Hebrew Handwritten Text: Optimal Solution for Recognition of Handwritten/Mixed English and Hebrew Text in Tax Deduction Forms

Question

I have a use case for tax deduction forms that have significant variation among them. They might be multi-page and have different structures. The general outlay is a table with some possible extra information around it. There are a list of columns that should be in each of them, but many of the documents lack some columns or have the information elsewhere in the document.

I am debating whether to use a neural model or train multiple template models. One of my main concerns is that some of the documents may have handwritten/mixed handwritten and typed-out text/mixed English and Hebrew text. I am wondering how this should affect my choice of solution.

Thank you!

Answer 1

Hi @Shahar Spencer,

Thank you for question. I recommend using a custom neural model to extract the required information from the tax deduction forms.

Regarding the handwritten/mixed English and Hebrew text, Document Intelligence supports the recognition of handwritten text in only English, French, German, Italian, Portuguese, Japanese, Korean, Chinese Simplified and Spanish. However, it does not currently support Hebrew handwriting recognition.

However, you can train a custom model to recognize the printed text in both English and Hebrew. You can also train multiple template models for different document structures and use a composed model to analyze the documents.

See: Supporting language documentation.

Also, I encourage you to submit a feature request for Hebrew handwritten text through the Azure Feedback Forum. Here's the link to the Azure Feedback Forum: Post idea · Community (azure.com)

Hope this helps. And, if you have any further query do let us know.

---

If this answers your query, do click Accept Answer and Yes for was this answer helpful.