Thanks — I’m responding to your question about whether LiveKit Inference supports ElevenLabs voices. Yes. LiveKit Inference supports ElevenLabs text‑to‑speech (TTS) voices and provides several ElevenLabs TTS models you can use without managing a separate ElevenLabs account. Authentication, billing, and integration are handled by LiveKit Cloud. Quick details you can use right away • Supported models include: ◦

elevenlabs/eleven_flash_v2

◦

elevenlabs/eleven_flash_v2_5

◦

elevenlabs/eleven_turbo_v2

◦

elevenlabs/eleven_turbo_v2_5

◦

elevenlabs/eleven_multilingual_v2

• You can specify ElevenLabs TTS when creating an agent session either with a string descriptor: ◦ Example:

tts="elevenlabs/eleven_turbo_v2_5:Xb7hH8MSUJpSbSDYk0k2"

• Or with the

inference.TTS

class for more control: ◦ Example:

from livekit.agents import inference

    session = AgentSession(
        tts=inference.TTS(
            model="elevenlabs/eleven_turbo_v2_5",
            voice="Xb7hH8MSUJpSbSDYk0k2",
            language="en"
        ),
        # ... other parameters
    )

Voices and languages • LiveKit supports all default ElevenLabs voices (examples: Alice, Chris, Eric, Jessica). • Several models support multilingual output —

eleven_flash_v2_5

,

eleven_turbo_v2_5

, and

eleven_multilingual_v2

support 30+ languages including English, Spanish, French, German, Hindi, and more. What’s not supported here • Custom, community, or cloned voices from ElevenLabs are not supported through LiveKit Inference. To use those, you must connect your own ElevenLabs account via the ElevenLabs plugin. Billing and auth • No separate ElevenLabs API key is required when using LiveKit Inference. Usage is billed through your LiveKit Cloud invoice. Would you like: • Help choosing a model/voice for a specific language or use case (IVR, narration, real‑time agent voice, etc.)? • An example for integrating ElevenLabs TTS into a specific AgentSession flow or SDK (web, Python, or else)? I’m happy to walk through an example matched to your use case. Sources: Introducing LiveKit Inference: A unified model interface for voice AI | ElevenLabs TTS plugin guide | LiveKit Documentation | ElevenLabs TTS | LiveKit Documentation