Quick Start Guide

Get up and running with AICHE voice-to-text in under 2 minutes. Works on Mac, Windows, Linux, iPhone, iPad, Apple Watch, Android, Chrome, and Obsidian.

4 min read
Last updated: April 7, 2026
installationsetupbasics

AICHE turns natural speech into clean, polished text at the cursor and gives you an intelligent voice workspace. One key to record, speak naturally, get text inserted, pasted, copied, or saved depending on platform. Here's how to get up and running in under 2 minutes.

The 3-step AICHE loop: Activate, Speak Naturally, Smart Insert and Sync

1. Pick Your Platform & Install

AICHE runs on 9 platforms. Download the appropriate package from our Downloads Page, which automatically detects your operating system.

Desktop (Mac, Windows, Linux)

  1. Download from aiche.app/profile?section=downloads
  2. Install:
    • macOS: Open the DMG and drag to Applications, or run brew install --cask aiche-desktop
    • Windows: Run the installer executable, or install via winget
    • Linux: Choose your preferred method: .deb (Debian/Ubuntu/Mint with APT repo), .rpm (Fedora/RHEL), AppImage, Flatpak, or Snap (featured by Canonical in the Ubuntu App Center with one-click install)
  3. Launch AICHE and click Sign In (Google, Apple, GitHub, or email)
  4. Grant microphone permission when prompted

Mobile & Wearables (iPhone, iPad, Android, Apple Watch)

  1. Download from the App Store (iOS/iPadOS) or Google Play Store (Android)
  2. Complete the 30-second onboarding and sign in
  3. Grant microphone permission
  4. Apple Watch: After installing on iPhone, open the Watch app on your phone and install AICHE from Available Apps

Extensions & Integrations

  • Chrome Extension: Install from the Chrome Web Store (also works in Brave, Edge, and Arc). Click the extension icon to sign in.
  • Obsidian Plugin: Install via Obsidian Community Plugins to transcribe thoughts and meeting notes directly into your vault.
  • Public REST API (Pro): POST audio programmatically from Python scripts, shell scripts, or CI/CD pipelines.

2. Record Your First Note

On Desktop

  1. Open any application where you type—editor, browser, Slack, terminal, email
  2. Place your cursor where you want text to appear
  3. Press the global recording hotkey:
    • Mac: ⌃ + ⌥ + R (Control + Option + R)
    • Windows & Linux: Ctrl + Alt + R
  4. Speak naturally. Do not dictate punctuation marks like "comma" or "period"
  5. Press the hotkey again to stop
  6. Clean, punctuated text appears directly at your cursor

On Mobile

  1. Open the app and tap the record button
  2. Speak your thoughts
  3. Tap again to stop
  4. Clean text appears in your feed, formatted and copied to clipboard

On Chrome Extension

  1. Press Alt + V (or click the extension icon)
  2. Speak your input — an on-page floating pill tracks your recording
  3. Press Alt + V again to stop
  4. Text auto-pastes directly into the active web form or chat box (with an instant Undo button on the pill)
  5. Press Alt + H anytime to open your full notes history in the Chrome side panel

3. Mobile Superpowers: Capture On The Go

AICHE is engineered to make mobile capture frictionless and dependable outside your desk:

  • Android Home-Screen Quick-Record Widget: Start and save voice notes directly from your home screen with a single tap, without launching the app.
  • Apple Watch Standalone Dictation: Tap your wrist and speak—no duration caps. Transcribes on your watch and syncs via WatchConnectivity so ideas land everywhere you work.
  • Offline Encrypted Queue: If you are in transit, on a flight, or in a subway with no network, record anyway. Audio is securely stored in a local AES-256 encrypted queue and automatically processes and syncs the moment your connection returns.
  • 28 Localized Languages: Full multilingual mobile interface support across iOS and Android.

4. The AICHE v2 Workspace

Beyond cursor dictation, AICHE v2 brings a complete suite of voice-first knowledge and document capabilities:

AICHE v2 workspace capabilities map: 5 core workflows across cloud and on-device models

1. Smart Insert & System Dictation

Transcribes at 400 WPM directly into your active window. AI cleanup strips filler words (um, uh, like), normalizes stutters, and adds punctuation and paragraphs based on your natural cadence.

2. Voice Code for AI Coding Agents (Pro)

Designed specifically for developers using Claude Code, Cursor, Codex, and Antigravity. Toggle with Ctrl + Option + . (Mac) or Ctrl + Alt + . (Windows/Linux). Includes:

  • Pause-Aware Auto-Send: Stop speaking for a beat and your prompt automatically dispatches to the agent—no Enter key or mouse required.
  • Spoken Confirmations: Approve or reject agent shell actions by voice ("approve" / "reject").
  • Software Development Profile: Recognizes CLI flags, regex, snake_case, and library identifiers.

3. Meetings & Calls

Capture multi-person discussions or online VoIP meetings (loopback system audio for Zoom, Google Meet, Microsoft Teams). Automatically separates speakers with diarization labels, extracts key decisions, and generates actionable task summaries with owners and deadlines.

4. Write & Spoken Editing

Draft documents from scratch or open existing notes to revise them by voice. Switch to Edit mode, select text, and speak editing instructions like "turn this into a short list" or "summarize key tradeoffs".

5. Ask Your Library (Grounded Q&A)

Query across all your historical notes and meeting transcripts by voice or text. Ask questions like "What did we decide about the migration deadline?" and receive answers grounded with clickable source passage chips linking back to the exact conversation.

(Note: Advanced v2 modules—Meetings, Write, Ask, and On-Device models—debuted on our Linux v2 desktop flagship and are actively rolling out across macOS, Windows, and mobile).


5. Privacy & On-Device AI

  • Zero-Retention Cloud Processing: Transcribed audio is processed and discarded immediately. No audio is ever stored or used for model training.
  • End-to-End Encrypted Cloud Sync: Notes are encrypted on-device with AES-256-GCM and an Argon2id passphrase key before syncing. Only you hold the decryption key.
  • On-Device Local AI (Pro): Run dictation, speaker diarization, and summaries locally on your machine without an internet connection (quantized CPU models, no dedicated GPU required).

6. Plans Snapshot

Plan Price (Annual) Devices Core Capabilities
Free Trial 7 days free 10 Unlimited dictation, full access, no credit card required
Personal $3.99/mo ($47.99/yr) 3 Unlimited voice notes, system dictation, meeting summaries, 99 languages
Pro $8.33/mo ($99.99/yr) 10 Everything in Personal + Voice Code, Software Dev profile, REST API, call loopback, on-device AI
Team $9.99/mo per seat 10 / seat Admin seats, roles, shared custom vocabulary, unified billing

Start your free trial at aiche.app/pricing.


7. Next Steps & Resources