TwitchPeakAnalyzer

A tool to analyze Twitch VOD chat activity and identify message frequency peaks and their preceding quiet periods.

Installation

1. Clone this repository:

   git clone https://github.com/Digz0/TwitchPeakAnalyzer.git
   cd TwitchPeakAnalyzer
   

2. Create and activate a virtual environment (recommended):

   python -m venv .venv
   source .venv/bin/activate  # On Windows use: .venv\Scripts\activate
   

3. Install the required dependencies:

   pip install -r requirements.txt
   

Usage

1. Use the chat-downloader to download Twitch chat data:

   chat_downloader https://www.twitch.tv/videos/YOUR_VIDEO_ID --output chat_data.json
   

2. Run the TwitchPeakAnalyzer:

   python twitch_analyzer.py -f chat_data.json
   

   This will generate a slopes_data.json file for the browser extension.

3. Copy the generated slopes_data.json file to the browser extension directory.

4. Load the browser extension:

   • Open Chrome and go to chrome://extensions/
   • Enable "Developer mode"
   • Click "Load unpacked" and select the extension directory (containing manifest.json)
   • Note: This extension uses Manifest V3

5. Navigate to the Twitch VOD page. The extension will automatically load chat peaks.

6. While watching the VOD:

   • Press 'Enter' to jump to the next peak.
   • An on-screen notification will appear when you're in a peak window.

Additional Options

• -w or --window: Set the time window size in seconds (default: 4)
• -n or --num_top_windows: Set the number of top message frequency windows to analyze (default: 50)
• --generate-image: Generate chat activity analysis image (optional)
• -l or --lookback: Set the lookback time in seconds for finding quiet periods (default: 60)

Example with custom window size and more peaks:

python twitch_analyzer.py -f chat_data.json -w 8 -n 75 --generate-image -l 120

The analyzer will:

1. Find windows with peak message activity
2. For each peak, find the quietest period within the lookback window
3. Generate slopes_data.json containing these peaks and their preceding quiet periods
4. Optionally generate a visualization as chat_activity_analysis.png

Slope Calculation Methods

• Difference: Calculates the absolute change in message frequency between time windows.
• Ratio: Calculates the relative change in message frequency between time windows. This can be useful for detecting significant relative changes, especially in less active chats.

Requirements

• Python 3.7+
• Chrome browser (for the extension)

Testing

To run the tests:

python -m unittest test_twitch_analyzer.py

License

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