James (male, American)Sheila (female, American)Margaret (female, American)Rob (male, American)Hugo (male, British)Howard (male, British)Fred (male, British)Harriet (female, British)David (male, British)Mary (female, British)Sky (female, British)Maryam (female, Arabic)Jordi (male, Catalan)Heidi (female, German)Elena (female, Spanish)Marie (female, French)Manon (female, French)Lea (female, French)Camille (female, French)Himanshu (male, Hindi)Ananya (female, Hindi)Olivia (female, Hungarian)Lorenzo (male, Italian)Haruto (male, Japanese)Sung (male, Korean)Jin (male, Korean)Lotte (female, Dutch)Lars (male, Norwegian)Leon (male, Polish)Gabriela (female, Portuguese)Irina (female, Romanian)Tatiana (female, Russian)Oran (male, Turkish)Taras (male, Ukrainian)Huong (female, Vietnamese)Mei (female, Chinese)

ExpressivenessNumber

How strongly the voice delivers the text, from 1 to 5. Higher is more expressive and less like the voice. Unset, Mistral picks. In the text itself, [emotion=happy] changes the emotion from that point and [pace=slow] the pace.

Range: 1–5 (step 0.5)

Output FormatSelect

Default: mp3

MP3WAVFLACOpus

Related models

Explore similar models

Start building with Voxtral TTS 3

No API keys required. Create AI-powered workflows with Voxtral TTS 3 in minutes — free.

Get Started Free Explore All Models

Voxtral TTS 3

Documentation & links

Parameters & options

Explore similar models

Voxtral Mini Transcribe 3

Voxtral Small Transcribe 3

Mistral Large 4

Gemini 3.8 Flash-Lite TTS

Qwen Audio 3.0 TTS Plus

Inworld TTS-2

Start building with Voxtral TTS 3