Announcing Amazon Textract Insights: Verifiable generative AI for documents
Announcing Amazon Textract Insights: Verifiable generative AI for documents
Amazon announced the general availability of Amazon Textract Insights, a generative AI feature available through a new API within Amazon Textract that lets users describe desired fields in natural language and submit a document or mixed multi-document packet to receive structured results with confidence scores and page-level grounding, with no model training required. Users define extraction in JSON Schema blueprints with natural language field descriptions and inference types, can start from AWS public blueprints or author custom ones, and can enable splitting on files up to 3,000 pages from Amazon S3. The service supports PDF, PNG, JPEG, and TIFF inputs in English, French, German, Italian, Portuguese, and Spanish, and is priced per page.
Source: AWS Machine Learning Blog · aws.amazon.com