Hello, I needed assistance with something. When you use the
main site, you get an option to “download”, after OCR-ing. In the API,
you only get back an input, which looks nothing like the text in the
downloaded file, example, heres the downloaded file output link:
The TextSaver: Save Your Text Online - Fast, Free and Easy! Here’s the API output text link:
The TextSaver: Save Your Text Online - Fast, Free and Easy! API flags, these are the same as
the web request: ‘language’: ‘eng’, ‘OCREngine’: 3, ‘isTable’ : True,
‘detectOrientation’ : True, ‘isSearchablePdfHideTextLayer’ : True,
‘scale’: True,
1 Like
The OCR API returns the OCR result in a JSON formatted response.
If you only need the text, then please look for the “ParsedText”: part of the JSON response. This is exactly the same text as displayed on the online ocr front page.
1 Like
No, As shown in the output, it’s not the same output. It’s not formatted.
If I enter the download buttons output into chatGPT, it understands the content. If I put the parsedText. It does not understand. It’s not pure text, it’s formatted oddly.
1 Like