[andrejk]

Azure has a separate service to read forms and formatted docs. https://azure.microsoft.com/en-us/services/cognitive-service...

[cdolan]

Thanks I was not aware! I’ll have to figure out how to easily test it (AWS tectrsct is actually really nice in that regard)

[bpchaps]

Do you know how good it is? I have a LOT of structured documents that I need to OCR.

[CorneliaKara]

MSFT person here - give it a try! sign up, and you get a free trial that can allow you to easily benchmark.

[Havoc]

Interesting. That is actually something I've been looking for and I do have a msdn sub.

fyi that defaulted to indian rupee as default currency for me (UK based & zero indian connections). Weird

[dav43]

Have you tried the tabula free program? I use it for some finance work reading filings.

http://tabula.technology/

[bpchaps]

Yep! It's great, but is maybe 60% there, so I'm looking for something that can extract much more structure from a document. I doubt what I'm looking for will exist for another 10 years, though.

[leeoniya]

is it feasible to create loose templates for where the data is and extract that way? i have a mothballed project that did pretty well. it was able to discern different templates from a mass of documents.

[ocrcustomserver]

I'm curious, if you email me a sample I can tell you what's possible.