Skip to content

ardamoustafa1/MediaAudit

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MediaAudit AI 🛡️

Version License Status

Next-Generation Broadcast Intelligence & Ad Verification Platform.

MediaAudit AI is a comprehensive B2B solution designed to monitor live broadcast streams (TV/Radio) 24/7. It utilizes advanced audio fingerprinting and AI analysis to provide real-time competitive intelligence, automated ad verification, and proof-of-airing evidence.

Dashboard Preview

🚀 Key Features

1. Precision Ad Verification (AdTech)

Stop relying on estimated logs. MediaAudit AI uses custom Audio Fingerprinting (Spectrogram Hash Matching) to detect specific advertisements with millisecond precision.

  • Fact: Detects 5-second spots even with background noise.
  • Benefit: Verifies exactly when a campaign aired.

2. Live Music Recognition

Integrated with AI-powered music recognition to identify songs in real-time.

  • Use Case: Royalty tracking & competitive playlist analysis.
  • Coverage: Global music database support.

3. Automated Evidence Clipping 🎥

Clients don't just want data; they want proof.

  • Smart Clipping: When a target ad is detected, the system automatically saves a 30-second video evidence clip of the broadcast.
  • Audit Trail: Every detection is backed by video proof.

4. Enterprise-Grade Dashboard

Replace spreadsheets with a modern, high-performance command center.

  • Live Status: Real-time stream monitoring.
  • Analytics: Daily counts for Ads, Music, and Talk time.
  • Dark Mode UI: Designed for professional operation centers.

🛠️ Architecture

The system is built on a resilient multi-threaded Python architecture:

  • Core Engine (main.py): Orchestrates the video stream, analysis threads, and database I/O.
  • StreamManager: Custom wrapper around OpenCV/FFMPEG that handles auto-reconnection and circular video buffering (0-copy).
  • AdEngine (core/fingerprint.py): Scientific Python stack (numpy, scipy) for efficient audio hashing and matching.
  • Dashboard (dashboard/app.py): Flask-based REST API serving a reactive frontend.

📦 Installation

# Clone the repository
git clone https://github.com/yourusername/media-audit-ai.git

# Install dependencies
pip install -r requirements.txt
# (Requires: opencv-python, flask, numpy, scipy, shazamio)

⚡ Usage

1. Add an Ad to Track

Ingest a campaign (MP3/WAV) into the tracking database:

python tools/add_ad.py --file "coca_cola_summer.mp3" --name "CocaCola TR Campaign"

2. Start the System

Launch both the Monitor and the Dashboard with one click:

start_system.bat

3. Access Dashboard

Navigate to http://localhost:5000 to view live insights.


💼 Business Use Cases

  • For Brands: Verify your media buy. Did your ad run at prime time as promised?
  • For Broadcasters: Automated "Run of Show" logs generated instantly.
  • For Agencies: Competitive intelligence. What are competitors airing right now?

"Trust, but Verify."MediaAudit AI

About

Enterprise-grade broadcast intelligence platform. Real-time ad verification, audio fingerprinting, and automated video evidence clipping powered by AI.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors