Built on our open-source work in Vietnamese OCR and document layout analysis, we digitize documents end-to-end: detect layout, read Vietnamese text accurately, and extract key fields from financial statements, invoices, and forms into clean structured data.
Pick a sample document and run the pipeline: layout analysis, text recognition, then structured extraction.
Press Run OCR to see the recognised text.
Simulated with sample documents (organisations, IDs and signatures are placeholders). The structured extraction is computed directly from the recognised lines. The real pipeline runs YOLOv7 layout analysis + VietOCR recognition.
High-accuracy Vietnamese text recognition, robust to fonts, scans, and photos.
AI-based detection of titles, paragraphs, tables, and figures to understand document structure.
Key-information extraction from financial statements and reports into analysis-ready datasets.
Reconstruct tables from scans into spreadsheets and databases while preserving structure.
Batch processing pipelines for large archives: deskew, denoise, OCR, extract, and index.
Combine OCR with LLMs to classify documents, resolve ambiguity, and validate extracted fields.
Input
YOLOv7
VietOCR / TrOCR
NLP / LLM
JSON / DB
Automatically extract balance-sheet and income-statement figures from scanned reports for credit analysis and data platforms.
Cut manual data entry: invoices, contracts, and forms are read, validated, and pushed into ERP/accounting systems.
Turn decades of paper archives into a searchable digital library with full-text OCR and metadata extraction.
Open-source optical character recognition pipeline tuned for Vietnamese text in scanned documents and photos
Key-information extraction from scanned financial statements into structured, analysis-ready data
YOLOv7-based detection of titles, paragraphs, tables, and figures for document understanding pipelines