Fix OCR problems — missed text, wrong characters | yomeru.ai
If text isn't detected or characters are wrong, start with reprocessing in the correct mode. Most OCR issues come from image quality or mode mismatch.
If the text extraction results are not what you expected, one of the fixes below should help. Most issues come down to image quality or choosing the right processing mode.
Quick fix
Before anything else, try reprocessing the page with the correct mode selected (Manga vs Novel). This solves the majority of recognition problems.
Common Issues
When Reprocessing Helps
Reprocessing is worth trying when you selected the wrong mode (Manga vs Novel), when the first attempt had a temporary glitch, or after you have replaced the source image with a higher-quality scan. It will not improve results if the underlying image is blurry, extremely low-resolution, or uses highly artistic hand-drawn lettering.
Still Having Trouble?
If none of the above fixes your issue, contact support through the website's contact form. Include a description of what you are seeing and, if possible, a screenshot of the problematic page.
Related Pages
Troubleshooting — OCR, uploads, processing, and login | yomeru.ai
Find fixes for missed speech bubbles, failed uploads, stuck processing queues, and login or 2FA problems — or contact support directly.
Fix upload errors — file size, format, and credit issues | yomeru.ai
Images must be under 10 MB (JPG, PNG, WebP). ZIPs and PDFs up to 100 MB. Check the error message — it names exactly which file caused the problem.