BocchiTheTank131/AudioVideoTranscripter ? reverse-engineered prompt

Build me a Windows desktop app for transcribing audio and video files locally, with no cloud upload, no account, and no telemetry. I want to drop in a file, pick a speech model, and get a transcript, subtitles, keywords, and speaker info saved on my computer. It should handle single files and batch jobs, remember settings in projects, and let me reopen a project later with edits intact. Please include a simple preview player with seek, volume, subtitle overlays, and clickable timestamps in the transcript. I also want search, replace, export to TXT, SRT, and VTT, plus a way to save and reopen project files.

Make it reliable for long recordings, with speaker and confidence warnings, optional cleanup for obvious bad transcriptions, and voice activity detection turned on by default if the local model is installed. Keep everything offline once models are downloaded, and make sure the app is easy to run on Windows first. If you need to check the current docs for any library details, look them up online.