Skip to main content
The full REST surface documented across this API Reference section — discovery endpoints, TTS synthesis, STT transcription, audio denoising, and the Call Center API — is also published as a single OpenAPI 3.1 document:
Use it to:
  • Generate a typed client in any language (openapi-generator, openapi-typescript, etc.)
  • Point an OpenAPI-aware MCP or agent tool at Lokutor’s REST API directly
  • Validate requests/responses against a schema in CI
This spec covers the HTTP surface only. The real-time voice-agent, TTS, and STT protocols run over WebSocket, which OpenAPI doesn’t model well — those are documented as prose + message tables at WebSocket Protocol, Real-time Streaming, and Voice Chat instead.
If you’re integrating an AI agent or LLM against Lokutor rather than generating a client, see llms-full.txt — a single flat-text file with the complete documentation, including the WebSocket wire protocol verbatim, sized for a one-shot fetch into a context window.

​
cURL