flaky-beard-91685
02/28/2025, 7:00 AMagent = VoicePipelineAgent(
vad=ctx.proc.userdata["vad"],
stt=deepgram.STT(),
llm=openai.LLM(model="gpt-4o-mini"),
tts=elevenlabs.TTS(
voice=elevenlabs.Voice(
id="2zRM7PkgwBPiau2jvVXc", name="Monika Sogam", category="standard"
)
),
chat_ctx=initial_ctx,
fnc_ctx=CallActions(api=ctx.api, participant=participant, room=ctx.room),
interrupt_min_words=3,
)
gray-monkey-83525
02/28/2025, 7:00 AMVoicePipelineAgent
, you need to adjust the Text-to-Speech (TTS) settings. However, the ElevenLabs TTS plugin does not support speed adjustments directly. You might consider using a different TTS provider that allows speed modifications, such as Cartesia, which supports speed adjustments as mentioned in the LiveKit Slack discussion.
Would you like more details on how to switch to a TTS provider that supports speed adjustments?
Sources:
- LiveKit Slack discussion 👍 0 button 👎 0 button 💡 Hint: Mention @runllm in the thread for followups.