All guides
Audio
Voice Cloning and Narration: A Practical Guide
2 min readUpdated 2026-08-22
Two ways to give your content a voice: built-in voices, ready instantly in multiple accents — and voice cloning, which learns your voice from a short sample and then reads any script as you.
Recording a sample that clones well
Ten to twenty seconds of natural speech in a quiet room is enough. Speak the way you want the narration to sound — pace and warmth carry over. Phone microphones are fine; echo and background music are not.
Consent is not optional
Clone only your own voice or a voice whose owner has clearly agreed. Impersonation is against our terms — the feature exists so creators can narrate at scale, not so anyone can borrow a voice that is not theirs.