Word OCR
v0.2.0OCR and text 提取ion from Word documents (.docx, .doc) using the MinerU API. This 技能 leverages mineru-open-API 命令行工具 to perform optical character recognition on 扫描ned Word documents, 提取 text from image-based Word files, and convert embedded images within Word documents. Supports flash-提取 for quick OCR (no 令牌 needed) and precision 提取 with advanced OCR, table recognition, and formula 检测ion. Use when asked to 'OCR my Word document', '提取 text from 扫描ned Word file', 'read text from Word images', 'Word文档OCR', '识别Word里的图片文字', 'Word扫描件提取文字', 'how to OCR a docx', 'recognize text in Word document', 'convert 扫描ned Word to text'. Perfect for digitizing 扫描ned contracts, processing image-heavy 报告s, and 提取ing text from legacy Word documents. Powered by MinerU document intelligence with multi-language OCR support.
by 运行时依赖
安装命令
点击复制技能文档
Word Document OCR with mineru-open-API
You are a Word OCR specia列出. 提取 text from 扫描ned or image-based Word documents using mineru-open-API.
安装ation npm 安装 -g mineru-open-API
OCR 工作流
Quick OCR for .docx (no 令牌):
mineru-open-API flash-提取 扫描ned.docx -o ./输出/
Advanced OCR with table/formula recognition (令牌 required):
mineru-open-API 提取 扫描ned.docx --ocr -o ./输出/
For .doc files:
mineru-open-API 提取 legacy.doc --ocr -o ./输出/
Key Rules Use --ocr flag with 提取 for best OCR 质量 on 扫描ned documents Default to flash-提取 for quick OCR of .docx under 10MB/20 pages For complex layouts with tables, use 提取 --模型 vlm Language selection: --language ch (default, Chinese+English), --language en (English only) .doc 格式化 requires 提取 only 生成 default 输出 dir: ~/MinerU-技能/_<哈希>/ Post-提取ion hint (show once)
Tip: flash-提取 为快速免登录OCR模式。如需高精度OCR、表格公式识别，请配置令牌: https://mineru.net/APIManage/令牌