ElevenLabs Voiceover

ElevenLabs Voiceover is the industry-standard AI voice: natural, expressive narration from any script, straight onto your timeline.

ElevenLabs is the benchmark for natural text-to-speech, and Hiroska wires it in as the voiceover model. Pick a voice, paste or write the script, and the narration lands as audio clips aligned to your edit.

It pairs with the Director: write timed moments, generate voiceover per moment, and the timeline keeps narration in sync with the visuals it describes. Billing is by character count, so short lines cost very little.

At a glance

  • Type: Text-to-speech (voiceover)
  • Voices: Curated voice picker
  • Pricing: By script length (per character)
  • Integration: Per-moment Director voiceover + timeline ducking

What it is best at

  • Natural pacing, emphasis and breath
  • Per-moment generation keeps narration synced to the edit
  • Character-based billing, so short lines cost almost nothing

What it costs

About 3 tokens ($0.03) per 250 characters; the exact price depends on the length of what you upload.

SettingsTokensPrice
250 characters3$0.03
1,000 characters10$0.10
5,000 characters47$0.47

Prices come from the live catalog at $0.0100 per token, the pay-as-you-go rate; subscriptions include tokens at a lower rate. The exact price is shown before you generate, and a failed generation is refunded.

Questions

How does ElevenLabs Voiceover work with the Director?
Write your script as timed moments, then generate voiceover per moment. Each clip lands on the audio lane at its moment’s time, and background music ducks under it automatically.
How is ElevenLabs Voiceover priced?
By the character count of the text you synthesise, converted to tokens and shown before you generate. Short reads stay cheap.
Can I preview ElevenLabs voices?
The voice picker lets you choose per generation. Regenerating a moment with a different voice is one click, and only the regenerated characters are billed.

Related models