Online OCR workspace

Convert images and PDFs into clean text.

UnlimitedOCR brings an OCR.ac-style upload workflow into a ShipAny application, with Baidu OCR wired as the first provider.

Drop file or paste image URL

Supports JPG, PNG, BMP, PDF, and OFD. PDF files are processed by page through the provider.

Upload an image, scan, receipt, form, or PDF to extract text here.

OCR tools

Built for common document capture jobs.

Image to Text

Extract readable text from JPG, PNG, and BMP files.

PDF to Text

Send PDF pages to Baidu OCR and collect clean text.

Screenshot OCR

Turn product screenshots and UI captures into copyable text.

Forms & Receipts

Recognize invoice, receipt, and form content for workflows.

Workflow

A fast path from document to usable text.

1Upload a document or paste an image URL

2Recognize with baidu_general_basic

3Copy text or inspect normalized JSON

API

One normalized endpoint for providers.

The app exposes a local Next.js route that keeps Baidu credentials server-side and returns normalized OCR text, lines, provider, and raw response data.

POST /api/ocr
Content-Type: application/json

{
  "provider": "baidu_general_basic",
  "imageBase64": "...",
  "url": "https://example.com/document.png"
}

Is this using the ShipAny template?

Yes. The project keeps the ShipAny Next.js structure, theme system, Tailwind setup, layouts, and shared UI components.

Does live OCR work now?

It works after BAIDU_OCR_API_KEY and BAIDU_OCR_SECRET_KEY are configured in the runtime environment.

Can it support more providers?

Yes. The /api/ocr route is shaped around provider selection so Google, Azure, DeepSeek self-hosted, or Unlimited-OCR can be added next.