Using an external Labelling with Azure AI Document Intelligence

Question

We are a small development team currently working on a project that heavily utilizes your "Document Intelligence Studio" with the "Custom neural model.” 

We are thrilled with the whole product and are getting amazing results. However, we are facing difficulties with data labelling.

We need to label a few hundred documents, and the tools provided within the "Document Intelligence Studio" are limited for such large-scale tagging which would involve a whole labelling team. Therefore, we need to use a different labelling solution. We have explored the "Azure machine learning data labelling projects" that use the popular COCO dataset. Despite being an Azure product, we could not find any information on how to use the labelled dataset in our Document Intelligence project. Hence, I am contacting you directly to seek more information.

I would appreciate it if you could provide me with information on labelling tools/formats that we can use in our "Document Intelligence Studio."

Thank you in advance.

Answer 1

Hello @Utopia , Unfortunately, there is no special tool other than Document intelligence studio for labeling at this moment.

Here are some possible workarounds you can try and see if that helps you:

• As we know in the studio, we already have auto labeling, new Human In The Loop features for custom models these should help in fast-tracking the labeling process.
• Pre-labeling speeds up the labeling process where you only need to focus on fixing the specific fields the model needs to improve on.
• Additionally With the latest 2023-07-31 GA, we can train custom models easier by lowering the number of labeled documents needed to train a custom neural model to a single document!  

I would recommend you to go through this documentation: https://techcommunity.microsoft.com/t5/azure-ai-services-blog/azure-ai-document-intelligence-new-capabilities-including/ba-p/3887375

I hope this helps.

Regards,
Vasavi

-Please kindly accept the answer and vote 'yes' if you feel helpful to support the community, thanks.