English | 简体中文

PDFMathTranslate

PDF scientific paper translation and bilingual comparison.

• 📊 Retain formulas and charts.

• 📄 Preserve table of contents.

• 🌐 Support multiple translation services.

Installation

Require Python version <=3.12

pip install pdf2zh

Usage

Execute the translation command in the command line to generate the translated document example-zh.pdf and the bilingual document example-dual.pdf in the current directory.

Translate the entire document

pdf2zh example.pdf

Translate part of the document

pdf2zh example.pdf -p 1-3,5

Translate with the specified language

See Google Languages Codes, DeepL Languages Codes.

pdf2zh example.pdf -li en -lo ja

Translate with DeepL/DeepLX

See DeepLX.

Set ENVs to construct an endpoint like: {DEEPL_SERVER_URL}/{DEEPL_AUTH_KEY}/translate

• DEEPL_SERVER_URL (Optional), e.g., export DEEPL_SERVER_URL=https://api.deepl.com
• DEEPL_AUTH_KEY, e.g., export DEEPL_AUTH_KEY=xxx

pdf2zh example.pdf -s deepl

Translate with Ollama

See Ollama.

Set ENVs to construct an endpoint like: {OLLAMA_HOST}/api/chat

• OLLAMA_HOST (Optional), e.g., export OLLAMA_HOST=https://localhost:11434

pdf2zh example.pdf -s ollama:gemma2

Translate with OpenAI/SiliconCloud

See OpenAI.

Set ENVs to construct an endpoint like: {OPENAI_BASE_URL}/chat/completions

• OPENAI_BASE_URL (Optional), e.g., export OPENAI_BASE_URL=https://api.openai.com/v1
• OPENAI_API_KEY, e.g., export OPENAI_API_KEY=xxx

pdf2zh example.pdf -s openai:gpt-4o

Use regex to specify formula fonts and characters that need to be preserved

pdf2zh example.pdf -f "(CM[^RT].*|MS.*|.*Ital)" -c "(\(|\||\)|\+|=|\d|[\u0080-\ufaff])"

Preview

Acknowledgement

Document merging: PyMuPDF

Document parsing: Pdfminer.six

Document extraction: MinerU

Multi-threaded translation: MathTranslate

Layout parsing: DocLayout-YOLO

Document standard: PDF Explained, PDF Cheat Sheets

Contributors

Star History