← Back to Blog

Convert Old Scanned Books to Audiobooks: Complete OCR Guide

You have an old book — maybe a rare edition, a family heirloom, or an out-of-print title that never made it to digital. The pages are physical, not files. Can you turn it into an audiobook? Absolutely. Here is how.

Step 1: Scan or Photograph the Pages

You have two options: use a flatbed scanner for the highest quality, or use your smartphone camera for convenience. Modern phone cameras produce surprisingly good results for OCR (Optical Character Recognition).

  • Smartphone approach: Use an app like Adobe Scan, Microsoft Lens, or the built-in iPhone document scanner. These apps automatically straighten pages, adjust contrast, and optimize for text recognition.
  • Scanner approach: A flatbed scanner at 300 DPI produces excellent results. Most libraries have free scanners available.
  • Tip: Good lighting is essential for phone scans. Photograph near a window during daylight. Avoid shadows falling across the text.

Step 2: OCR — Extracting Text from Images

OCR software converts images of text into actual, editable text that TTS engines can read. The technology has improved dramatically — modern OCR handles old typefaces, slight skew, and even some handwriting.

Free OCR options include Google Drive (upload images and open as Google Docs), Tesseract (open source), and online tools like OCR.space. For challenging scans — old books, unusual fonts, yellowed pages — commercial tools like ABBYY FineReader provide noticeably better accuracy.

Step 3: Clean Up the Text

OCR is good but not perfect. Common issues include:

  • Hyphenated words at line breaks that did not rejoin properly
  • Page numbers and headers mixed into the text
  • Footnote markers that interrupt the flow
  • Character substitutions: "rn" misread as "m", "l" misread as "1"

A quick pass through the text fixing obvious errors takes 30 to 60 minutes for a typical book and dramatically improves the audiobook quality. Focus on the first paragraph of each chapter — these are the most noticeable.

Step 4: Convert to Audiobook

Save your cleaned text as a PDF, EPUB, or plain text file and upload it to an AI audiobook maker. The platform handles chapter detection, voice synthesis, and audio formatting automatically.

Some platforms, including AudioBookByMe, accept scanned images directly — the OCR step is built into the pipeline. Upload your page photos and the system extracts text, structures chapters, and generates audio without you needing to handle text extraction manually.

Which Books Work Best?

Cleanly printed books from the 20th century onwards produce excellent OCR results. Older books with ornate typefaces or degraded print require more cleanup. Books with lots of images, charts, or non-standard layouts are more challenging — the OCR may mix caption text with body text.

The best candidates are out-of-print non-fiction, rare philosophy texts, family histories, and academic works that never received a digital edition. These are often exactly the books that would benefit most from an audio version.

Legal Considerations

Creating an audiobook from a scanned book for personal use is generally fine. Publishing or distributing it requires the book to be in the public domain or for you to have permission from the copyright holder. Books published before 1929 in the US are safely in the public domain. Check copyright status before distributing commercially.

Upload scanned pages directly

AudioBookByMe handles OCR, chapter detection, and narration in one pipeline. Upload page images and download a finished audiobook.

Create Your Free Audiobook

Related Reading

Free: ACX Audio Requirements Checklist

The complete spec sheet for publishing on Audible, Apple Books, and more — plus tips to speed up your audiobook production.

No spam, ever. Unsubscribe anytime.