Turn a scanned Amharic PDF into editable Ge'ez text
Upload a scanned Amharic book, report or document and get every page back as editable Unicode Amharic, in Word and as plain text: each fidel syllable read with its vowel order, Ethiopic punctuation such as ። and ፣ kept, uncertain words highlighted. The first page is free without an account; up to 300 pages is $5.
- Ge'ez (Ethiopic) script in Unicode, searchable and editable
- Each syllable read with its vowel mark, not just its base consonant
- Ethiopic punctuation (። ፣ ፤ ፥) and numerals kept as printed
- One block of text per scanned page
- Choose Tigrinya instead for Tigrinya documents in the same script
Why Amharic OCR goes wrong
Ge'ez is a syllabary: each character is a consonant with a vowel shown by a small change to its shape, a stroke, a ring or a foot. Misreading that small mark changes the syllable and often the word. Scans with low resolution or faded print make those marks hard to see, and general OCR tools trained mostly on Latin text tend to confuse them.
Here each page is read separately at high resolution, with instructions that the text is Amharic in Ge'ez, that each syllable must be read with its vowel order, and that Ethiopic punctuation and numerals stay as printed. Words the reader is unsure of are marked and highlighted rather than guessed.
Getting a good result
- Scan at 300 dpi or take sharp, straight phone photos in good light
- Upload the whole document at once; every page is read separately
- Check highlighted words against the scan before using the text
- For Tigrinya documents, choose Tigrinya so the right spelling rules are used
What people convert
- Amharic books and pamphlets for research or a new edition
- Ethiopian certificates and official letters before translation
- Church and community documents
- Reports and studies published only as scanned PDFs
How the tool works, what it costs and how your files are handled is on the Scanned PDF to Word page; prices for every tool are on the pricing page.
Last updated
Questions and answers
Can it read Tigrinya too?
Yes. Choose Tigrinya; it uses the same Ge'ez script with its own spelling and punctuation habits.
How many pages for $5?
Up to 300 pages in one job. Unlimited covers 600 pages per job.
Which font does the Word file use?
Nyala, a common Ethiopic font on Windows; if it is missing, Word substitutes another Ethiopic font, and you can change it in one step.
Is the first page free?
Yes, without an account. A scan of up to 3 pages is free in full.