[PCcoding666]

Summary

Rewrites docs/VIDEO_SCRIPT.md so the YouTube intro video matches the app as it ships today. The old script made several claims the app doesn't support, including a frosted-glass main window, SFSpeechRecognizer, filler-word removal, a "ten-odd MB" footprint and "download, unzip, double-click". The new script:

• 9 shots, about 3 minutes, each with what's on screen, the Chinese narration, on-screen text and recording notes:
  1. hook
  2. what LCT is
  3. first launch (in-app permissions, hardware-based model recommendation)
  4. dual-lane meeting
  5. live language switch
  6. export
  7. overlay and global hotkeys
  8. offline
  9. requirements and download
• Publish prerequisites: the notarized DMG (blocked on the Apple Developer account), PR fix(macos): handle capture commands while the main window is closed #15, and two behaviours not yet verified in the GUI (offline operation; switching to a language that needs an on-device model download). Each has a fallback if it doesn't pan out.
• Claim ↔ source table: every narration claim is tied to the code or a measurement, for example the 393–984 ms per-line translation latency on an M2 Pro / 16 GB.
• "Don't say" list: corrects the old claims. It records that the 4B model currently sits at about 10 GB in Ollama because of the default 32K context, so the video shouldn't show memory until that's fixed.
• Recording checklist and YouTube title, description, chapters.

Also updates the root README's macOS row, which still listed only SFSpeechRecognizer.

Test plan

• Each shot's UI text checked against the source on master (WelcomeView, MainView, AppDelegate status-bar menu, SettingsView, SessionExporter output).
• Measured numbers taken from the running app and Ollama on 2026-10-10/11.
• Shots 5 (optional line) and 8 need a GUI check before recording; see the prerequisites table.

🤖 Generated with Claude Code

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.