I just connected Vision App to my Base and have a few questions:
First, my Use Case:
We are entering Orders data from a receipt into Airtable, so I thought and searched if there was a way to maybe scan the Receipt and extract the text on the Receipt.
I have set a field as the Attachment field, but everytime I add a new Record and add a new scanned Receipt image file to the Attachment field, and then click “Analyze” in the app window, it says it will bulk extract all the attachment field files of all Records, including existing Record ones I have already added earlier and already “Analysed”. It says it will cost x credits, which includes all the files of the existing Records as well. Is there a way to only click to Analyse new Records that have been added? Or another way to setup so we can upload a few scanned Order files and extract just those, each time?
I am using Text extract, and all the text is saved into a single text field that I chose after initial extraction. Is there a way to extract some of the text and place them in different fields? The presumption is the app will learn the image and repeat to place pieces of texts into different fields?
I do see in “example” in Google’s website about Vision, that it could extract pieces of text and numerically identify them as parts, for example.