Announcing an updated text detection model for Amazon Textract

Amazon Textract now provides an updated text detection model that improves accuracy, expands language support, and enhances multilingual table extraction. The updated model applies to the DetectDocumentText and AnalyzeDocument APIs and requires no changes to your Amazon Textract API requests or response schema. Your applications are served by the updated model automatically, so you can continue calling Amazon Textract exactly as before.

The updated model expands language coverage to more than 40 languages, including Albanian, Arabic, Basque, Belarusian, Bosnian, Bulgarian, Chinese (Simplified and Traditional), Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Hebrew, Hindi, Hungarian, Icelandic, Italian, Japanese, Korean, Latvian, Lithuanian, Luxembourgish, Norwegian, Polish, Portuguese, Romanian, Russian, Serbian Cyrillic, Serbian Latin, Slovak, Slovenian, Spanish, Swedish, Turkish, Turkmen, and Ukrainian, enabling reliable processing of previously unsupported scripts. Accuracy improvements include enhanced long-word recognition and handwriting detection. Additionally, multilingual table extraction is more robust on merged cells, skewed or rotated tables, and partial or corrupted tables. These improvements help you accurately process invoices, forms, identity documents, and other multilingual or complex documents, including those with tables, at scale.

The updated Amazon Textract model is available in all AWS Regions where Amazon Textract is available. To learn more, see the Amazon Textract product page and the Amazon Textract Developer Guide.