Thanks — I’m responding about controlling ELEVEN_LABS TTS speed. You can adjust ElevenLabs TTS speed via the

speed

parameter when configuring the ElevenLabs TTS in LiveKit. Two common integration methods: • For the LiveKit ElevenLabs plugin (when you manage your own ElevenLabs key), set `voice_settings.speed`:

from livekit.plugins import elevenlabs

session = AgentSession(
    tts=elevenlabs.TTS(
        voice_id="ODq5zmih8GrVes37Dizd",
        voice_settings=elevenlabs.VoiceSettings(
            speed=1.5  # e.g., 1.0 = normal; 0.25–4.0 supported
        )
    )
)

• For LiveKit Inference (LiveKit-hosted ElevenLabs), pass

speed

inside

extra_kwargs

(or

modelOptions

in Node.js):

from livekit.agents import inference

session = AgentSession(
    tts=inference.TTS(
        model="elevenlabs/eleven_turbo_v2_5",
        voice="Xb7hH8MSUJpSbSDYk0k2",
        extra_kwargs={
            "speed": 1.5
        }
    )
)

Notes: •

1.0

is normal speed. Valid range is typically 0.25 (slower) to 4.0 (faster). • Use whichever integration matches your setup (plugin vs. inference). If you tell me which method you’re using (plugin with your own ElevenLabs key or LiveKit Inference) and the language/voice, I can give a short ready-to-run snippet tailored to your environment. Sources: ElevenLabs TTS plugin guide | LiveKit Documentation | ElevenLabs TTS | LiveKit Documentation