SanadGah's processing engine converts Persian printed documents into text and structured data using OCR and linguistic post-processing. API-first, deployable within your own infrastructure.
Security & data governance commitments
{
"document_id": "doc-001",
"accuracy": 0.973,
"pages": 18
}
قرارداد همکاری شماره سند: ۱۴۰۴/۰۲۸۷ تاریخ انعقاد: ۱۴۰۴/۰۳/۱۵ طرفین قرارداد: ۱. شرکت آریا فناوری پارس، به نمایندگی علی احمدی، به سمت مدیرعامل ۲. شرکت داده‌پردازان نوین، به نمایندگی سارا محمدی، به سمت مدیر فنی موضوع قرارداد: همکاری در توسعه و پشتیبانی سامانه‌های نرم‌افزاری سازمان، شامل طراحی، پیاده‌سازی و نگهداری... مدت قرارداد: ۱۲ ماه از تاریخ انعقاد مبلغ قرارداد: ۲,۴۵۰,۰۰۰,۰۰۰ ریال شرایط پرداخت: به‌صورت مرحله‌ای و مطابق پیوست مالی...
Organizations face paper archives, weak OCR tools, and manual processes. SanadGah turns these challenges into structured, searchable data.
Millions of pages of paper documents, image-only PDFs, and scanned files — with no searchable text, no data extraction capability, and effectively archived yet unusable.
Converting Persian printed documents into searchable text and structured data through batch processing.
General-purpose OCR tools show low accuracy on connected letters, contextual letterforms, and Persian printed text; output typically requires extensive manual review.
A specialized Persian OCR engine with linguistic post-processing (NLP) to reduce common errors in Persian printed documents.
Manually entering data from documents takes hours of human labor per document; high human error and effectively zero scalability.
Batch and parallel processing through an API — no manual intervention required per document.
OCR returns raw text, not structured data; the extracted text can't be connected to ERP, CRM, or archival systems and requires manual reprocessing.
Key field extraction from contracts and invoices (initial version) and structured JSON output through a unified API.
SanadGah converts Persian printed documents — through a multi-stage processing pipeline — into searchable text, key fields, and structured JSON output, via API and in batch.
Submitting documents via REST API, individually or in batch. PDF and image support.
Rotation correction, contrast enhancement, and noise removal — to improve input quality before OCR.
Text extraction from Persian printed documents using a specialized Persian OCR engine, designed for connected letters and contextual Persian letterforms.
Correcting OCR output with natural language processing (NLP) — reducing errors in Persian words, Persian numerals, and punctuation.
Identifying key fields from contracts and invoices (parties, date, amount, document number) — initial version, extensible.
Delivering results as JSON and full text via API and webhooks — ready to connect to enterprise systems.
SanadGah is designed API-first — submit documents via REST API, receive results via webhook, and get structured JSON output for integration with your organization's existing systems.
Submit documents individually or in batch, using each organization's dedicated API key. PDF and image support with immediate processing ID response.
Once processing of each document completes, the result is sent to the organization's endpoint via webhook — with no need for polling or synchronous waiting.
Processing results include full text, extracted key fields, and processing metadata — ready for mapping to ERP, CRM, or archival systems.
API documentation, code samples, and integration guides. Official SDKs for common languages are on the product roadmap.
At SanadGah, instead of numeric claims, we evaluate processing accuracy on a sample of your organization's real documents — with defined criteria, transparent reporting, and a focus on Persian printed documents.
A sample of your organization's Persian printed documents is collected, and the OCR output is compared against reference text. The final report includes the overall accuracy rate and a breakdown of errors.
Evaluation is performed based on three metrics: Character Accuracy, Word Accuracy, and accuracy of key fields in contracts and invoices.
Errors are reported broken down by type: connected letters, Persian numerals, punctuation, and specialized terms — to identify which aspect requires improvement.
Linguistic post-processing (NLP) is applied to OCR output to correct common Persian errors. This layer is continuously improved with new evaluation data.
After evaluation, a full report — including the accuracy rate, sample output, error analysis, and a deployment model recommendation — is made available to the organization.
The current focus is on Persian printed documents. Handwritten documents, low-quality historical archives, and texts with non-standard fonts may have lower accuracy and are on the product roadmap.
We don't claim accuracy; we evaluate. Every number below is verifiable on your own documents — before any contract.
97.3%
Persian OCR accuracy
In sample evaluations
2.4s
Avg. processing time
Per 18-page document
100%
Data sovereignty
In On-Premise deployment
0
Model training on your data
Across all deployments
Currently accepting pilot organizations
If you want to be among the first to receive a full accuracy evaluation on your own documents, apply now.
What sets SanadGah apart from general-purpose OCR tools, foreign cloud services, and traditional document management solutions — specifically for Persian documents.
A dedicated focus on Persian printed documents, with linguistic post-processing (NLP) to correct OCR output and reduce errors in connected letters and contextual Persian letterforms.
A parallel architecture for batch document processing. REST API and webhooks for integration with enterprise systems — with no manual intervention required per document.
On-premise deployment with no dependency on foreign cloud services — built for organizations whose data must not leave their own infrastructure.
Rather than quoting a number, we evaluate accuracy on a sample of your organization's real documents. A technical team based in Iran, supporting deployment and processing optimization.
Persian printed document processing across legal, insurance, enterprise archiving, and administrative correspondence — with an initial focus on contracts, invoices, administrative letters, and forms.
Document processing with transparency, data governance, and access control — designed for organizations whose data must not leave their own infrastructure.
With on-premise deployment, documents can be processed within the organization's internal infrastructure with no dependency on foreign cloud services. In this model, data does not leave the organization for processing and is managed in accordance with the organization's data residency policies.
Data submitted by customers is not used for model training or service improvement. This is a contractual commitment and data policy — not merely an internal practice.
Every processing request, submission time, status, and API access is traceable. Processing logs are made available to the customer for transparency and accountability.
Input and output documents can be stored or automatically deleted in accordance with the organization's defined retention policy. Upon contract termination, full data export or deletion is available.
From accuracy evaluation on your own documents to integration, deployment, and technical support — our team alongside you at every stage.
We evaluate a sample of your organization's real documents before signing a contract — no numeric claims, only observable results on your own data.
REST API integration, webhook setup, and — if you choose — on-premise deployment within your organization's infrastructure, along with technical documentation.
Technical support from our team in Iran, with hands-on assistance during deployment and processing optimization tailored to your organization's document types.
Support for usage-based payment, SaaS subscription, On-Premise license, and enterprise contract — with payment in Iranian Rial and official invoicing.
Transparency about what is ready today, what is in development, and what is planned — without unfulfillable promises.
The exact timing for each of these items is provided in the demo session, based on your organization's needs. No capability is advertised as available before it is actually ready.
In a single session, we process a sample of your organization's real documents, compare the results against reference text, and deliver an accuracy report. Decisions based on results — not promises.
SanadGah team response typically on the first business day