VOCR
VOCR is an OCR and AI-powered screen recognition tool for macOS, designed to help VoiceOver users navigate inaccessible interfaces and understand visual content. It integrates directly with VoiceOver, allowing for seamless navigation of recognized text and AI-driven image analysis.
Demo
OCR
Computer Use
Check out the demo for Computer Use where I put VOCR through a series of different UI tasks.
WARNING: USE AT YOUR OWN RISK!
VOCR is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY, expressed or implied, of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. Please see the GNU General Public License for more details.
Download
Get the latest version here: VOCR v3.0.0-beta.5.
Setup
To ensure VOCR works properly, follow these steps in order:
- Install: Unzip the downloaded file and move the VOCR application to your Applications folder.
- Open VoiceOver Utility and check allow VoiceOver to be controlled with AppleScript in the general category.
- Launch & Grant Permissions: Launch VOCR. A permissions window will appear. Click the action buttons for each row to grant the following:
- Accessibility: Required for OCR and window interaction.
- Screen Recording: Required to capture screen content.
- VoiceOver (Automation): Required for cursor control and speaking results.
- Camera: Optional for capturing images with camera.
- Notifications: Optional for update alerts.
- Note that for each permission, you will be directed to the System Settings window, as required by macOS.
- Screen Configuration:
- Ensure the Screen Curtain is OFF (
VO + Shift + F11). - Hide VoiceOver Visuals (
VO + Command + F11) to prevent them from interfering with scans.
- Ensure the Screen Curtain is OFF (
- Restart & Verify: Once Accessibility and Screen Recording permissions are granted, restart VOCR. Press
Command + Shift + Control + Wto perform your first scan. You should hear a beep followed by “finished.”
Features
OCR Modes
- Scan Window (
Cmd + Shift + Control + W): Recognizes text in the currently focused window. - Scan VO Cursor (
Cmd + Shift + Control + V): Recognizes the specific element under the VoiceOver cursor (useful for video players or social media images). - Real-Time OCR (
Cmd + Shift + Control + R): Continuously monitors the screen and reports new content, such as live subtitles.
AI Integration
VOCR can communicate with any platform compatible with the OpenAI Chat Completion API. Examples include Claude, Gemini, OpenAI, OpenRouter, and local engines such as Ollama and Llama.cpp.
- Ask AI (
Cmd + Shift + Control + A): Ask a question about the last scan or an image file in Finder. - Explore with AI (
Cmd + Shift + Control + E): Analyzes the image to identify and describe different layout areas. - Camera Capture (
Cmd + Shift + Control + C): Take a photo with your webcam and analyze it with AI. - Start/Stop Computer Use (
Cmd + Shift + Control + U): Let AI control apps using mouse and keyboard commands to perform a task. - Pause/Resume Computer Use (
Cmd + Control + P): Only available during computer use. - Toggle Speak Assistant Message: Command+Control+S (Only Available during Computer Use)
To manage models and API keys, go to the VOCR Menu > Presets > Preset Manager.
When the task is completed, the token usage and trace log will be copied to the clipboard.
If you notice Computer Use enters a loop while attempting the same task, either cancel the task or pause and resume it with a different instruction.
Keyboard Shortcuts
Global Shortcuts
| Shortcut | Action |
| :— | :— |
| Cmd + Shift + Control + S | Open VOCR Menu / Settings |
| Cmd + Shift + Control + W | Scan Window |
| Cmd + Shift + Control + V | Scan VoiceOver Cursor |
| Cmd + Shift + Control + A | Ask AI a Question |
| Cmd + Shift + Control + R | Toggle Real-Time OCR |
| Cmd + Shift + Control + E | Explore with AI |
| Cmd + Shift + Control + U | Computer Use |
| Cmd + Shift + Control + C | Camera Capture |
Navigation Shortcuts (Active after a scan)
| Shortcut | Action |
| :— | :— |
| Cmd + Control + Arrows | Move through elements |
| Cmd + Shift + Control + Arrows | Move by character |
| Cmd + Control + L | Report current coordinates |
| Cmd + Control + I | Identify current object with AI |
| Cmd + Control + Shift + Q | View recognized QR codes and barcodes |
| Escape | Exit navigation |
Settings
Access the VOCR Menu with Cmd + Control + Shift + S to customize:
- OCR > Autoscan: Automatically scan after clicking an item.
- OCR > Detect Object: Locate icons and objects without text.
- OCR > Positional Audio: Audio feedback that maps mouse location to sound frequency and panning.
- AI > Use Preset Prompt: Use the prompt from the selected preset without opening the Ask AI dialog.
- AI > Speak Assistant Message: To hear more detail from the assistant during Computer Use.
- Launch on Login: Automatically start VOCR when you log in.
Troubleshooting
- “Nothing Found”: Ensure the Screen Curtain is turned OFF.
- No Speech during Navigation: Ensure the VoiceOver (Automation) permission is granted in the Permissions window.
- Permission Issues: If permissions are not being recognized after a macOS update, use the Reset option in the VOCR menu or run the reset script from the terminal:
cd /path/to/vocr/folder ./reset_permissions.shAlternatively, you can manually reset via terminal:
tccutil reset All com.chikim.VOCR sudo tccutil reset All com.chikim.VOCR - Local AI: If using Ollama, ensure you have pulled a vision model:
ollama pull qwen3.5.
Enjoy using VOCR!