Skip to main content

Install and set up InkSpoke

This page gets InkSpoke onto your computer and through its short first-run setup. You'll download the app, install it like any other desktop program, then walk a six-step wizard that grants the permissions InkSpoke needs and confirms your starting models and workspaces. Budget about five minutes.

1. Download​

Go to inkspoke.app/download and grab the build for your operating system. InkSpoke runs on Windows, macOS, and Linux.

After it's installed, InkSpoke keeps itself current — it checks for a newer stable release on startup and can update in the background, so you rarely need to return to the download page.

2. Install​

Installation is the standard flow for each platform:

PlatformHow to install
WindowsRun the downloaded installer and let it finish, then launch InkSpoke from the Start menu.
macOSOpen the download and move InkSpoke into your Applications folder, then launch it.
LinuxRun the downloaded package for your distribution, then launch InkSpoke.
First launch shows the setup wizard

The very first time you open InkSpoke, it detects that setup hasn't run yet and opens the onboarding wizard automatically. If you ever want to run it again, you can — see Re-run setup later.

3. The first-run wizard (6 steps)​

The wizard is a small, centered window with a step indicator and a Skip / Back / Continue footer. On the last step, Continue becomes Finish.

┌─────────────────────────────────────────────────┐
│  InkSpoke setup                                 │
├─────────────────────────────────────────────────┤
│                                                 │
│   Grant microphone access                       │
│   InkSpoke needs your mic to hear you speak.    │
│                                                 │
│            [ Grant microphone access ]          │
│                                                 │
│   ○ ○ ●  ○ ○ ○     Step 3 of 6                  │
├─────────────────────────────────────────────────┤
│  Skip                        Back   Continue ▸  │
└─────────────────────────────────────────────────┘

Here's the whole path:

Every step is skippable, and any permission you skip can be granted later — so if you're in a hurry, click through and come back.

Step 1 — Welcome​

A quick introduction: the InkSpoke name, the tagline "Speak. We'll write it the way you want," and three things to expect — offline-first speech-to-text, optional AI refinement for chat, email, and notes, and the fact that it works inside any app. Nothing to configure here; click Continue.

Step 2 — Industry​

InkSpoke asks "What kind of work do you do?" and shows a grid of cards (plus an Other… card that opens a free-text box). Your choice tunes your workspaces' vocabulary so transcription and refinement start out closer to your domain's terms. Pick the closest match, or skip if none fit.

Step 3 — Microphone access​

InkSpoke can't hear you without the microphone, so this step has a Grant microphone access button. Approve the OS prompt and the button shows ✓ Granted.

Microphone is required for dictation

If you skip this step, dictation won't work until you allow microphone access. You can grant it any time from your operating system's privacy settings (macOS System Settings → Privacy & Security → Microphone; Windows Settings → Privacy → Microphone) — no reinstall needed.

Step 4 — Selected-text capture​

This permission lets InkSpoke read the text you've highlighted in another app, which powers Command Mode — select a sentence, speak an instruction like "make this formal," and InkSpoke rewrites it in place. What it requests depends on your OS:

PlatformWhat InkSpoke usesWhat you grant
macOSAccessibilityClick Grant accessibility access and enable InkSpoke under System Settings → Privacy & Security → Accessibility.
WindowsUI AutomationNothing extra — it works without a separate permission.
LinuxClipboard-based captureNo separate system prompt.

This step is optional. Plain dictation works without it; you only need it for Command Mode and other selected-text features.

Step 5 — Choose your default model​

This step is informational — it shows the setup InkSpoke ships with so you know what you're starting from. No selection happens here.

PurposeDefaultWhere it runs
Speech-to-textWhisper SmallOn your device, offline, free
AI refinementPlatform AIIn the cloud, during your Pro trial

So out of the box, your speech is transcribed entirely on-device (nothing leaves your computer), while AI refinement uses InkSpoke's built-in Platform models as part of your Pro trial. You can change either later — swap to a larger on-device model, or bring your own provider key — from AI Models in Settings.

What's the Pro trial?

New installs get a Pro trial with an audio allowance rather than a fixed calendar countdown. Once you've dictated for a while you may hit the allowance; see The free trial for the details.

Step 6 — Ready​

The final screen confirms "You're all set" and lists the starter workspaces InkSpoke seeds for you:

🎙  Speech / Voice     💬  Chat
✉   Email              📝  Note Taking

Each workspace carries its own vocabulary and tone, and InkSpoke can switch between them automatically based on the app you're writing in. Click Finish to close the wizard.

Re-run setup later​

Changed your mind, or skipped a step you now want? You can run the whole wizard again from Settings → Configuration → General → Rerun Onboarding.

Power users

You don't have to touch the wizard to change anything it set. Permissions live in your OS privacy settings, your default models live under AI Models, and every hotkey and preference lives in Settings → Configuration. Re-running onboarding is just the guided path, not the only one.

You're ready to dictate​

With microphone access granted, you're set. The default activation hotkey is Alt + Space on Windows and Linux, or ⌥ + Space on macOS — press it, speak, and press it again to see your first words appear.

Next steps​