Azure Document Intelligence Studio - Are there any keyboard shortcuts I could use when mapping fields into a table in a Custom Extraction Model?

Question

Within Azure Document Intelligence Studio, I've created a Custom Extraction Model. Now, I'm mapping out OCR'd text to each corresponding field, and I would like to know if there are any shortcuts that could help improve efficiency.

From what I understand, Shift + 1, 2, 3, etc. map out text boxes to the main field section and that's great, but what if I'm mapping out fields from within a table? Is there a similar shortcut?

For instance, what if I want to assign a new field to column "product_desc", is there a shortcut available for that?

Additionally, are there any shortcuts to insert and remove a table row? That would also help a lot.

Thanks in advance!

Answer 1

Hi RAFAEL BELOKUROWS!

Welcome to Azure AI Q and A forum. Thank you for posting your query here.

All the tips and tricks are documented in this section of Custom Labelling tips.

There is not much keyboard shortcuts to share. But you can follow below practice for faster labeling

1. Search for specific words - Enter the word in Search box of Document intelligence resource and hit enter to find the section.
2. Auto label tables - Start by selecting the table icon on the page and select on the auto label button.
3. Shift selects - Click on the first word of the span you want to label, hold down the Shift key, click on the last word of the span, it will select the entire span of text
4. Region labeling- How to Use Region Labeling:
   1. Open the Document: Load the document in Azure Document Intelligence Studio.
   2. Select the Region Labeling Tool: Choose the region labeling tool from the toolbar.
   3. Draw Bounding Box: Click and drag to draw a bounding box around the region of text you want to label.
   4. Assign Label: Assign the appropriate label to the bounded region.
5. Label overlapping fields - If you expect your analyze results to contain overlapping fields, Draw a bounding box around the relevant region of text. Ensure that the bounding boxes can overlap as needed. Assign the appropriate labels to each region. If a word appears in both fields, ensure it is included in the overlapping bounding boxes. Each word can be labeled for up to two fields. Both complete and partial overlaps are supported.

Reference - Custom labelling tips.

Thank you.