Gradium TTS Is on LiveKit Inference, Free Through October 9

Gradium text-to-speech is now a LiveKit Inference provider, and voice cloning for your voice agents runs through the LiveKit Cloud dashboard. You select Gradium with one model string, `gradium/default`, using the LiveKit API key you already have. There's no Gradium account and no second API key to manage. Sign up for LiveKit Cloud: https://cloud.livekit.io/signup?utm_source=youtube&utm_medium=video&utm_campaign=devrel&utm_content=NpGSC4fVIoA This walkthrough covers what Gradium does with the hard cases voice agents hit constantly (phone numbers, dates, alphanumeric codes, and URLs), how to create a voice clone in the LiveKit Cloud dashboard from about ten seconds of audio, and how to point your agent at either a flagship Gradium voice or your own clone. Voice cloning creates a synthetic voice model from a short reference recording, then generates new speech in that voice. With Gradium on LiveKit Inference, about 10 seconds of reference audio is enough to create a usable clone. Gradium TTS is free on LiveKit Inference through October 9, 2026. The code shown is Python, running against LiveKit Inference with a low latency text to speech path. 📚 Resources 📚 - Agent docs: https://docs.livekit.io/agents/?utm_source=youtube&utm_medium=video&utm_campaign=devrel&utm_content=NpGSC4fVIoA - Gradium TTS on LiveKit Inference: https://docs.livekit.io/agents/models/tts/gradium/?utm_source=youtube&utm_medium=video&utm_campaign=devrel&utm_content=NpGSC4fVIoA - Custom voices and voice cloning: https://docs.livekit.io/agents/models/tts/custom-voices/?utm_source=youtube&utm_medium=video&utm_campaign=devrel&utm_content=NpGSC4fVIoA 🤝 Join the Community: https://community.livekit.io/?utm_source=youtube&utm_medium=video&utm_campaign=devrel&utm_content=NpGSC4fVIoA #livekit #ai #voiceai