Is it possible to read PDF's using Virtual Agent?

[SaiNishwanR]

I need to extract all the text from a PDF when uploaded to a form as an attachment, save them and use all words to look for specific keywords.

Document Intelligence is not an option as it is paid sadly, so is there any other option if Virtual Agent is not possible? This has to be performed on the platform itself so i cannot use API calls to perform this operation.

[Dr Atul G- LNG · Accepted solution]

Hi @SaiNishwanR 

No, it might be later but also alone VA can't dom, it needs the help of Gen AI/ Now Assist to work which is again paid.

[kaustubhdub]

Hello regarding the genAi use for extracting the pdf content, can we do it with the agent itself without involving the document intelligence? If yes, can you tell me how? as i tried but the agents were not extracting the data. 

[BillMartin]

you will need an OCR technology to read image to text. you may want to use open-source technology or create one.

To create an OCR (Optical Character Recognition) technology in JavaScript, you can leverage libraries like Tesseract.js, a pure JavaScript port of Tesseract, which is a powerful and open-source OCR engine developed by Google.