Skip to main content

Mistral STT plugin guide

How to use the Mistral STT plugin for LiveKit Agents.

Available in
Python
|
Node.js

Overview

This plugin allows you to use Voxtral  as an STT provider for your voice agents.

Installation

Install the LiveKit Mistral AI plugin from PyPI or npm:

uv add "livekit-agents[mistralai]~=1.5"
pnpm add @livekit/agents-plugin-mistralai

Authentication

The Mistral AI integration requires a Mistral AI API key .

Set the MISTRAL_API_KEY in your .env file.

Usage

Use Mistral AI STT in your AgentSession or as a standalone transcription service. For example, you can use this STT in the Voice AI quickstart.

from livekit.plugins import mistralai
session = AgentSession(
stt=mistralai.STT(
model="voxtral-mini-latest"
),
# ... llm, tts, etc.
)
import * as mistralai from '@livekit/agents-plugin-mistralai';
const session = new voice.AgentSession({
stt: new mistralai.STT({
model: "voxtral-mini-latest"
}),
// ... llm, tts, etc.
});

Parameters

This section describes some of the available parameters. See the plugin reference for a complete list of all available parameters.

modelstringDefault: voxtral-mini-latest

Name of the Voxtral STT model to use.

languagestring

The language code to use for better transcription accuracy if already known. Only used with batch models.

context_biaslist[str]

Words or phrases to guide the model toward good spelling or names or domain-specific vocabulary. Up to 100 items can be provided. Only used with batch models.

target_streaming_delay_msint

Target streaming delay in milliseconds for realtime mode. Only used with realtime models.

vadvad.VAD

Voice Activity Detector used to trigger audio flush for realtime models (which lack server-side endpointing). When not provided, Silero VAD is auto-loaded with default settings. Only used with realtime models.

Additional resources

The following resources provide more information about using Mistral AI with LiveKit Agents.