homelab-00/TranscriptionSuite ? reverse-engineered prompt

Build me a fully local speech to text app that runs on Windows, macOS, and Linux. I want a desktop dashboard where I can record from my mic or system audio, import audio or video files, and get accurate transcripts quickly without my audio leaving my machine. It should support multiple transcription backends, including Whisper style models, and handle speaker labels so I can see who said what.

Please make it work for long recordings, live dictation, and file import, with easy export to plain text and subtitle files. I’d also like a notebook view with a calendar style history, full text search, and a simple way to chat with my notes through an OpenAI compatible provider. Add global keyboard shortcuts for starting and stopping recording and pasting text at the cursor. If it makes sense, include a way to run the backend locally, with Docker on most systems and native support where possible. Look up current docs online if you need to.