Document Intelligence Region in billing differs from the resource settings
Hi all, we are currently using Azure AI Document Intelligence. All resources are set to Europe, whereas the billing shows that "Azure Applied AI Services" (acccounts) is located in the US East. Is there any setting to move all services to the…
Azure Search Index with both text and images
Hi, I want to implement a RAG architecture where the input data is in pdfs and contain text, images and tables. Working only with text is straightforward and I've built such applications before. However the addition of images in the new use case makes…
Document Intelligence Studio output using Python SDK
Hi, I am using document intelligence studio and see that tables are extracted in exactly same manner when as it look in the image. But when doing the same using Python code is very confusing. I have a booklet of 300 pages and each page either has table…
A huge latency added to Document OCR request
I'm experiencing a latency issue with the Document Intelligence service while applying OCR on a large number of images of books. Each image takes about 5 seconds to respond, which I believe is not solely due to the OCR step. I have checked the Azure…
begin_analyze_document encounters (InternalServerError) An unexpected error occurred.
Currently, in the process of processing one kind of documents, all with the same format and almost the same length. We were able to process the first few docs last week. However, this morning, we keep encountering this error. How should we go about…
How to extract the value when the option is circled indicating that it is selected?
I am new in using Azure document intelligence studio and I am using General document model to extract key value pairs and paragraphs content. I have one requirement where if the field is circled, I should get this value as selected. Is there any way I…
Address Field Type in Custom Extraction Model
Hello, I am currently working on a custom extraction model in Document Intelligence Studio (Azure AI) based on the standardized Invoice model. Everything is functioning well, except for the address fields. In the standard Invoice model, these fields are…
Issues with Document Intelligence Character Extraction
Hello MS QA Community, I'm encountering an issue with Document Intelligence (DI) where it's not extracting characters accurately. For instance, in the image provided, there's a large heading containing the phrase '2.0M in THF', but DI is extracting 'THF'…
prebuilt-layout page orientation marks points inconsistent
I have been trying out analyze_document with model prebuilt-layout (2024-02-29-preview). I have noticed that the points of the polygons for lines and words encode the text orientation. That is, when looking at the text horizontally (rotating the image if…
how to add a field to represent currency like there is in the prebuilt invoice document in a custom neural model
how to add a field to represent currency like there is in the prebuilt invoice document in a custom neural model? i want to train my own model to get higher accuracy
Unable to use new API for Document Intelligence
I am trying out the code in this link with a valid key and endpoint Layout model sample code, but I am getting an error message. I am unable to understand what is the reason for this error as I am using the sample document in the sample code to test it…
Standardizing Table Schema from Prebuilt - Invoices Using Azure Document Intelligence
I am using the Prebuilt Invoice Model from Azure Document Intelligence. While the model works well overall, I am facing an issue with the table schema. The invoices I receive come from various sources, each with a different table structure. My goal is to…
Getting "Cannot connect invoice 2022-08-31 grpc server from 127.0.0.1:50051 after 150 seconds." when running container locally.
I keep getting this error when running my container locally using docker compose up. "formrecognizerinvoice[0] azure-cognitive-service-invoice | Cannot connect invoice 2022-08-31 grpc server from 127.0.0.1:50051 after 150 seconds." This…
How to train document intelligence custom built model to respect labeled table's order of items when the table is split into columns?
Hello, I am using the custom built model (neural model) to teach document intelligence what to look for specifically in price tables for some meat price sheets. Often, these vendor meat price sheets have their tables of prices split into two columns. You…
Azure AI | Document Intelligence Studio:InteractionRequiredAuthError
login_required: AADSTS50058: A silent sign-in request was sent but no user is signed in. The cookies used to represent the user's session were not sent in the request to Azure AD. This can happen if the user is using Internet Explorer or Edge, and the…
How to upload a analyzed document done with documentAnalysisClient.AnalyzeDocumentAsync call in code, into the training dataset in code, which mimics the 'Mark as to uploaded into the training dataset' button in Azure DI
After analyzing a doc with the documentAnalysisClient.AnalyzeDocumentAsync, I need a way to upload the tested doc directly into the training dataset of its corresponding blob with the already labeled fields. Basically I need to mimic the yellow…
Azure Cognitive Search + Open AI + Image Content
I'm using Azure cognitive search which uses blob data source to integrate with Open AI. It seems to work fine when i upload PDFs and create search index and refer them in the code. But if the PDF contains images (say a graphical image with text on it) or…
Azure AI Document Intelligence : On-premises Containers
I want to use Azure Document Intelligence containers on-premises. For this, I created a Kubernetes file, referencing the Docker Compose file mentioned here. The containers are running fine. Now, I need some guidance on how to use my containers to…
Different OCR results between PDF and TIFF document
Using the Azure AI Document Intelligence, General documents parser, we are experiencing different OCR results in specific fields if we send the same document as a PDF or TIFF. Has anyone experienced something similar?
Unable to extract images/figures from DocumentIntelligence analyze document client method for docx and pptx files
I have a docx file with text,tables and images. When I'm using the document intelligence prebuilt layout tool https://documentintelligence.ai.azure.com/studio/layout, I'm getting json output but in that there is are no "figures"…