The quickest way to get an answer out of Gemini is to ask for it out loud. Set Clavio's wake word to “Gemini” and just say “Gemini, compare these two pricing models and tell me which one scales better.” Your question lands in the prompt and sends itself the moment you stop — nothing to tap, hold, or click. Keep Clavio on Fast and Gemini reads your question exactly as you asked it, at the speed of speech. Then, when the answer sends you off to write the doc or the email you were researching for, the same voice writes that too.
Say its name: set the wake word to “Gemini”
Clavio lets you choose a wake word for each app, so for Gemini you set it to the assistant's own name — “Gemini.” Now you're not launching a dictation tool; you're talking to the assistant. You say “Gemini, summarize this thread in five bullets,” it answers to its name, types your question into the prompt, and — with auto-send on — sends it the instant you stop. No key to hold, no mic button to tap, no reaching for the trackpad — just your voice.
Fast mode: your question, word for word
A prompt to Gemini isn't a letter — it shouldn't be smoothed over or reworded before it's sent. So keep Clavio on Fast: it types your words exactly as you said them, with no rewriting and no processing pause, the second you stop. That's what a good question needs — Gemini reads precisely what you asked, not a tidied-up paraphrase of it. Add the names you actually use — products, people, acronyms, the topic you're digging into — to Clavio's dictionary, and “BigQuery,” “Gemini 2.5,” or a client's name come through clean every time instead of spelled by guess.
Think out loud: long questions and fast follow-ups
A session with Gemini is a back-and-forth — a long, half-formed question, then a follow-up, then another. That's exactly the kind of thing that's slow to type and quick to say. Dictation lets you ask the big, messy question the way it's actually in your head — “okay, and what if we did it the other way round, would that change the migration order?” — without trimming it down to something easier to type. You keep the momentum of a real conversation: speak the thought, read the answer, speak the next one. With auto-send on, each turn fires the moment you finish, so the exchange moves at talking speed instead of typing speed.
One voice for the whole loop — the doc, the email, the reply
The answer is rarely the end of it. You asked Gemini to research something because you had to write something — so when the answer's in, dictate the Google Doc straight from the same voice. Draft the Gmail the research was for, and here you may want it polished, so Clavio switches to a natural tone for that window while Gemini stays on Fast. Someone replies and you fire back in a casual voice. Clavio remembers the tone for each app, so one wake word carries you from the question all the way to the finished writing — and since it only answers to your voice, a coworker saying “Gemini” nearby never sets it off. That's the whole point: you research and you write, and all of it is just talking.
Recommended Clavio settings for Gemini
| Setting | Recommended | Why |
|---|---|---|
| Wake word (per app) | Make it “Gemini” | Speak to the assistant by name — “Gemini, summarize this” — instead of a generic trigger, with no key or mic button to tap. |
| Polish level | Fast (off) | Types your question instantly, exactly as asked. No rewriting, no processing pause — Gemini reads your literal words. |
| Auto-send | On | Fully hands-free: the moment you stop talking, the prompt sends itself, and each follow-up fires at talking speed. |
| Personal dictionary | Add your terms | Keeps product names, people, and acronyms in your questions — Gemini 2.5, BigQuery, a client's name — coming through clean. |
| Tone for the follow-through | A natural polish | When you move from prompting Gemini to writing the Google Doc or Gmail it was for, Clavio switches to a tidy tone, per app. |
Prompt Gemini hands-free with Clavio
Clavio is an AI dictation app for Mac. It sits in your menu bar and types finished text wherever your cursor is — the Gemini prompt in your browser or the app, your Google Docs, Gmail, Slack. You set a wake word per app (say “Gemini” to prompt Gemini), and choose the polish per app too: Fast with no cleanup so your question reaches the model exactly as asked, or a natural tone for the doc or email you write next. A personal dictionary keeps your product and people names intact, and an on-device voiceprint means only your voice sets it off. Free to try: 3,000 words a month, no card, macOS 14+ on Apple Silicon. Pro is £12/month for unlimited.
Keep reading
Common questions
Doesn't Gemini already have a microphone button?
It does — you tap the mic in the Gemini box, speak one question, and it only works inside Gemini. Clavio is different in two ways that matter day to day: you give it the wake word “Gemini,” so there's nothing to tap, and it works everywhere your work lives, not just the prompt — the Google Doc and the Gmail draft your question was for. That reach, and never reaching for the mic, is why people use it.
Can I prompt Gemini fully hands-free?
Yes. Give Gemini the wake word “Gemini,” turn auto-send on, and just say “Gemini, do this” — your question is typed and sent the moment you stop talking. No key, no mic button — just your voice, including every follow-up.
Will it reword my prompts or mangle names?
Not on Fast — it types your words exactly, no rewriting, so Gemini reads the question you actually asked. To keep product names, people, and acronyms precise, add them to the dictionary. By default Fast hands the model your exact wording.
Does it work with Gemini in the browser and the app?
Yes. Clavio types wherever your cursor is, so it doesn't care whether Gemini is a browser tab at gemini.google.com or an app window — the wake word and auto-send work the same in both. The same setup follows you to any other app on the Mac too.
Can I use the same voice for the doc or email afterward?
Yes — that's the main reason to use a system-wide app instead of the mic in the prompt box. Clavio types into Google Docs, Gmail, and Slack the same way it types into the Gemini prompt, and remembers the tone per app: exact for the question, tidy for the doc, casual for a chat reply. One wake word covers research and writing both.