Fact · Speech recognition
1.08percent
On typewritten historical records, the OCR output had a character error rate of 1.08 percent against a labelled sample.
Checked against its source on October 17, 2026
In the source's own words
“OCR achieved a Character Error Rate (CER) of 1.08% and a Word Error Rate (WER) of 5.06%”
Enriching Historical Records: An OCR and AI-Driven Approach for Database Integration, Zahra Abedi, arXiv, 2025, Abstract.
Open the original- What it measures
- The study ran OCR on typewritten historical records and compared the text with a labelled subset of 10 percent of the files, using both character and word error rates.
- What it does not show
- A single study of one document collection. The same OCR engine on other scans, fonts or languages can give different figures.