Task 04 OCR Skill Creation
Task 04 OCR Skill Creation
VANTAGE FUNDAMENTALS
Task 04 OCR Skill Creation
Task 04 OCR Skill Creation
During this task, we will use the “OCR Skill” from the Skill Catalog and utilize the
Vantage API to facilitate end-to-end document processing. The most common usage
scenario for document processing in Vantage is calling Vantage Skills from other
software systems. Therefore, we will use the Swagger tool in this task to help you
visualize the workflow and the API code that needs to be created.
By completing this task, you will learn how to:
• Create a new OCR Skill
• Process documents using the Swagger tool
Note: Please be aware that screenshot depictions of the GUI in this task may differ from your
Vantage environment, depending on which version you are using.
Be sure to carefully follow the task instructions when creating and naming folders, Skills, and
files.
Note: You can learn more about these properties on the Help page in Vantage.
7. On the Export tab, select JSON>Preserve document structure, PDF, DOCX and TXT formats.
The result should look like this:
18. Copy and paste the Skill id into Notepad, and then
type the string “Skill ID” in the line above the
pasted value. We will need this value later.
27. The next step is to get transaction info. In Swagger, scroll down, then click on the following GET
line: /api/publicapi/v1/transactions/{transactionId}.
28. Click Try it out.
29. Paste your transaction id to the corresponding field, as shown in the screenshot below:
Note: In this step, you could alternatively select from any of the export formats we previously
selected: PDF, DOCX, JSON, or TXT. To be aligned with later instructions, we recommend you
copy the PDF format File ID.
34. Click Try it out, then enter the transactionId, then the fileId (you stored both in Notepad),
then click the blue Execute button.
35. Click Download file (see below).
36. Open the file to check the results. This originally was a JPG file, which is an image format with
no text layer. Now we have a PDF file that looks the same, but there is an underlying text layer.
Click Ctrl+F and search for the word amount. You will find four instances; this is now a fully
searchable document!
Typically, this PDF file, and the other files in formats you initailly selected for export, are used in
various critical downstream business processes.
Great job! You have successfully tested the OCR Skill to do end-to-end document processing using
the Swagger tool!