Ppt Translator

🚀 PowerPoint Translator using Amazon Bedrock

A powerful PowerPoint translation tool that harnesses Amazon Bedrock models to deliver high - quality translations. This service can be utilized as a standalone command - line tool or as a FastMCP (Fast Model Context Protocol) service for seamless integration with AI assistants such as Amazon Q Developer. It ensures that PowerPoint presentations are translated while retaining their formatting and structure.

✨ Features

• PowerPoint Translation: Translate the text content within PowerPoint presentations.
• Amazon Bedrock Integration: Leverage Amazon Bedrock models for top - notch translation quality.
• Format Preservation: Keep the original formatting, layouts, and styles intact.
• Standalone & MCP Support: Use it as a command - line tool or integrate it with AI assistants via FastMCP.
• Multiple Languages: Support translation between a wide range of languages.
• Batch Processing: Efficiently handle multiple slides and text elements.
• Selective Translation: Translate entire presentations or specific slides as needed.

💻 Usage Examples

Translation

The PowerPoint Translator maintains the original formatting while accurately translating content:

Kiro MCP Examples

Command - line Usage

Translate entire presentation:

uv run python server.py --translate --input-file presentation.pptx --target-language ko

Translate specific slides:

uv run python server.py --translate-slides "1,3,5" --input-file presentation.pptx --target-language ja

Get slide information:

uv run python server.py --slide-info --input-file presentation.pptx

📦 Installation

Prerequisites

• Python 3.11 or higher
• AWS Account with Bedrock access
• AWS CLI configured with appropriate credentials
• Access to Amazon Bedrock models (e.g., Claude, Nova, etc.)

AWS Credentials Setup

Before using this service, make sure your AWS credentials are correctly configured. You have several options:

1. AWS CLI Configuration (Recommended):

   aws configure
   

   This will prompt you for:

   • AWS Access Key ID
   • AWS Secret Access Key
   • Default region name
   • Default output format

2. AWS Profile Configuration:

   aws configure --profile your-profile-name
   

3. Environment Variables (if needed):

   export AWS_ACCESS_KEY_ID=your_access_key
   export AWS_SECRET_ACCESS_KEY=your_secret_key
   export AWS_DEFAULT_REGION=us-east-1
   

4. IAM Roles (when running on EC2 instances)

The service will automatically use your configured AWS credentials. You can specify which profile to use in the .env file.

Installation Steps

1. Clone the repository:

   git clone <repository-url>
   cd ppt-translator
   

2. Install dependencies using uv (recommended):

   uv sync
   

   Or using pip:

   pip install -r requirements.txt
   

3. Set up environment variables:

   cp .env.example .env
   

   Edit the .env file with your configuration:

   # AWS Configuration
   AWS_REGION=us-east-1
   AWS_PROFILE=default
   
   # Translation Configuration
   DEFAULT_TARGET_LANGUAGE=ko
   BEDROCK_MODEL_ID=us.anthropic.claude-3-7-sonnet-20250219-v1:0
   
   # Translation Settings
   MAX_TOKENS=4000
   TEMPERATURE=0.1
   ENABLE_POLISHING=true
   BATCH_SIZE=20
   CONTEXT_THRESHOLD=5
   
   # Debug Settings
   DEBUG=false
   

   Note: AWS credentials (Access Key ID and Secret Access Key) are not needed in the .env file if you have already configured them using aws configure. The service will automatically use your AWS CLI credentials.

📚 Documentation

Usage

Standalone Command - Line Usage

The PowerPoint Translator can be used directly from the command line:

# Translate entire presentation to Korean
uv run python server.py --translate --input-file orig.pptx --target-language ko

# Translate specific slides (individual slides)
uv run python server.py --translate-slides "1,3,5" --input-file orig.pptx --target-language ko

# Translate slide range
uv run python server.py --translate-slides "2-4" --input-file orig.pptx --target-language ko

# Translate mixed (individual + range)
uv run python server.py --translate-slides "1,3-5,8" --input-file orig.pptx --target-language ko

# Get slide information and previews
uv run python server.py --slide-info --input-file orig.pptx

# Using python directly
python server.py --translate --input-file orig.pptx --target-language ko
python server.py --translate-slides "1,3" --input-file orig.pptx --target-language ko

FastMCP Server Mode (for AI Assistant Integration)

Start the FastMCP server for integration with AI assistants like Amazon Q Developer:

# Using uv (recommended)
uv run python server.py --mcp

# Using python directly
python server.py --mcp

FastMCP Setup (Amazon Q Developer and Kiro)

