Skip to main content
Start with what you want to build. Each product guide owns its setup and options; the links below take you directly to the relevant section. New to the API? Start with the STT and TTS quickstart.

Agents through UI or API

Build a voice agent in the console or create its profile with the API. Both edit the same saved agent, whose conversations run on Omni.

API primitives

Use a focused API when your application already owns the rest of the workflow.

Dub an existing recording

Choose Dub for a saved audio or video file, and Live Translation for a live two-person conversation. These products have separate language rosters and connection requirements.

Call intelligence and answer detection

Use the call intelligence overview to choose between transcripts, summaries, custom extracted fields, and compliance results.

Framework adapters

Keep your framework when it should own the conversation, tools, and LLM: LiveKit Agents, Pipecat, or Vercel AI SDK. PyAI supplies STT and TTS in those integrations.

Next

Authentication covers keys and scopes. SDKs covers Python and TypeScript clients. Integrations covers imports and connected apps. Use Errors and limits for shared failures and Production readiness before moving customer traffic.