start(1)

📦 version 1.1.1, in nixpkgs unstable

👑 START(1) — the girlies' guide to start

Okay bestie, so this is a page from the General Commands Manual — a whole manual, and this one little page is all about start. Buckle up, it's short but iconic. 💅

🌸 NAME

start — start recording (send SIGUSR1 to daemon) ✨

🦋 SYNOPSIS

start [--type] [--clipboard] [--paste] [--file] [--model] [--profile]
[--auto-submit] [--no-auto-submit] [--shift-enter-newlines]
[--no-shift-enter-newlines] [--no-osd] [--smart-auto-submit]
[--no-smart-auto-submit] [-h|--help]

💖 DESCRIPTION

Start recording (send SIGUSR1 to daemon). That's it, that's the whole vibe — you poke the daemon, it starts recording. No cap. 🎀

✨ OPTIONS

  • --type — override the output mode so it simulates keyboard typing ⌨️
  • --clipboard — override the output mode so it goes to the clipboard only 📋
  • --paste — override the output mode to paste (clipboard + Ctrl+V) 🫶
  • --file [<FILE>] — write the transcription to a file. Use --file all on its lonesome to use file_path from config, or --file=path.txt when you want to name an explicit path, iykyk 🗂️
  • --model <MODEL> — use a specific model for this transcription (e.g., large-v3-turbo) 👑

Fr fr any engine's model works here, not just the configured engine's: the daemon loads it alongside the default model and keeps it loaded for the next override (unless on_demand_loading is set). The names are the ones voxtype info models shows; an unknown or undownloaded model fails immediately — no gentle letdown, just instant fail. 💀

  • --profile <NAME> — use a named profile for post-processing (e.g., --profile slack). Profiles are defined in config.toml under [profiles.name] 🎀
  • --auto-submit — auto-submit (press Enter) after this transcription ⏎
  • --no-auto-submit — disable auto-submit for this transcription (overrides config) 🙈
  • --shift-enter-newlines — use Shift+Enter for newlines in this transcription 🌸
  • --no-shift-enter-newlines — disable Shift+Enter newlines for this transcription (overrides config) 🚫
  • --no-osd — suppress the on-screen display for this recording only 🫥

It leaves osd.enabled in config.toml untouched — like, completely untouched, no notes. Intended for external tools that draw their own dictation UI and do not want voxtype's overlay on top of it. Respect the other main characters. 💜

  • --smart-auto-submit — enable smart auto-submit for this recording (say "submit" to press Enter) 🗣️
  • --no-smart-auto-submit — disable smart auto-submit for this recording 🔇
  • -h, --help — print help (see a summary with -h) 💖
Heads up bestie: -h gives you the summary version of the help. That's not a bug, that's just how she is. 🌈

And that's the whole page, mwah. 💋 Signing off from the General Commands Manual, page start(1) — same name up top, same name at the bottom, symmetrical queen that she is. 👑✨

📦 where this page comes from