Skip to main content
Integrations and SDKs

LiveKit integration

Learn how to integrate Speechmatics STT with LiveKit Agents.

LiveKit Agents is a framework for building real-time, voice-enabled AI applications that connect with LiveKit rooms. With the Speechmatics STT plugin, you enable your voice agent to transcribe live audio in real-time, along with speaker diarization, turn detection and noise robustness.

LiveKit integrations are perfect for:

  • Voice AI: voice assistants, chatbots, and IVR systems
  • Transcription: live events or recordings
  • Accessibility: screen readers and assistive technologies, in-app help widgets
  • Media: news broadcasts, automated announcements

Features​

  • Realtime transcription: instant, accurate speech-to-text
  • Speaker diarization: identify and separate multiple speakers automatically
  • Turn detection: detect natural speech boundaries and pauses
  • Noise robustness: maintain accuracy in challenging environments
  • Global language support: works with diverse accents and dialects
  • Partial results: receive interim transcriptions for faster response times
  • High accuracy: industry-leading word recognition for natural conversations

Quickstart​

Requirements​

Installation​

uv add "livekit-agents[speechmatics]~=1.2"

Usage​

import os
import asyncio

from livekit.agents import AgentSession, cli
from livekit.plugins import speechmatics

async def run_agent(room):
    session = AgentSession(
        stt=speechmatics.STT(
            api_key=os.environ["SPEECHMATICS_API_KEY"],
        ),
    )
    await session.join(room)

if __name__ == "__main__":
    cli.run_app(run_agent)

For detailed examples, please see the Speechmatics Academy.