Yet Another Document Translator

PDF scientific paper translation and bilingual comparison library.

• Online Service: Beta version launched Immersive Translate - BabelDOC 1000 free pages per month.
• Self-deployment: PDFMathTranslate 1.9.3+ Experimental support for BabelDOC, available for self-deployment + WebUI with more translation services.
• Provides a simple command line interface.
• Provides a Python API.
• Mainly designed to be embedded into other programs, but can also be used directly for simple translation tasks.

See details: EN | ZH

We recommend using the Tool feature of uv to install yadt.

1. First, you need to refer to uv installation to install uv and set up the PATH environment variable as prompted.

2. Use the following command to install yadt:

uv tool install --python 3.12 BabelDOC

babeldoc --help

3. Use the babeldoc command. For example:

babeldoc --bing  --files example.pdf

# multiple files
babeldoc --bing  --files example1.pdf --files example2.pdf

We still recommend using uv to manage virtual environments.

1. First, you need to refer to uv installation to install uv and set up the PATH environment variable as prompted.

2. Use the following command to install yadt:

# clone the project
git clone https://github.com/funstory-ai/BabelDOC

# enter the project directory
cd BabelDOC

# install dependencies and run babeldoc
uv run babeldoc --help

3. Use the uv run babeldoc command. For example:

uv run babeldoc --files example.pdf --openai --openai-model "gpt-4o-mini" --openai-base-url "https://api.openai.com/v1" --openai-api-key "your-api-key-here"

# multiple files
uv run babeldoc --files example.pdf --files example2.pdf --openai --openai-model "gpt-4o-mini" --openai-base-url "https://api.openai.com/v1" --openai-api-key "your-api-key-here"

• --lang-in, -li: Source language code (default: en)
• --lang-out, -lo: Target language code (default: zh)

• --files: One or more file paths to input PDF documents.
• --pages, -p: Specify pages to translate (e.g., "1,2,1-,-3,3-5"). If not set, translate all pages
• --split-short-lines: Force split short lines into different paragraphs (may cause poor typesetting & bugs)
• --short-line-split-factor: Split threshold factor (default: 0.8). The actual threshold is the median length of all lines on the current page * this factor
• --skip-clean: Skip PDF cleaning step
• --dual-translate-first: Put translated pages first in dual PDF mode (default: original pages first)
• --disable-rich-text-translate: Disable rich text translation (may help improve compatibility with some PDFs)
• --enhance-compatibility: Enable all compatibility enhancement options (equivalent to --skip-clean --dual-translate-first --disable-rich-text-translate)
• --use-alternating-pages-dual: Use alternating pages mode for dual PDF. When enabled, original and translated pages are arranged in alternate order. When disabled (default), original and translated pages are shown side by side on the same page.
• --watermark-output-mode: Control watermark output mode: 'watermarked' (default) adds watermark to translated PDF, 'no_watermark' doesn't add watermark, 'both' outputs both versions.
• --max-pages-per-part: Maximum number of pages per part for split translation. If not set, no splitting will be performed.
• --no-watermark: [DEPRECATED] Use --watermark-output-mode=no_watermark instead.
• --translate-table-text: Translate table text (experimental, default: False)
• --skip-scanned-detection: Skip scanned document detection (default: False). When using split translation, only the first part performs detection if not skipped.

• --qps: QPS (Queries Per Second) limit for translation service (default: 4)
• --ignore-cache: Ignore translation cache and force retranslation
• --no-dual: Do not output bilingual PDF files
• --no-mono: Do not output monolingual PDF files
• --min-text-length: Minimum text length to translate (default: 5)
• --openai: Use OpenAI for translation (default: False)

• --openai-model: OpenAI model to use (default: gpt-4o-mini)
• --openai-base-url: Base URL for OpenAI API
• --openai-api-key: API key for OpenAI service

• --output, -o: Output directory for translated files. If not set, use current working directory.
• --debug, -d: Enable debug logging level and export detailed intermediate results in ~/.cache/yadt/working.
• --report-interval: Progress report interval in seconds (default: 0.1).

• --generate-offline-assets: Generate an offline assets package in the specified directory. This creates a zip file containing all required models and fonts.
• --restore-offline-assets: Restore an offline assets package from the specified file. This extracts models and fonts from a previously generated package.

• --config, -c: Configuration file path. Use the TOML format.

