Contents
How to recreate your voiceGetting a clone that sounds like youWhat a recreated voice is good forFrequently asked questionsRecreating a voice means building an AI model of how someone speaks — tone, pace, accent — and then using it to read new scripts. In Rekreate the cloned voice, powered by ElevenLabs, becomes the narrator for your Reels, UGC ads, avatar videos and carousels.
The point is consistency at volume: publishing ten videos a week in your own voice without recording ten times, and keeping the same voice when you translate into another market.
- Clone your own voice from a short, clean recording.
- Use it for video narration, avatar dialogue and ad variations.
- Consent is not optional — clone only your voice or a voice you have written permission to use.
- Stock voices are available too if you don't want to record anything.
Recreate your voice
Sign in, add a voice sample and generate narration for your next video.
Free to try.
How to recreate your voice
Record a clean sample
One to three minutes, quiet room, one voice, no music. Read normally — the clone copies the energy of the sample, so read like you talk in your videos.
Create the voice
Upload the sample in Rekreate and give the voice a name. It appears in your voice library alongside the stock voices.
Write or recreate the script
Type the script, or take one from a recreated viral video so the words already follow a proven structure.
Generate and place it
Use the voice for narration, for an AI avatar talking head, or over AI-generated scenes, then add captions and publish.
Getting a clone that sounds like you
| Do | Why |
|---|---|
| Record in a soft room | Curtains, carpet, a closet full of clothes — reflections are what make clones sound thin |
| Keep one speaker only | Background voices get folded into the model |
| Match your delivery | A flat read produces a flat clone; speak at the pace you use on camera |
| Avoid music and effects | The model learns the noise as part of your voice |
| Punctuate the script | Commas and full stops control pauses more than any setting |
| Spell out odd names | Write brand names phonetically if the first take mispronounces them |
What a recreated voice is good for
- Faceless channels. Your voice over AI-generated scenes from Seedance or Veo — personal, without being on camera.
- UGC ads at volume. Five hook variations narrated identically, so the test measures the hook and not the read.
- Avatar videos. The avatar speaks in your voice instead of a stock one, which is what makes it feel like your brand.
- Corrections. Rewrite one line and regenerate it, instead of re-recording a whole video.
- Other languages. Keep a recognisable voice when you publish the same script in another market.
Only clone voices you are allowed to clone. Recreating someone else's voice without written consent — a celebrity, a colleague, a relative — is impersonation, is banned on most platforms, and is illegal in a growing number of countries. Disclose AI narration where the platform requires it.
Frequently asked questions
How do I recreate my voice with AI?
Record one to three minutes of clean speech, upload it as a voice in Rekreate, and the cloned voice becomes available for narration in every video tool in your account.
How much audio do I need to clone a voice?
A couple of minutes of clear, single-speaker audio is enough for a usable clone. More clean material generally means a more natural result than a longer but noisy recording.
Can I recreate someone else's voice?
Only with their written permission. Cloning a public figure or anyone who has not consented is impersonation and violates both platform rules and, in many places, the law.
Is AI voice cloning free?
New accounts get a free trial. After that, voice generation uses Rekreate credits together with the video and image tools.
Can the cloned voice speak other languages?
Yes — you can generate speech in other languages while keeping a recognisable voice, which is how brands localise the same script across markets.
Will viewers notice it is AI?
A good clone from a clean sample is hard to distinguish in short-form video, especially with music and captions. Long, emotional monologues are where synthetic voices still show.
Create your next viral video with Rekreate
Find a proven short-form video, recreate it with your script, avatar and voice, and generate scenes with Seedance, Kling and Veo in one workspace.
Start free trial →