Dictate your prompts and long context; text lands at the terminal cursor.
System-level input — text lands wherever the cursor is · zero integration
No plugin needed: Gemini CLI's generous free tier pairs perfectly with spoken input — saying an idea is faster than typing it, and transcription stays on-device.
| Dimension | Spoketype | Notes |
|---|---|---|
| Integration | No plugin — works out of the box | Text lands wherever the cursor is |
| Works in | Terminals, editors, any input field | No per-app setup |
| Transcription runs | On this computer | Audio never leaves the device |
| Hotkey | Customizable | Default ⌥Space (Mac) / Ctrl+Alt+Space (Windows) |
| Price | Free | No account, no purchases |
Gemini CLI is a Google product; this page describes using Spoketype alongside it and is not an official integration.
Gemini CLI's free tier is generous, so the real cost of experimentation becomes typing prompts and pasting context. Spoketype removes that too: speak the idea, and it lands at the cursor.
One — dictate long context: paste file paths, speak the explanation, combine both. Two — iterate prompt variants quickly; revise aloud, once per version. Three — give multi-step instructions in one breath and let the agent run.
Anyone using Gemini CLI as a daily lab, reworking prompts at high frequency.
macOS & Windows · No plugin · Audio never leaves
Related: Claude Code · Codex·OpenCode