[
    {
        "role": "user",
        "content": "how%20do%20i%20scan%20printed%20text%20directly%20from%20JAWS%3F"
    },
    {
        "role": "assistant",
        "content": "Yes \u2014 JAWS has a built-in feature called *Convenient OCR* that can scan printed text (via a scanner or camera) or images and convert it into readable text. Here\u2019s how you can set it up and use it, along with tips to help it work well.\n\n---\n\n## What You\u2019ll Need\n\n- A **TWAIN-compatible** flatbed scanner *or* a camera device (like the Freedom Scientific Pearl camera) connected to your computer. ([sastltd.zohodesk.eu](https://sastltd.zohodesk.eu/portal/en/kb/articles/jaws-set-the-scanning-source-for-convenient-ocr?utm_source=openai))  \n- JAWS installed and running with Convenient OCR enabled.  \n\n---\n\n## Turning On and Configuring the Scanner\n\n1. Open the **Utilities** menu in JAWS (JAWS key + J).  \n2. Navigate to the **Convenient OCR** section. ([sastltd.zohodesk.eu](https://sastltd.zohodesk.eu/portal/en/kb/articles/jaws-set-the-scanning-source-for-convenient-ocr?utm_source=openai))  \n3. Select **Cameras and Scanners**.  \n4. Under that, pick the connected scanner or camera device (look for a TWAIN source). Use the arrow keys and press ENTER to confirm. ([sastltd.zohodesk.eu](https://sastltd.zohodesk.eu/portal/en/kb/articles/jaws-set-the-scanning-source-for-convenient-ocr?utm_source=openai))  \n\n---\n\n## How to Scan Printed Text\n\nHere are the commands (hotkeys) to scan, depending on what you want to capture. Start by pressing **JAWS key + Spacebar**, then **O** to enter the OCR layer, and then a second letter to pick what to scan. ([sastltd.zohodesk.eu](https://sastltd.zohodesk.eu/portal/en/kb/articles/jaws-convenient-ocr-layered-hotkeys?utm_source=openai))\n\n| Task | Key Combination |\n|---|---|\n| Scan document using camera or scanner | INSERT+SPACE, O, A |\n| Recognize an image file (e.g. JPG, PNG) | INSERT+SPACE, O, F |\n| OCR an open PDF document | INSERT+SPACE, O, D |\n| Recognize selected image control (e.g. on web or app) | INSERT+SPACE, O, C |\n| OCR current application window | INSERT+SPACE, O, W |\n| OCR the whole screen | INSERT+SPACE, O, S |\n\nAfter initiating the OCR, JAWS shows the recognized text in the **Results Viewer**, where you can read using your usual reading or navigation keystrokes. Press ESC to close the viewer. ([tdx.umsystem.edu](https://tdx.umsystem.edu/TDClient/36/DoIT/KB/PrintArticle?ID=1197&utm_source=openai))\n\n---\n\n## Reading Printed Text from Paper\n\nTo scan printed documents (books, magazines, etc.):\n\n1. Place the printed page face-down on the flatbed scanner (or face-up under the Pearl camera).  \n2. Use **INSERT+SPACE**, **O**, **A** to perform OCR of the paper. JAWS will capture the image via the scanner/camera and then run OCR on it. ([sastltd.zohodesk.eu](https://sastltd.zohodesk.eu/portal/en/kb/articles/jaws-convenient-ocr-layered-hotkeys?utm_source=openai))  \n\n---\n\n## Working with PDFs That Are Just Images\n\nSometimes a PDF isn\u2019t tagged and is just a scanned image \u2014 JAWS can\u2019t read it normally. In that case, there are two options:\n\n- When you open such a PDF, JAWS often detects that it's empty or inaccessible and prompts: \u201cUse Convenient OCR?\u201d Then you can choose **yes**. ([aph.org](https://www.aph.org/app/uploads/2021/06/Prevent-Document-Frustration-JAWS-and-PDFs.pdf?utm_source=openai))  \n- If you miss the prompt, you can press **INSERT+SPACE**, **O**, **D** to manually run OCR on the PDF. It may take a few seconds depending on the size of the document. ([sensorysolutions.co.za](https://sensorysolutions.co.za/wp-content/uploads/2022/11/Using-PEARL-with-JAWS.pdf?utm_source=openai))  \n\nAfter processing, text appears in the Results Viewer \u2014 you can navigate, search, and copy it. ([sensorysolutions.co.za](https://sensorysolutions.co.za/wp-content/uploads/2022/11/Using-PEARL-with-JAWS.pdf?utm_source=openai))  \n\n---\n\n## Tips for Best Results\n\n- Use decent lighting for camera scans and make sure the page is flat.  \n- Text should be at least about **8 point** or larger and clearly printed (plain fonts work well). Very small fonts or fancy/script fonts are harder. ([tdx.umsystem.edu](https://tdx.umsystem.edu/TDClient/36/DoIT/KB/PrintArticle?ID=1197&utm_source=openai))  \n- Simple layouts work best\u2014columns, tables, graphics can confuse the OCR.  \n- Be patient: large files or scans with lots of pages may take more time. ([sensorysolutions.co.za](https://sensorysolutions.co.za/wp-content/uploads/2022/11/Using-PEARL-with-JAWS.pdf?utm_source=openai))  \n\n---\n\nIf you want, I can walk you through setting this up with your specific scanner or show you how to use these features step-by-step from scratch \u2014 let me know what device you're using."
    }
]