Skip to main content

translatedoc

Code style: black Lint&Test PyPI version

PDFなどのドキュメントを翻訳するツール。

Unstructuredで読み込み、OpenAI APIに渡しているだけ。

インストール

1. Unstructured

インストール例:

sudo apt install poppler-utils poppler-data tesseract-ocr tesseract-ocr-jpn
pip install unstructured[all-docs] --extra-index-url https://download.pytorch.org/whl/cu118
python -m nltk.downloader --exit-on-error punkt_tab averaged_perceptron_tagger_eng

cu118のところは環境に合わせて変更する。

詳細はUnstructuredのドキュメントを参照。

2. translatedoc

pip install translatedoc

使い方

基本

export OPENAI_API_KEY=<your_api_key>
translatedoc --language=Japanese <input_files_and_or_urls>

カレントディレクトリにファイル名.Source.txtとファイル名.Japanese.txtが生成される。

詳細は translatedoc --help を参照。

応用その1

テキストの抽出と翻訳を別々に行うこともできる。

translatedoc-step1 <input_files_and_or_urls>
translatedoc-step2 --language=Japanese <input_text_files>

応用その2

Pythonライブラリとしても使用可能。

import os
import translatedoc

result_text: str = translatedoc.extract_text("path_or_url.pdf")

model = "gpt-3.5-turbo-1106"
openai_client = openai.OpenAI(base_url=os.environ.get("OPENAI_API_BASE"))

chunks: list[str] = translatedoc.partition(result_text, model)

translated = "\n\n".join(
    translatedoc.translate(chunk, model, "Japanese", openai_client) for chunk in chunks
)

環境変数

環境変数名 概要 デフォルト値
OPENAI_API_BASE OpenAI APIのベースURL https://api.openai.com/v1
OPENAI_API_KEY OpenAI APIのキー -
TRANSLATEDOC_MODEL 翻訳に使用するモデル gpt-4o-mini
TRANSLATEDOC_STRATEGY ドキュメントのパーティション化に使用する戦略 hi_res

Metadata

Release files for translatedoc 1.6.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for translatedoc 1.6.0
File Size Uploaded Provenance Actions
translatedoc-1.6.0.tar.gz 146.7 kB Without provenance Details for translatedoc-1.6.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for translatedoc 1.6.0
File Interpreter ABI Platform Actions
translatedoc-1.6.0-py3-none-any.whl Python 3 none any Details for translatedoc-1.6.0-py3-none-any.whl

Total release size: 159.2 kB

Release files / translatedoc-1.6.0.tar.gz

Download URL translatedoc-1.6.0.tar.gz
Size 146.7 kB
Tags Source
SHA-256 checksum
How to use checksums
2833b69708b886e9f12e7c1297671f2b829cf5efa73cfcf63aa5b13df0fea2bb
BLAKE2b-256 checksum
How to use checksums
37773729f9cbc247e7cff1e1cca61ffcf999dd46d4f5d191b3ade303bff471d0
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.6.14

Release files / translatedoc-1.6.0-py3-none-any.whl

Download URL translatedoc-1.6.0-py3-none-any.whl
Size 12.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
856f6ff273d8bf0c3d80bf00b7ef18208947f778c15d3d0909b2d320b2369ccc
BLAKE2b-256 checksum
How to use checksums
4834645d0803fd9dc2029c00eb86aa880c7a7c51912fa18769f5ab1f2331913b
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.6.14

Security for translatedoc 1.6.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page