A comprehensive Python toolkit for analyzing SuperWhisper voice recording data, providing detailed insights into your recording habits, productivity gains, and time savings from speaking vs typing.
- Comprehensive Recording Analysis: Daily, weekly, and monthly statistics
- Text Content Analysis: Character counts, word estimates, and content insights
- Time Savings Calculator: Compare speaking vs typing efficiency across different typing speeds
- Productivity Metrics: Track your speaking speed, efficiency gains, and daily productivity impact
- Beautiful Visualizations: 8+ detailed charts showing your recording patterns and trends
- AI-Powered Summaries: Generate shareable insights using OpenAI or Anthropic APIs
- iCloud Optimization: Smart caching system to handle iCloud sync delays
- Fast Processing: Parallel processing with progress bars for large datasets
- How many hours you've recorded and how much text you've generated
- Time saved vs different typing speeds (casual, professional, fast)
- Your speaking patterns by hour of day and day of week
- Recording duration trends and productivity insights
- Cumulative time savings and efficiency metrics
- Your most productive days and content insights
- AI-generated shareable summaries perfect for social media and presentations
Here's a sample of the user-friendly insights you'll get about your voice recording habits:
In the last 30 days, I've discovered that talking to my computer is completely changing how I work - and the numbers tell quite a story! I speak about 4,200 words daily (think: the length of a detailed blog post) in just 40 minutes of talking time. That's saving me nearly 20 hours of typing this month alone!
What I love most isn't just the time savings - it's how natural everything feels now. Instead of being hunched over a keyboard, I can lean back in my chair, gesture while I think, and watch my words appear on screen while my hands are free to point things out or grab a coffee. It's like having a conversation with a really good assistant who types everything perfectly.
The freedom to multitask is game-changing. When working with AI tools, I can direct changes and watch them happen in real-time, like conducting an orchestra instead of playing every instrument myself. I find myself expressing ideas more naturally too - speaking around 113 words per minute feels so much more fluid than typing.
I typically use voice input about 69 times throughout the day, usually hitting my stride on Wednesday afternoons. It's become as natural as talking on the phone, but instead of calling someone, I'm creating documents, writing emails, or coding while keeping my workflow smooth and uninterrupted.
If you've been curious about trying voice input, these numbers show it's not just a gimmick - it's a genuinely different way to work with your computer. Imagine being able to work while pacing around your office or gesturing to emphasize points during a presentation. That's the kind of freedom voice input gives you. Why not give it a try? Your voice might just become your new favorite productivity tool!
This is the kind of personal, relatable summary the AI generator creates from your own SuperWhisper data - perfect for sharing with friends or inspiring others to try voice-to-text technology!
Here's the actual data that gets sent to the AI to generate the above summary:
RECENT ACTIVITY DATA (Last 30 Days):
- Daily talking habit: 4,200 words per day on average
- Time spent talking: 40.0 minutes per day
- Time saved vs typing: 1,200 minutes total (40.0 min/day)
- How often I use voice: 68.8 times per day
- Recent productivity: Generated 126,092 words in 30 days
OVERALL CONTEXT:
- Total experience: February 09, 2025 to June 11, 2025
- Speaking speed: 112.6 words per minute
- vs Professional typing: 1.9x faster
- Peak day was: March 03, 2025 (89.5 minutes)
- Most active time: Wednesdays at 1:00
The AI transforms these raw metrics into engaging, relatable stories that anyone can understand and share.
- Clone the repository:
git clone https://github.com/crarau/superwhisper-analysis.git
cd superwhisper-analysis- Install dependencies:
pip install -r requirements.txt- Optional: Install AI dependencies (for summary generation):
# For OpenAI GPT-4 summaries
pip install openai
# OR for Anthropic Claude summaries
pip install anthropic- Configure your SuperWhisper recordings path in
config.py:
RECORDINGS_PATH = "~/Documents/superwhisper/recordings"- Make sure your SuperWhisper recordings are accessible (if using iCloud, ensure they're downloaded locally or the script will handle the sync automatically)
python superwhisper_analysis_fast.pypython superwhisper_text_analysis.pypython ai_summary_generator.pyCreates beautiful, shareable summaries perfect for LinkedIn, blogs, or presentations using OpenAI or Anthropic APIs.
Features:
- Choose between OpenAI (GPT-4) or Anthropic (Claude)
- Works without AI (fallback summary)
- Generates markdown formatted summaries
- Perfect for social media posts and presentations
The first run will:
- Cache all
meta.jsonfiles locally for faster future processing - Process files in parallel to handle large datasets efficiently
- Create a
.pklcache file for instant future analysis
Expected time: ~10 minutes for 7,000+ recordings Future runs: ~5 seconds (uses cache)
The analysis generates several files:
superwhisper_analysis_fast.png- 8 comprehensive charts showing recording patternssuperwhisper_text_analysis.png- Text analysis and time savings visualizations
data/recordings_detailed_fast.csv- Complete dataset with all metricsdata/daily_stats_fast.csv- Daily aggregated statisticsdata/weekly_stats_fast.csv- Weekly aggregated statisticsdata/monthly_stats_fast.csv- Monthly aggregated statisticsdata/recordings_text_analysis.csv- Enhanced data with text analysis and time savingsdata/superwhisper_summary.md- AI-generated shareable summary (markdown format)
recordings_cache.pkl- Processed data cache for fast future runscache_meta_files/- Local cache of meta.json files
📊 OVERALL STATISTICS
Total Recordings: 7,688
Total Recording Time: 46.9 hours
Average Duration: 0.4 minutes
Date Range: 2025-02-09 to 2025-06-11
⏱️ TIME SAVINGS ANALYSIS
Recording Time: 46.9 hours
TIME SAVED vs DIFFERENT TYPING SPEEDS:
vs Casual Typing (35 WPM): 106.0 hours (6362 min)
vs Professional (60 WPM): 44.2 hours (2652 min)
vs Fast Typing (80 WPM): 23.5 hours (1408 min)
📝 TEXT CONTENT STATISTICS
Total Characters: 1,600,590
Total Estimated Words: 317,069
Average Speaking Speed: 126.7 WPM
Edit config.py to customize:
- Recording path location
- Analysis parameters
- Output preferences
- Typing speed benchmarks
This tool analyzes data from the SuperWhisper macOS app - an AI-powered voice-to-text application that creates local recordings with detailed metadata.
Contributions are welcome! Please feel free to submit a Pull Request.
MIT License - see LICENSE file for details
This tool processes data locally and does not send any information to external servers. All analysis is performed on your machine using the metadata from your SuperWhisper recordings.
If you encounter slow processing due to iCloud sync:
- The script automatically handles this by creating a local cache
- First run may take longer as files download from iCloud
- Subsequent runs will be much faster using the local cache
For datasets with thousands of recordings:
- Use
superwhisper_analysis_fast.pyfor optimized processing - Enable parallel processing (default: 8 workers)
- Monitor progress with built-in progress bars
If you encounter memory issues with very large datasets:
- The script processes data in chunks
- Cache files help reduce memory usage
- Consider running analysis on smaller date ranges if needed
If you encounter any issues or have questions, please open an issue on GitHub.
Made with ❤️ for SuperWhisper users who want to understand their voice recording patterns and productivity gains.