Document Intelligence platform

From complex documents to structured data

Classify documents, extract fields and tables, preserve layout context, and return application-ready JSON—built for Southeast Asian formats, scripts, and workflows.

Explore the API
  • Structured JSON
  • Regional languages
  • Configurable schemas
document_intelligence.jsonStructured
Supplier
Invoice no.
Tax
Total
{
  "document_type": "invoice",
  "language": "vi",
  "fields": {
    "supplier": "...",
    "invoice_no": "...",
    "tax": "...",
    "total": "..."
  },
  "line_items": [...]
}

One document pipeline

From complex documents to structured data

TurboLens combines regional OCR, layout understanding, and configurable extraction so teams can move from unstructured files to usable application data.

  1. 01

    Ingest

    Send PDFs, scans, or document images through an API-first processing flow.

  2. 02

    Classify

    Identify the document type and route it to the appropriate extraction path.

  3. 03

    Extract

    Capture configured fields, tables, layout structure, stamps, and visual elements.

  4. 04

    Connect

    Return structured JSON shaped for applications, data stores, and reviewer tools.

Beyond basic OCR

Understand the document, not only the text

Basic OCR returns characters. Document Intelligence preserves the relationships between fields, tables, layout regions, and visual signals so the output is easier to use in real workflows.

Key-value extraction

Capture named fields and their values from varied document layouts.

Tables & line items

Preserve row, column, and header relationships for downstream processing.

Layout understanding

Recognize sections, forms, checkboxes, figures, and multi-column structures.

Stamp & seal detection

Locate official markings and include their regions in reviewer workflows.

Image cleanup

Reduce watermark and background interference before text extraction.

Chart & figure analysis

Turn relevant visual elements into structured, reviewable information.

Regional document coverage

Built around the documents your teams actually process

Browse the OCR catalogue

Invoices & receipts

Supplier details, tax fields, line items, totals, and payment references.

Explore documents

Identity documents

Regional national IDs, passports, bilingual fields, and machine-readable zones.

Explore documents

Contracts

Parties, clauses, dates, obligations, signatures, and structured sections.

Explore documents

Trade & logistics

Bills of lading, packing lists, customs declarations, and delivery records.

Explore documents

Forms & certificates

Tax forms, applications, medical certificates, and government documents.

Explore documents

Apply it to your workflow

See Document Intelligence by industry

Explore how extraction, classification, and structured outputs support insurance, logistics, public-sector, and custom enterprise workflows.

Browse solutions

Frequently Asked Questions

Document Intelligence combines OCR with document classification, layout understanding, and structured extraction. It turns PDFs, scans, and document images into organized data that applications and reviewers can use.

Basic OCR primarily returns text. TurboLens also captures field relationships, tables, document sections, and configured output structures so the result fits downstream workflows.

TurboLens supports regional invoices, receipts, identity documents, passports, contracts, tax forms, customs documents, medical certificates, and other configured enterprise document types.

Teams submit documents through API-first workflows and receive structured JSON. Output schemas and processing logic can be aligned to the fields and document variants used by the application.