Podcast transcript generator

This build does not bundle a speech-to-text model, so it does not create words from audio. What it does locally: it reads your episode file, plans the Whisper-style processing chunks, turns a transcript you already have (from any transcription service or your own notes) into a clean cue list, plays each cue against the local file, and exports TXT, SRT, or VTT. Nothing uploads at any step. Processing speed and model size only apply to an actual transcription run, which is not part of this tool.