Azure-docs: Couldn't get OCR results as in the demo page

[whataride]

I tried the script, and all is good. However, when I used the images from the demo page here: https://azure.microsoft.com/en-us/services/cognitive-services/computer-vision/?v=18.05, "Read text in images" session, no text were detected. I am wondering why is the discrepancy.

Specifically, the images I tried are:

1. https://azurecomcdn.azureedge.net/cvt-115aaf389870d14afd6954e357004438425b429102bc9f77723150fc408b90c8/images/shared/cognitive-services-demos/read-text/read-1-thumbnail.png

2. https://azurecomcdn.azureedge.net/cvt-91abb847976d001a8ecbbab8773dd40e316af17ae348d2590bddca28dccf85b7/images/shared/cognitive-services-demos/read-text/read-3-thumbnail.png

---

Document Details

⚠ Do not edit this section. It is required for docs.microsoft.com ➟ GitHub issue linking.

• ID: ab15e2d6-7f06-3037-d47d-2d76fb8456ac
• Version Independent ID: 396d1582-f04b-ad66-88ee-ae614ecd0ece
• Content: Computer Vision Python quickstart OCR - Microsoft Cognitive Services
• Content Source: articles/cognitive-services/Computer-vision/QuickStarts/python-print-text.md
• Service: cognitive-services
• GitHub Login: @noellelacharite
• Microsoft Alias: nolachar

[SaurabhSharma-MSFT]

@whataride Thanks for your feedback! We will investigate and update as appropriate.

[YutongTie-MSFT]

@whataride Thanks for the feedback! I have assigned the issue to the content author to investigate further and update the document as appropriate.

@noellelacharite Hi, this is a doc-related issue/ product-question issue, can you please check it to see if there something wrong with our document? Thanks a lot!

[Gullfaxi171]

hi ! same issue here, any news to explain the difference ?

[hjpark914]

hello ! I have same issue too. any news to explain the difference ?

[tchristiani]

@YutongTie-MSFT please reassign to @PatrickFarley

[nandagopal1992]

I think this might be because you are using different API than the one in the demo page. The demo page uses recognize text API rather than the OCR API

[skysrc]

Same here...not able to get the same result as in demo page via vision api.

[hjpark914]

Hi. I think I solved it. I will summarize tomorrow

[PatrickFarley]

@hjpark914 ping

[skysrc]

Apparently, the demo page are using "vision/v2.0/recognizeText" api. So i had switched to this for the solution.

[PatrickFarley]

Hi @whataride, @skysrc is correct. "OCR" and "recognizeText" are two different APIs. OCR is the legacy engine, while recognizeText is the newer engine. I will try to make this more clear in the docs.

please-close

@kamathn for visibility.