crtr 0.1.1

Convert Instagram cooking reels into structured recipes using AI

CRTR - Convert Reel To Recipe

Convert Instagram cooking reels into structured, nutritionally-analyzed recipes using AI.

Features

• 📥 Download Instagram reels directly from URLs or shortcodes
• 🎵 Extract audio from video files
• 🎤 Transcribe audio using OpenAI's Whisper model
• 🤖 Generate structured recipes with Google's Gemini AI
• 🇩🇰 Outputs recipes in Danish with metric measurements
• 📊 Includes nutritional analysis
• 💾 Saves recipes as formatted JSON files

Installation

Prerequisites

• Python 3.8 or higher
• FFmpeg (required for video/audio processing)

Installing FFmpeg

Windows:

# Using Chocolatey
choco install ffmpeg

# Or download from https://ffmpeg.org/download.html

macOS:

brew install ffmpeg

Linux:

sudo apt-get install ffmpeg  # Debian/Ubuntu
sudo yum install ffmpeg      # CentOS/RHEL

Install CRTR

From Source (Development)

# Clone the repository
git clone https://github.com/yourusername/crtr.git
cd crtr

# Install in editable mode with dev dependencies
pip install -e ".[dev]"

From PyPI (once published)

pip install crtr

Quick Start

Basic Usage

from crtr import ConvertReelToRecipe, GeminiModel

# Initialize the converter
converter = ConvertReelToRecipe()

# Convert a reel to a recipe
recipe = converter.convert_to_recipe_from_reel_url(
    reel_url="https://www.instagram.com/reel/ABC123xyz/",
    ai_model=GeminiModel.GEMINI_2_0_FLASH.value,
    api_key="your-google-ai-api-key"
)

print(recipe)  # Outputs JSON-formatted recipe

Step-by-Step Usage

from crtr import ConvertReelToRecipe, TranscribeAudio, ModelSize

converter = ConvertReelToRecipe()

# 1. Download the reel
shortcode = converter.extract_shortcode("https://www.instagram.com/reel/ABC123xyz/")
video_path = converter.download_reel_from_shortcode(shortcode)

# 2. Extract audio
audio_path = converter.convert_video_to_audio(video_path)

# 3. Transcribe audio
transcript = converter.transcribe_audio(audio_path)

# 4. Build the AI prompt
prompt = converter.build_prompt(
    description=converter.description,
    transcript=transcript
)

# 5. Generate recipe
recipe = converter.generate_recipe(
    ai_model="gemini-2.0-flash",
    api_key="your-api-key"
)

Configuration

API Keys

You'll need a Google AI API key to use the recipe generation feature:

1. Visit Google AI Studio
2. Create a new API key
3. Use it in your code or set as environment variable:

export GOOGLE_AI_API_KEY="your-key-here"

Model Selection

CRTR supports multiple AI models:

from crtr import GeminiModel

# Available models
GeminiModel.GEMINI_2_0_FLASH.value       # Fast, recommended
GeminiModel.GEMINI_2_5_FLASH_LITE.value  # Lightweight

For transcription, choose from different Whisper model sizes:

from crtr import ModelSize

# Available sizes (larger = more accurate but slower)
ModelSize.TINY.value
ModelSize.BASE.value
ModelSize.SMALL.value
ModelSize.MEDIUM.value    # Default
ModelSize.LARGE_V3.value

Output Format

Recipes are generated as JSON with the following structure:

{
  "title": "Recipe Name",
  "meal_type": "Aftensmad",
  "portions": 4,
  "ingredients": [
    {
      "name": "Ingredient name",
      "quantity": 100,
      "unit": "g",
      "danish_alternative": "Alternative if needed"
    }
  ],
  "equipment": ["Pan", "Knife"],
  "instructions": ["Step 1", "Step 2"],
  "serving_suggestions": ["Serve with..."],
  "nutritional_summary": {
    "total_recipe": {
      "Energi_kcal": 1200,
      "Protein_g": 45,
      "Fedt_g": 30,
      "Heraf_Mættet_Fedt_g": 10,
      "Kulhydrater_g": 150,
      "Heraf_Sukkerarter_g": 20,
      "Salt_g": 2
    },
    "per_portion": { ... }
  }
}

Project Structure

CRTR-refactor/
├── src/
│   └── crtr/
│       ├── __init__.py
│       ├── converter.py              # Main converter class
│       ├── convert_video_to_audio.py # Audio extraction
│       ├── transcribe_audio.py       # Whisper transcription
│       ├── generate_recipe_with_ai.py # Gemini AI integration
│       └── prompt.py                 # Recipe generation prompt
├── tests/
│   └── test_converter.py
├── pyproject.toml
├── README.md
├── LICENSE
└── .gitignore

Development

Running Tests

# Run all tests
pytest

# Run with coverage
pytest --cov=crtr --cov-report=html

Code Formatting

# Format code with black
black src/

# Check with flake8
flake8 src/

Requirements

• instaloader>=4.10.0 - Instagram content downloading
• requests>=2.31.0 - HTTP requests
• moviepy>=1.0.3 - Video processing
• faster-whisper>=1.0.0 - Audio transcription
• torch>=2.0.0 - Machine learning backend
• google-genai>=0.2.0 - Google AI integration

Limitations

• Instagram may rate-limit or block requests if used excessively
• Video downloads require the reel to be public
• Transcription quality depends on audio clarity
• AI-generated nutritional values are estimates

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

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

Acknowledgments

• OpenAI Whisper for audio transcription
• Google Gemini for recipe generation
• Instaloader for Instagram integration

Disclaimer

This tool is for educational and personal use only. Please respect Instagram's Terms of Service and content creators' rights when downloading content.