asrbox

本地优先的音视频转写与字幕工作台：25 个本地 ASR 模型（Whisper、Faster Whisper、FunASR、Qwen3-ASR、Granite Speech、Cohere Transcribe、ARK-ASR、Voxtral、FireRedASR2、MLX）、macOS/Windows 桌面端与 Docker/Web 部署、字幕编辑导出及 LLM 字幕核对。 Local-first transcription workspace with 25 local ASR models, macOS/Windows desktop apps, Docker/Web deployment, and LLM subtitle proofreading.

GitHub

Official source

Start locally with the CLI guide