Runtime Speech Recognizer is an Unreal Engine plugin that brings offline, cross-platform, multilingual speech recognition to your projects, powered by OpenAI’s Whisper. It supports 95+ languages, runs entirely on-device with no internet required, and works on all platforms supported by Unreal Engine (UE 4.27 and newer).
The video below walks through the setup process and demonstrates three common workflows, including voice-activated command recognition with Silero VAD for noisy environments.
Links:
Fab: Runtime Speech Recognizer (Real-Time, Offline, AI) | Fab
Documentation: Overview | Georgy Dev Docs
Windows demo: Redirect
Discord: Redirect
https://dev.epicgames.com/community/learning/tutorials/M4np/offline-speech-recognition-in-unreal-engine-with-runtime-speech-recognizer-openai-whisper