Your voice. Cloned from 15 seconds.
Regent clones your voice from a single 15-second sample and uses it across every reel, every script, every post. No more recording takes.
What it is
A voice clone is built once from a short, clean recording. Regent then drives every generated reel, narration, or audio asset using that voice — so the output sounds like you, not a stock TTS narrator.
How it works
- 1Record a 15-second sample in a quiet space, or upload an existing clean clip.
- 2Regent processes the sample and locks the clone to your account.
- 3Every script Regent writes can be rendered in your voice automatically — no extra step.
- 4Re-record any time to replace the clone (for example, after a voice change or accent shift).
Specs
- Sample length
- 15 seconds
- Output use
- Reels, narration, audio posts
- Account binding
- Voice locked to your creator account
- Re-clone
- Anytime
We have an Autopilot mode. Flip it on and no manual approvals are needed — you sleep, Regent handles everything from research to post.
Questions about voice cloning
How long does the voice sample need to be?+
Exactly 15 seconds of clean speech. Longer samples can be uploaded but Regent only uses the first ~15 seconds for the clone.
Can I clone someone else’s voice?+
No. Regent only clones a voice that belongs to you, or a voice for which you hold explicit, documented consent. Uploading another person’s voice without authorization is a violation of our Acceptable Use Policy and the AI Avatar & Voice Consent.
Can I change my cloned voice later?+
Yes. Upload a new 15-second sample anytime and the clone updates. Useful after a voice change, recording-environment change, or rebrand.
Does the voice clone work in languages other than English?+
Multilingual mode is available — the clone reads scripts in the target language using the original voice timbre. Quality is best in English at v1.
Related features
Public beta — 100 creator cap