I'm glad you're interested in integrating a custom speech-to-text (STT) implementation for the ElevenLabs Scribe model in LiveKit! To help you get started smoothly, here’s a simple step-by-step guide using the

livekit-plugins-elevenlabs

package: 1. Install the Plugin: First, let's install the ElevenLabs plugin for LiveKit Agents with pip:

pip install "livekit-agents[elevenlabs]~=1.2"

2. Authentication: You'll need an ElevenLabs API key, which you can get from here. Once you have it, set it as an environment variable like this:

ELEVEN_API_KEY=your_api_key_here

Alternatively, you can pass the API key directly when you initialize the STT instance—whichever works best for you. 3. Initialize the STT Instance: Next, use the

STT

class from the

livekit.plugins.elevenlabs

module. Here’s an example to guide you:

from livekit.plugins import elevenlabs

stt_instance = elevenlabs.STT(
    language_code="en",  # Optional: specify language
    tag_audio_events=True  # Optional: include audio event tags like (laughter)
)

4. Integrate into AgentSession: Finally, incorporate the STT instance into an

AgentSession

or

VoicePipelineAgent

. For example:

session = AgentSession(
    stt=stt_instance,
    tts=elevenlabs.TTS(),  # Optional TTS
    # ... other components
)

This integration supports non-streaming transcription with interim results and conveniently uses the

scribe_v1

model by default. For more details and additional tips, feel free to check out the STT integrations overview in the LiveKit documentation. If you have any questions along the way, I’m here to help! Sources: ElevenLabs TTS integration guide | LiveKit Docs | livekit-plugins/livekit-plugins-elevenlabs/livekit/plugins/elevenlabs/tts.p...