Private beta30 seats for the $9 test film.Check a niche

DocsMaking films

Voice library and voice cloning

Pick a natural narration voice from the library, or clone your own voice with your consent.

Status: Beta

Beta

This runs in the studio and opens to testers with the private beta. Details can still change. Changelog

The voice library

On the channel page under Look, choose Voice. Each voice has a short sample in each language it supports, and you can filter by language, tone and pace. Set a default voice per channel; you can change it per film in the editor.

All voices produce word timing, which VidSwarm uses for captions, for cutting scenes on the beat of the narration and for re-voicing single sentences.

Pronunciation

Names and places are where synthetic voices go wrong. In the editor, select a word and type how it should sound, or ask the agent: “Say Nijmegen like a Dutch speaker would”. Saved pronunciations apply to every film of the channel.

Cloning your voice

  • Choose “Add my voice” on the voice page.
  • Read the consent sentence shown on screen. It must be the same person as the voice being cloned.
  • Record or upload a few minutes of clean speech: one speaker, no music, little room echo.
  • Wait for the voice to be built, then listen to the test lines before using it.

You can clone only your own voice or the voice of someone who records the consent sentence themselves. Cloned voices are private to your account and can be deleted at any time from the same page.

Re-voicing

To change one sentence, edit it in the script panel; only that sentence is re-voiced and the scene timing adjusts. Re-voicing the whole film, for example with a different voice, costs 0.5 min after the included improvements.