A user reported a tricky issue with a few PDF files. This issue shows only if you use OCR Engine2 and have language auto-detection enabled:
- OCREngine=2
- language=auto
In this case - for only a few PDF files - the OCR API returns the error message "E301: Unable to process the file. Please contact support."
The issue is rare. But if you encounter it, the workaround is to avoid the language autodetection and simply specify a concrete OCR language e. g. language=ger. Note that English OCR is always included. So even if you specify German or Chinese as OCR language, OCREngine2 will also detect English text correctly.
PS: If you encounter this error, please contact tech support at team AT ocr.space and send us one or two of these “problem PDF files” for testing to that we can debug this issue better.
PPS: The issue is not related to this older (fixed) issue, even so the API error code was the same.