TL;DR: A voice first workflow means your voice is the default input and the keyboard is the fallback, not the other way around. With Infina you hold Option to dictate into any Mac app, hold fn to speak a command, and when you want to go all the way, double-tap Cmd for hands-free mode: sit back, eat lunch, and from across the room just say "type" plus your words, "send" to press Enter, "open Cursor" to switch apps, and repeat. No other dictation app completes that prompt, send, and switch-apps loop hands-free in plain English. Infina is $99 once (as of July 2026), on-device by default, with a 7-day refund.
This guide is mostly mechanics: which keys, which words, and what a full day looks like when you stop typing first.
What a voice first workflow actually is
Most people who own a dictation app still type 95 percent of their day. They dictate when they remember to, which is rarely, because reaching for a special mode is friction.
A voice first workflow flips the default. Every email, message, prompt, and note starts as speech, and the keyboard is reserved for the two things it is genuinely best at: editing and precision work.
The payoff is speed you can feel. Speaking is about three times faster than typing, and the gap compounds across a day of communication; the numbers are in dictation vs typing speed.
The blocker was never the idea, it was the tooling. If dictating requires a hotkey hold, an Enter press, and a Cmd-Tab for every message, your hands never actually leave the keyboard, so the habit never sticks.
The toolkit: three controls to learn
Infina's entire control surface is three gestures. Learn them once and they cover the whole day.
Hold Option: push-to-talk dictation anywhere
Hold the Option key, speak, release. Your words are typed into whatever app is focused: Mail, Slack, Notes, a browser, a terminal.
Transcription runs entirely on your Mac, on the Apple Neural Engine, so it is fast, works offline, and your audio never leaves your device. This is the gesture you will use fifty times a day, and getting fluent with it is the subject of how to voice type faster.
Hold fn: voice command
Hold fn and tell your Mac to do something instead of type something: open apps, switch tabs, real OS-level actions. Dictation puts words on screen; commands move the machine.
Double-tap Cmd: hands-free mode
Double-tap Cmd and Infina starts listening without any key held. It ships off by default and is labeled experimental, but it is the mode that makes the workflow genuinely voice first, because it removes the last keyboard touches.
The hands-free loop is three spoken words:
- Say "type" plus your words: "type can we move the standup to two". Infina types it into the focused app. No hotkey; a sentence starting with "type" is itself the trigger.
- Say "send": Infina presses Enter.
- Say "open Slack" (or "open Notes", "open Claude Code"): Infina switches apps. Then loop.
Prompt, send, switch, from a few feet away, hands around a sandwich. Dictation apps still make you touch the keyboard to trigger and to send; Infina completes the whole loop by voice. The deep dive on why this loop matters is hands-free voice prompting.
A voice first workflow, morning to evening
Here is what the three controls look like stitched into an actual day.
Morning: email by voice
Open Mail, click into a reply, hold Option, and talk. A five-paragraph reply is about ninety seconds of speaking.
Triage becomes a rhythm: read with your eyes, answer with your voice, archive, next. The keyboard only comes out to fix a name or tighten a sentence.
Midday: messages and quick replies
Slack and iMessage are where voice first pays fastest, because messages are short, conversational, and constant. Hold Option, say the reply, release.
For lunch at your desk, double-tap Cmd instead. "Type sounds good, shipping the fix after lunch", then "send", and your hands never left the plate.
Afternoon: AI prompts by voice
If you prompt Claude Code, Cursor, or ChatGPT all day, this is where the workflow stops being a convenience and becomes an edge. Good prompts are long, and long prompts are exactly what voice is best at.
Hold Option and describe what you want in full sentences, with all the context you would normally be too lazy to type. Or go hands-free and run agents from across the room: "type now add retry logic to the upload path", "send", "open Cursor", repeat. Keeping several agents busy by voice is its own guide: hands-free Claude Code.
Evening: notes and capture
Ideas do not arrive while you are seated in typing position. With hands-free on, "open Notes" then "type" plus the thought captures it while you pace, stretch, or tidy the desk.
End the day the way you started it: eyes on the screen for review, voice for everything that needs words.
Honest limits
- English only in the base product. The optional $10/month cloud add-on brings more languages.
- Raw output by design. Base Infina types what you said with fast on-device formatting, which is exactly right for prompts, messages, and notes. If you want polished punctuation and grammar everywhere, the same $10/month add-on (7-day free trial) uses large language models via our cloud AI providers (Together AI and Groq), which beats the $15/month subscription apps at their own game: you own the app, and you rent the polish only if you want it.
- Mac only, Apple Silicon required for the on-device models.
- Hands-free is experimental and off by default. Push-to-talk is the mature core, and a voice first day works on it alone.
- No free trial. There is a 7-day no-questions money-back guarantee instead, which is a real trial with one extra email.
FAQ
What is a voice first workflow? A way of working where speaking is the default input for emails, messages, prompts, and notes, and the keyboard is reserved for editing and precision work. It works because speaking is about three times faster than typing for most people.
Do I need to go fully hands-free to work voice first? No. Most of the day runs on hold-Option push-to-talk dictation. Hands-free mode (double-tap Cmd, then "type", "send", "open [app]") is the layer you add when you want zero keyboard touches, and it ships off by default.
Does this work in every Mac app? Yes. Infina types at the OS level into whatever app is focused: Mail, Slack, browsers, terminals, editors. No per-app plugin is needed.
Is my audio sent to the cloud? Not by default. Transcription runs on your Mac on the Apple Neural Engine and works offline, and by default no audio or transcripts are stored. Cloud processing exists only as an optional paid add-on.
What Mac do I need? A Mac with Apple Silicon (M-series) for the on-device models. The base product is English-only; the cloud add-on adds more languages.
How much does Infina cost? $99 one-time as of July 2026, with every 1.x update included and a 7-day no-questions-asked money-back guarantee. No subscription; the optional cloud add-on is $10/month with its own 7-day trial. Details on pricing.
The bottom line
A voice first workflow is not about dictating the occasional email. It is about moving your default input from 40-something words a minute to talking speed, all day, in every app.
Infina is built to be that default: hold Option to dictate anywhere, hold fn to command your Mac, double-tap Cmd to drop the keyboard entirely and run the type, send, open loop by voice. No other dictation app closes that loop hands-free in plain English.
One $99 purchase (as of July 2026), on-device by default, 7-day refund. Set it up in the morning, and by the evening walkthrough above you will know whether you ever want to go back.