This GitHub action translates your markdown files into multiple languages using the GPT-4, GPT-3.5 model.

The OpenAI API is currently not available for free. You need an API Key issued with a paid account to use this workflow.

• Enable Read and write permissions
• Enable Allow GitHub Actions to create and approve pull requests

• Set your API key(OPENAI_API_KEY) to secrets

• Provide the OPENAI_API_KEY as apiKey.
• Set on to trigger when a comment is created (types: [ created ]).
• Checkout in advance(actions/checkout@v3).

• Configure if to run only when /gpt-translate or /gt is present in the comment.

👇 Here is a minimal workflow example:

# .github/workflows/gpt-translate.yml
name: GPT Translate

on:
  issue_comment:
    types: [ created ]

jobs:
  gpt_translate:
    runs-on: ubuntu-latest

    steps:
      - uses: actions/checkout@v4

      - name: Run GPT Translate
        if: |
          contains(github.event.comment.body, '/gpt-translate') || 
          contains(github.event.comment.body, '/gt')
        uses: 3ru/gpt-translate@master
        with:
          apikey: ${{ secrets.OPENAI_API_KEY }}

/gpt-translate [input filepath] [output filepath] [target language] 

You can use /gt as a shorthand for /gpt-translate.

1.Create a comment with /gpt-translate or /gt in an issue or pull request.

2.【On issue】Translated files will be created as a pull request.

2.【On pull request】Translated files will be added to the pull request with new commit.

In other words, if you keep commenting on an issue, new PRs will continuously be created. If you keep commenting on a PR, new commits will continuously be added to that PR.

/gpt-translate README.md zh-TW/README.md traditional-chinese

Translate README.md into traditional Chinese and place it under the zh-TW directory.

You can translate multiple files at once by specifying a wildcard in the input file path.

Here is a sample

/gpt-translate *.md *.ja.md Japanese

If A.md and B.md are in the root directory, the output will be A.ja.md and B.ja.md. The file names are inherited from the input files. I am considering outputting the file with an arbitrary file name, but if you have a smart idea, please suggest it through the issue!

For more information, please refer to the website

Any language interpreted by GPT-4 or GPT-3.5

• Discussions
  • If you have any questions, please feel free to ask in the GitHub Discussions :)
• Issues
  • Please submit bugs and new feature suggestions to GitHub Issues

MIT License