WhisperLiveKit System Requirements (CPU, RAM, Disk)

WhisperLiveKit is a self-hosted, real-time speech-to-text engine that runs fully on your own hardware. It streams audio from your microphone or any source and transcribes it live with low latency, identifies who is speaking, and can even translate across languages on the fly. It exposes OpenAI- and Deepgram-compatible REST endpoints, so it drops into existing voice applications without code changes. No cloud dependency, no per-minute fees — your audio never leaves your machine.