Mistral AI · Voice

Your cloned voice

Your videos narrated in your own voice.

Cost
2 cr
per minute
Quality
9 /10
Speed
9 /10
Available on
Growth
and higher plans

You record 15 to 25 seconds of your voice once. Voxtral TTS, Mistral's speech model, uses that sample to narrate every video with your tone, pace and accent. The sample stays in your Soloreach account and you can delete it at any time.

What it does well

  • Your tone, pace and accent from a short sample
  • Speaks your voice in 9 languages: English, French, Spanish, Portuguese, Italian, Dutch, German, Hindi and Arabic
  • Follows the emotion of your sample, no tags needed
  • Drives the lip-sync of your own avatar

How Soloreach uses it

  1. 1

    Voiceover

    Narrates the full script of each video.

  2. 2

    Your avatar

    The voice your avatar speaks with, lip-synced by Avatar IV.

  3. 3

    Translations

    Same video in another language, still in your voice.

When to pick another model

You publish in a language Voxtral doesn't speak yet: use a studio voice, which covers 70+ languages.

Other models · Voiceover

Compare every model