Can screen readers read JPEGs
No, screen readers can't read text inside a JPEG. That's not how it works. JPEGs store pictures as pixels - little colored dots - not as actual text characters. So when you have JAWS, NVDA, or VoiceOver trying to parse a JPEG, they're basically looking at a photograph of words. The words aren't really there. Not for them anyway. You need some other method to make that content accessible.
Why can't screen readers extract text from JPEGs?
Think of it this way - screen readers speak digital text, not pictures of text. A JPEG file? It's just a grid of pixels. No text layer hiding in there. No character encoding. No semantic structure at all. The screen reader encounters an image, full stop. Not the words inside it. This isn't some bug - it's just the fundamental difference between image formats and text-based ones like HTML, tagged PDFs, or DOCX files.
How can you make JPEG content accessible to screen readers?
You've got three main options here, honestly:
- Alt Text (Alternative Text): Stick a descriptive alt attribute on your
<img>tag in HTML. This is the go-to move. The screen reader reads that alt text instead of trying to figure out the image itself. - Optical Character Recognition (OCR): Grab something like Adobe Acrobat, Google Docs, or a dedicated OCR tool. It pulls text from the image. Then you can put that text as a visible layer behind the image or as a separate block nearby.
- Provide a text transcript: Just put the exact text from the JPEG in a visible (or hidden but accessible) text block right before or after the image. Simple.
What about PDFs and other image formats?
It's not just JPEGs. Screen readers can't read text from any pure image format - PNG, GIF, BMP, TIFF - none of them. A PDF works only if it's a "tagged PDF" with a real text layer. A scanned PDF is basically just a bunch of JPEGs stitched together. That won't work either.
Comparison of file formats and screen reader accessibility
| Format | Screen Reader Can Read Text? | Requires OCR? | Best Practice |
|---|---|---|---|
| JPEG | No | Yes | Use alt text or OCR |
| PNG | No | Yes | Same as JPEG |
| GIF | No | Yes | Same as JPEG |
| HTML | Yes | No | Use semantic tags |
| Tagged PDF | Yes | No | Ensure tags are present |
| Scanned PDF | No | Yes | Run OCR, then tag |
Checklist for ensuring JPEG accessibility
- Always add meaningful alt text to
<img>tags containing JPEGs. - If the JPEG contains critical data (charts, infographics, quotes), provide a text transcript or use OCR.
- Avoid using images of text unless absolutely necessary (e.g., for logos or specific design).
- Test with a real screen reader: NVDA (free) or VoiceOver (built into macOS/iOS).
- For complex images (like graphs), describe the data in a longdesc or adjacent text.
People also ask about JPEGs and screen readers
Can a screen reader read text from a screenshot?
Nope. A screenshot is saved as a JPEG or PNG - it's an image. The screen reader can't read any text in it unless you've added alt text or run OCR. Honestly, most accessibility experts say never send a screenshot as your only way of sharing text-based info. It's just bad practice.
Do screen readers announce that an image is a JPEG?
Not exactly. The screen reader will say "image" or "graphic" for a JPEG file. It might also read the filename if there's no alt text. So you'd hear something like "Image, Screenshot_2024.jpg." Zero help. Always include descriptive alt text so users actually get the content.
What is the best way to add alt text to a JPEG in HTML?
The syntax is: <img src="chart.jpg" alt="Bar chart showing 2024 sales increased by 20 percent">. Keep it concise but descriptive. Don't start with "Image of" or "Picture of" - the screen reader already announces it as an image. Focus on what the image means or does.
Can OCR tools make JPEGs fully accessible?
Maybe, but not perfectly. OCR tools can extract text, but the results aren't always flawless. Handwriting, low-res images, or complex layouts mess things up. After OCR, you gotta manually check the text and, if embedding it back into a document, make sure the text layer aligns and tags correctly. OCR is a bridge, not a magic fix.
Frequently asked questions (FAQ)
Can screen readers read JPEG files at all?
Screen readers can detect that a file is a JPEG image, but they cannot read any text inside it. They will announce "image" and may read the filename, but the visual content is inaccessible without alt text or OCR.
Is a JPEG considered accessible if it has alt text?
Yes, if the alt text accurately describes the content or function of the image, the JPEG is considered accessible. However, if the image contains important text, the alt text must include that text verbatim or a summary.
How do I check if my JPEG is accessible?
Use a screen reader (like NVDA or VoiceOver) to navigate to the image. If you hear a meaningful description, it is accessible. You can also use automated accessibility tools (like WAVE or Axe) that check for missing alt attributes.
Do screen readers work with JPEGs in PDF files?
Only if the PDF is a tagged PDF with a text layer. If the JPEG is embedded inside a PDF without a text layer, the screen reader cannot access the text. The PDF must be processed with OCR first.
Short Summary
- JPEGs are pixel-based: Screen readers cannot read text from JPEGs because the format stores visual pixels, not characters.
- Alt text is the primary solution: Always add descriptive alt text to the
<img>tag to convey the image's content to screen reader users. - OCR can help but is not perfect: Optical Character Recognition can extract text from JPEGs, but manual verification is needed for accuracy.
- Avoid images of text when possible: For critical information, use real HTML text instead of embedding text in JPEGs to ensure full accessibility.