How to Clone Your Voice with AI

Updated September 3, 2026

TL;DR: record one clean minute of yourself reading naturally, upload it on LeapFun's Starter plan or above, confirm the voice is yours, then audition with a real paragraph. Most disappointing clones come from noisy samples, not short ones. Full guide with the legal side on the voice cloning guide.

Before you record

Step by step

  1. Record the sampleRead at your natural pace, one take, no music, no processing. Record two or three candidates in the same session.
  2. Upload and confirm consentIn Voice Cloning, upload the file and confirm the voice is yours or that you have the speaker's permission. The confirmation is stored with the clone.
  3. AuditionGenerate the same paragraph with each candidate. Listen on headphones and on a phone speaker; keep the best and delete the rest.
  4. Use it everywhereThe clone appears under My Voices: pick it in Text to Speech, or assign it as narrator or a character in Audiobook Studio. Every export is loudness-mastered.

Getting it closer to you

What you can and can't clone

Your own voice, or a voice whose owner has agreed — a client under contract, a collaborator who has said yes. Not a public figure, not a family member who wasn't asked. LeapFun records consent for every clone and the acceptable-use policy applies.

Not ready to record? Audition preset voices first — real samples:

Frequently asked questions

How long does cloning take?

Minutes. Upload, confirm consent, and the clone is ready to audition; there is no training queue for instant clones.

Can I clone in Chinese?

Yes. Clones read English and Chinese; record the sample in whichever language you'll mostly use, and the timbre carries to the other.

What if my clone mispronounces my name?

Add a pronunciation rule in a Studio project — an alias respelling works on every voice tier — and it applies to every chapter.

Can I delete the sample later?

Yes. Remove the voice under My Voices and the stored sample goes with it.