Subgen wraps OpenAI’s Whisper automatic speech recognition model to transcribe video audio into subtitle files. It supports integration with media servers like Jellyfin, Plex, and Emby as triggers, along with Bazarr and Tautulli. The tool offers GPU acceleration via CUDA for NVIDIA and ROCm for AMD GPUs, configurable model selection, and outputs SRT subtitle format. It runs as a containerized service with environment variable configuration for model management and timeout controls.