account at any time — it's removed from storage immediately.
• Withdraw consent by deleting the content; we stop cloning it and delete the associated files.
• Request erasure of your data via support@tutdub.com — as standard practice, we honor the right to be forgotten and delete your metadata if you ask (though by design there's almost nothing left to erase).
• If a video contains other people's voices, confirming rights is your call — you're responsible for getting informed consent from each identifiable speaker before uploading.

One thing we ask of you

Because the output is AI-generated speech, the voice policy asks you to disclose it when you publish — a line like "Voice translated with

TutDub" in the description is enough. It's a small habit that keeps AI voices

honest, and it's the same transparency we expect of ourselves.

The short version

• No persistent voice model is ever stored — the clone exists only for the job and is deleted when it finishes.
• Finished videos are kept for 96 hours, then permanently deleted.
• Uploads are never used for training and never shared with advertisers.
• Consent and rights confirmation happen on the upload screen, every time.
• You can delete your content, withdraw consent, or request erasure at any time.

Read the full voice policy — it's short, and it's written for humans.

Want to see your own voice protected in practice? Start Free and dub a few minutes with your three welcome credits.

Read next

• Stop being scared how your voice sounds — voice cloning, reassured
• Translate your video into 14 languages, in your voice

FAQ

1. Does TutDub keep my voice after processing?

   No. The reference slice and working files are deleted when the job completes. No persistent voice model or voiceprint is created.

2. What exactly is stored after a dubbing job?

   Only the finished videos, and only for 96 hours. After that window they're permanently deleted. What remains is job metadata (job ID, timestamps, selected languages) used for billing and analytics — no voice data.

3. Can TutDub reuse my cloned voice later?

   No. The clone is discarded after the job and cannot be recreated from the metadata we keep.

4. Is my upload used to train AI models?

   No. Uploaded content is used only to produce your own dubbed output.

5. Do I need to upload a separate voice sample?

   No. The voice is cloned directly from your video — nothing extra to record or submit.

6. What if my video contains other people's voices?

   You confirm on upload that you hold the rights, and you're responsible for obtaining informed consent from each identifiable speaker before enabling cloning.

7. Do I have to tell viewers the voice is AI?

   Yes. Disclosing AI-generated voice when you publish isn't just our policy — it's a standard requirement of most content platforms, including YouTube. Any clear line works — for example, "Voice translated with #TutDub" in the description. It doesn't have to mention TutDub, but if you like the result, we'd be happy if you share TutDub with others.

8. How do I delete my data?

   Remove the video/output from your account anytime, or request erasure of retained voice-related metadata via support@tutdub.com.