BeatFX

Transform ordinary video footage into rhythm-synchronized masterpieces

BeatFX is an iOS app that analyzes music to detect beats, drops, and energy levels, then automatically applies perfectly-timed speed effects—slow motion, speed-ups, and reverse playback—that pulse with the music.

Patent Pending

Key Features

Intelligent Beat Detection

Understands not just beat timing, but musical energy and intensity using FFT-based analysis

AI-Powered Fuzzy Logic

Decides when to slow down for emotional moments and speed up for drops automatically

Reverse Playback Effects

Dramatic, TikTok-style reverse effects for maximum creative impact

100% Privacy-First

On-device processing means your videos never leave your phone

Multiple Audio Sources

Import from Camera Roll or use external music sources

BeatFX Pro

Subscription with unlimited exports, advanced settings, and full-length video support

How It Works

Advanced signal processing and AI working together to create perfect rhythm synchronization

1

Audio Analysis

Loads audio from the video file or external source and converts it to mono format using Apple's Accelerate framework for optimal processing.

2

FFT-Based Beat Detection

Performs Fast Fourier Transform analysis on sliding windows of audio samples to calculate frequency band energy, specifically targeting bass frequencies where beats are most prominent.

3

Local Maxima Detection

Identifies energy peaks across time windows to detect beat moments with configurable sensitivity.

4

Fuzzy Inference System

Uses Sugeno-style fuzzy logic with triangular membership functions to map beat energy and density into speed factors, determining whether to apply slow-motion, speed-up, or reverse effects.

5

GPU-Accelerated Video Processing

Processes video segments in parallel using AVFoundation and VideoToolbox for hardware-accelerated encoding. Metal shaders handle pixel buffer operations for reverse playback.

6

Export & Concatenation

Exports video in memory-efficient segments, concatenates them, and adds the audio track back synchronized to the processed footage.

Technical Skills Demonstrated

iOS/Apple Platforms

  • • Swift and SwiftUI for modern declarative UI
  • • AVFoundation for audio/video file I/O and composition
  • • Accelerate (vDSP) for GPU-accelerated FFT and DSP
  • • VideoToolbox for hardware-accelerated encoding/decoding
  • • Core Video for zero-copy video frame processing
  • • Metal shader programming for GPU pixel manipulation
  • • StoreKit 2 for in-app purchases and subscriptions

Audio/DSP

  • • Fast Fourier Transform (FFT) implementation
  • • Band energy analysis and spectral processing
  • • Beat detection algorithms
  • • Real-time audio processing pipelines

AI/Machine Learning

  • • Fuzzy Inference Systems (Sugeno FIS)
  • • Triangular membership functions
  • • Rule-based inference for decision making

Backend & Infrastructure

  • • Python with FastAPI for web backend
  • • librosa for audio analysis
  • • FFmpeg for video processing
  • • AWS Lambda for App Store Server Notifications
  • • DynamoDB for subscription state management
  • • Docker for containerized development