UnifyAPI
← Back to catalog

ocr.extract_text

mock data$0.002 / call

Extract Text from Image (OCR)

Extract text from any image or PDF URL using OCR — supports 20+ languages including English, Russian, Chinese, Japanese, Korean, Arabic. Returns recognized text. Handles PNG, JPG, GIF, BMP, PDF, TIFF (OCR.space)

AI & MLprovider: Ocr#ocr#ai

Input schema

urlrequired

URL of the image or PDF to extract text from (PNG, JPG, GIF, BMP, PDF, TIFF supported)

string
filetype

File type hint — set if URL has no extension or content-type is wrong

string
language

OCR language: "eng" (English, default), "rus" (Russian), "ger" (German), "fre" (French), "spa" (Spanish), "jpn" (Japanese), "kor" (Korean), "chs" (Chinese Simplified)

string
detect_orientation

Auto-detect and correct image orientation (default false)

boolean

Call via REST

curl -X POST https://unifyapi.pro/api/call/ocr.extract_text \
  -H "Authorization: Bearer uak_live_..." \
  -H "Content-Type: application/json" \
  -d '{"url":"<url>","filetype":"<filetype>","language":"<language>","detect_orientation":"<detect_orientation>"}'

Try it

Paste a UnifyAPI key (create one in the dashboard) and run a real call.