Azure OCR Kurdish language support

Question

Using Document Intelligence Studio and Azure AI services, computer vision can extract Kurdish text from image files. However, the extracted data shows significant deviations when compared to the original text.

Answer 1

Hi Dev_Admin,
Greetings & Welcome to the Microsoft Q&A forum! Thank you for posting your query!

I understand that you're encountering challenges with the accuracy of Kurdish text extraction using Document Intelligence Studio and Azure AI services. The lower accuracy could stem from the fact that prebuilt models may not be fully optimized for Kurdish text.
Here are steps to improve the results:

• If the prebuilt models don't meet your accuracy requirements, consider training a custom model using Document Intelligence Studio. This approach allows you to fine-tune the model specifically for Kurdish text, improving recognition.
• The accuracy of text extraction can be affected by the quality of the input image. Make sure that the images you are using are of high quality and have good contrast between the text and background.

If the issue persists, submit examples of original and extracted text to Microsoft Support via the Azure portal. This helps improve the service by contributing to model optimization.

Hope this helps. Do let us know if you have any further queries.

---

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