paddleocr-text-recognition

To extract text from images or PDFs, this Skill sends local file paths or URLs to the PaddleOCR API. It returns recognized text and location data, and you can customize the output or stream it directly to stdout.

Yes, you can process PDF documents and images directly from a web URL. The Skill accepts web URLs as input, fetches the remote file, and extracts the text content automatically via the PaddleOCR API.

Yes, text recognition requires authentication. You must set the environment variables PADDLEOCR_OCR_API_URL and PADDLEOCR_ACCESS_TOKEN to authenticate your requests and operate the extraction process.

Yes, PaddleOCR preserves text location data. This makes it suitable for structured document processing like invoices, receipts, forms, and tables where the spatial position of the extracted text is important.

The required dependency for image to text extraction is the httpx library. This Python package handles the HTTP requests needed to communicate with the PaddleOCR API for processing your files.

PaddleOCR provides options for output customization and direct stdout streaming. This allows you to pipe the extracted text and location data directly into other command-line tools or scripts without intermediate files.