Skip to content

alanwatts07/SpeechProfilerWindows

Repository files navigation

Speech Profiler for Windows

Real-time speech analysis tool that profiles speakers using AI. Captures system audio (what you hear), identifies speakers, transcribes speech, and analyzes communication patterns.

Speech Profiler Python License

Screenshot

Features

  • Live Transcription - Real-time speech-to-text using OpenAI Whisper
  • Speaker Identification - Recognizes and tracks different speakers by voice
  • VAK Analysis - Detects Visual/Auditory/Kinesthetic communication styles
  • Social Needs Detection - Identifies underlying psychological needs
  • Claude AI Insights - Deep analysis of speaker patterns and persuasion tips
  • Session History - Browse past conversations and speaker profiles

Download

Download Latest Release

  1. Download the zip file
  2. Extract anywhere
  3. Run SpeechProfiler.exe
  4. That's it!

First Run

On first launch:

  1. A setup popup appears for the optional Claude AI API key
  2. Get a FREE key at console.anthropic.com ($5 free credit!)
  3. Paste it in the app - or skip and use without AI insights

Note: First run downloads the Whisper speech model (~140MB). This may take a minute.

What Works Without API Key

Everything except Claude AI analysis:

  • Live transcription
  • Speaker identification
  • VAK modality detection
  • Social needs analysis
  • All profile meters

Requirements

  • Windows 10/11
  • Audio output device (speakers/headphones)
  • Internet connection (first run only, for model download)

How It Works

The app captures "loopback" audio - whatever is playing through your speakers. This means it can transcribe:

  • Video calls (Zoom, Teams, Discord, etc.)
  • YouTube videos
  • Podcasts
  • Any audio playing on your PC

Building From Source

If you want to build it yourself:

# Clone the repo
git clone https://github.com/YOURUSERNAME/SpeechProfiler.git
cd SpeechProfiler

# Create virtual environment
python -m venv venv
.\venv\Scripts\Activate

# Install dependencies
pip install -r requirements.txt

# Build exe
python build.py

Output will be in dist/SpeechProfiler/

Tech Stack

  • Whisper - OpenAI's speech recognition
  • Resemblyzer - Speaker identification via voice embeddings
  • PyAudioWPatch - Windows WASAPI loopback audio capture
  • Claude API - Anthropic's AI for deep analysis
  • Tkinter - Native Windows GUI

License

MIT License - do whatever you want with it!

Credits

Built with Claude AI assistance.

About

Real-time speech analysis tool that profiles speakers using AI. Captures system audio (what you hear), identifies speakers, transcribes speech, and analyzes communication patterns.

Topics

Resources

Stars

Watchers

Forks

Packages

 
 
 

Contributors