Released: Jul 13, 2025

V25.7 での更新項目

機能

• The new AsposeAI class provides the bridge between LLM models and the OCR engine. The AsposeAI class integrates external AI models (e.g. from Hugging Face) into OCR recognition results for spell-check correction and intelligent post-processing.

Released: Jun 13, 2025

V25.6 での更新項目

機能

• Added a debug mode to the API to allow users to view detection area results on the image.
• Added 'confidence' setting for text lines in RecognitionResult.
• Updated ONNX Runtime to version 1.22.0.

不具合の修正

• Fixed hOCR output formatting error.

Released: May 21, 2025

V25.5 での更新項目

機能

• Performance optimization: Decreased the recognition time for the default setting.

不具合の修正

• Fixed an issue with missed pages during PDF recognition.

Released: Apr 8, 2025

V25.4 での更新項目

機能

• Added support for markdown output format with document layout.
• Automatic language detection during recognition.
• Improved DOCX output format.

Released: Mar 26, 2025

V25.3 での更新項目

機能

• Added universal recognition of Arabic, Persian and English alphabets.
• Added support for the automatic analysis of image content and detection of layout blocks.
• Improved recognition speed.

不具合の修正

• Fixed an issue with saving recognition results to hOCR.

Released: Feb 27, 2025

V25.2 での更新項目

機能

• Exposed control over ONNX session options for advanced users.
• Added automatic detection of image language, supporting: English (Latin), Cyrillic, Arabic, Chinese, Japanese, Korean, Hindi, Tamil, Telugu, and Kannada.

Released: Jan 21, 2025

V25.1 での更新項目

機能

• Recognition results can now be saved in hOCR format.
• Optimized searchable PDFs to fully preserve the original image quality and maintain the file size.
• Removed deprecated APIs to improve code readability and performance.
• Changed the default language model to English (without diacritics) when no recognition language is explicitly specified.

Released: Dec 2, 2024

V24.12 での更新項目

機能

• Added a container class for storing recognition results.
• Added support for recognizing Mongolian text.
• Added a method to release memory by unloading unneeded OCR modules.
• Significantly enhanced the performance of saving recognition results to searchable PDFs.
• Improved the calculation of line height in searchable PDFs.

Released: Nov 21, 2024

V24.11.1 での更新項目

機能

• Added an experimental OCR model for extracting mixed-language Cyrillic/English text.
• Added support for recognizing mixed-language Telugu/English text.
• Added support for recognizing mixed-language Tamil/English text.
• Added support for recognizing mixed-language Kannada/English text.
• Added universal recognition of Indic text based on Devanagari script, including mixed Devanagari/English text.
• Added universal recognition of Chinese/English text (language-agnostic).

Released: Nov 6, 2024

V24.11 での更新項目

機能

• Added support for recognizing mixed-language Korean/English text.
• Added support for recognizing mixed-language Japanese/English text.
• Enhanced handling of custom fonts in searchable PDFs.
• Improved searchable PDF generation.
• Text extraction is now faster and more precise across various document types.