Dictate a prompt: a waveform from your real mic level, a timer, then transcribing.
pnpm dlx shadcn@latest add "https://hairlineui.com/r/voice-input.json"