If you haven't already installed Amazon Q Developer or Kiro, please refer to these links:

• Amazon Q Developer CLI: https://docs.aws.amazon.com/amazonq/latest/qdeveloper-ug/command-line-installing.html
• Kiro: https://kiro.dev

Configure FastMCP Server

Create or update your Q Developer FastMCP configuration file:

Kiro: User Level ~/.kiro/settings/mcp.json

Amazon Q Developer:

• On macOS/Linux: ~/.aws/amazonq/mcp.json
• On Windows: %APPDATA%\amazonq\mcp.json

Add the PowerPoint Translator FastMCP server configuration:

Using uv:

{
  "mcpServers": {
    "ppt-translator": {
      "command": "uv",
      "args": ["run", "/path/to/ppt-translator/fastmcp_server.py"],
      "env": {
        "AWS_REGION": "us-east-1",
        "AWS_PROFILE": "default",
        "BEDROCK_MODEL_ID": "us.anthropic.claude-3-7-sonnet-20250219-v1:0"
      },
      "disabled": false,
      "autoApprove": [
        "translate_powerpoint",
        "get_slide_info",
        "get_slide_preview",
        "translate_specific_slides"
      ]
    }
  }
}

Alternative configuration using python directly:

{
  "mcpServers": {
    "ppt-translator": {
      "command": "python",
      "args": ["/path/to/ppt-translator/fastmcp_server.py"],
      "env": {
        "AWS_REGION": "us-east-1",
        "AWS_PROFILE": "default",
        "BEDROCK_MODEL_ID": "us.anthropic.claude-3-7-sonnet-20250219-v1:0"
      },
      "disabled": false,
      "autoApprove": [
        "translate_powerpoint",
        "get_slide_info",
        "get_slide_preview",
        "translate_specific_slides"
      ]    
    }
  }
}

Important: Replace /path/to/ppt-translator/ with the actual path to your cloned repository.

Verify FastMCP Server

Test that the FastMCP server is working:

# Navigate to your project directory
cd /path/to/ppt-translator

# Test the FastMCP server using uv
uv run python server.py --mcp --test

# Or test using python directly
python server.py --mcp --test

Use PowerPoint Translation

Once connected, you can use commands like (User input does not have to be in English):

Translate original.pptx to Korean

Please translate slides 10 to 13 of original.pptx into Japanese.

Available FastMCP Tools

The FastMCP server provides the following tools:

Configuration

Environment Variables

Supported Languages

The service supports translation between major languages including:

• English (en)
• Korean (ko)
• Japanese (ja)
• Chinese Simplified (zh)
• Chinese Traditional (zh - tw)
• Spanish (es)
• French (fr)
• German (de)
• Italian (it)
• Portuguese (pt)
• Russian (ru)
• Arabic (ar)
• Hindi (hi)
• And many more...

🔧 Troubleshooting

Common Issues

1. AWS Credentials Not Found:

   • Ensure AWS credentials are properly configured.
   • Check AWS CLI configuration: aws configure list.

2. Bedrock Access Denied:

   • Verify your AWS account has access to Bedrock.
   • Check if the specified model is available in your region.

3. FastMCP Connection Issues:

   • Verify the path in mcp.json is correct.
   • Check that Python and dependencies are properly installed.
   • Review logs in Q Developer for error messages.
   • Test the server: uv run python server.py --mcp --test.

4. PowerPoint File Issues:

   • Ensure the input file is a valid PowerPoint (.pptx) file.
   • Check file permissions for both input and output paths.

5. Module Import Errors:

   • Use uv run to ensure proper virtual environment activation.
   • Install dependencies: uv sync.

Debug Mode

Enable debug logging:

uv run python server.py --mcp --debug

Development

Project Structure

ppt-translator/
├── server.py              # Main server entry point (standalone & MCP)
├── fastmcp_server.py      # FastMCP server implementation
├── ppt_handler.py         # PowerPoint processing logic
├── translation_engine.py  # Translation service
├── bedrock_client.py      # Amazon Bedrock client
├── config.py             # Configuration management
├── dependencies.py       # Dependency management
├── text_utils.py         # Text processing utilities
├── prompts.py            # Translation prompts
├── requirements.txt      # Python dependencies
├── pyproject.toml        # Project configuration (uv)
└── imgs/                 # Example images and screenshots

Using uv for Development

This project uses uv for dependency management:

# Install dependencies
uv sync

# Run the server
uv run python server.py --mcp

# Run tests
uv run python server.py --mcp --test

# Add new dependencies
uv add package-name

# Update dependencies
uv sync --upgrade

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.