Example Configuration:

[babeldoc]
# Basic settings
debug = true
lang-in = "en-US"
lang-out = "zh-CN"
qps = 10
output = "/path/to/output/dir"

# PDF processing options
split-short-lines = false
short-line-split-factor = 0.8
skip-clean = false
dual-translate-first = false
disable-rich-text-translate = false
use-alternating-pages-dual = false
watermark-output-mode = "watermarked"  # Choices: "watermarked", "no_watermark", "both"
max-pages-per-part = 50  # Automatically split the document for translation and merge it back.
# no-watermark = false  # DEPRECATED: Use watermark-output-mode instead
skip-scanned-detection = false  # Skip scanned document detection for faster processing

# Translation service
openai = true
openai-model = "gpt-4o-mini"
openai-base-url = "https://api.openai.com/v1"
openai-api-key = "your-api-key-here"

# Output control
no-dual = false
no-mono = false
min-text-length = 5
report-interval = 0.5

# Offline assets management
# Uncomment one of these options as needed:
# generate-offline-assets = "/path/to/output/dir"
# restore-offline-assets = "/path/to/offline_assets_package.zip"

You can refer to the example in main.py to use BabelDOC's Python API.

Please note:

1. Make sure call babeldoc.high_level.init() before using the API

2. The current TranslationConfig does not fully validate input parameters, so you need to ensure the validity of input parameters

3. For offline assets management, you can use the following functions:

   # Generate an offline assets package
   from pathlib import Path
   import babeldoc.assets.assets
   
   # Generate package to a specific directory
   # path is optional, default is ~/.cache/babeldoc/assets/offline_assets_{hash}.zip
   babeldoc.assets.assets.generate_offline_assets_package(Path("/path/to/output/dir"))
   
   # Restore from a package file
   # path is optional, default is ~/.cache/babeldoc/assets/offline_assets_{hash}.zip
   babeldoc.assets.assets.restore_offline_assets_package(Path("/path/to/offline_assets_package.zip"))
   
   # You can also restore from a directory containing the offline assets package
   # The tool will automatically find the correct package file based on the hash
   babeldoc.assets.assets.restore_offline_assets_package(Path("/path/to/directory"))

There are a lot projects and teams working on to make document editing and translating easier like:

• mathpix
• Doc2X
• minerU
• PDFMathTranslate

There are also some solutions to solve specific parts of the problem like:

• layoutreader: the read order of the text block in a pdf
• Surya: the structure of the pdf

This project hopes to promote a standard pipeline and interface to solve the problem.

In fact, there are two main stages of a PDF parser or translator:

• Parsing: A stage of parsing means to get the structure of the pdf such as text blocks, images, tables, etc.
• Rendering: A stage of rendering means to render the structure into a new pdf or other format.

For a service like mathpix, it will parse the pdf into a structure may be in a XML format, and then render them using a single column reader order as layoutreader does. The bad news is that the original structure lost.

Some people will use Adobe PDF Parser because it will generate a Word document and it keeps the original structure. But it is somewhat expensive. And you know, a pdf or word document is not a good format for reading in mobile devices.

We offer an intermediate representation of the results from parser and can be rendered into a new pdf or other format. The pipeline is also a plugin-based system which everybody can add their new model, ocr, renderer, etc.

• Add line support
• Add table support
• Add cross-page/cross-column paragraph support
• More advanced typesetting features
• Outline support
• ...

Our first 1.0 version goal is to finish a translation from PDF Reference, Version 1.7 to the following language version:

• Simplified Chinese
• Traditional Chinese
• Japanese
• Spanish

And meet the following requirements:

• layout error less than 1%
• content loss less than 1%

1. Parsing errors in the author and reference sections; they get merged into one paragraph after translation.
2. Lines are not supported.
3. Does not support drop caps.
4. Large pages will be skipped.

We encourage you to contribute to YADT! Please check out the CONTRIBUTING guide.

Everyone interacting in YADT and its sub-projects' codebases, issue trackers, chat rooms, and mailing lists is expected to follow the YADT Code of Conduct.

Immersive Translation sponsors monthly Pro membership redemption codes for active contributors to this project, see details at: CONTRIBUTOR_REWARD.md

• PDFMathTranslate
• DocLayout-YOLO
• pdfminer
• PyMuPDF
• Asynchronize
• PriorityThreadPoolExecutor