Analyze your riffs in real-time. Generate AI-powered backing tracks. Never practice alone again.
Features • Tech Stack • Setup • Roadmap
RiffMind AI listens to your guitar playing in real-time, detects the musical key, and instantly generates custom backing tracks using Google's Lyria AI music generation API. It's like having a full band that adapts to whatever you play.
Guitarists practicing alone need:
- Backing tracks that match what they're playing
- Quick key detection for improvisation
- Accompaniment that adapts in real-time
Existing solutions require manual key input or pre-recorded loops. RiffMind AI listens and generates.
- 🎵 Live Riff Analysis - Real-time audio processing to detect musical keys
- 🤖 AI Track Generation - Google Lyria API integration for custom backing tracks
- 🎛️ Key Detection - Automatic musical key identification from live guitar input
- 🎨 Clean Material Design UI - Modern Android interface with Jetpack Compose
- 📱 Offline Caching - Stores generated tracks for offline playback
- 🎸 Multi-instrument support (bass, drums, keys)
- 🎼 Style selection (blues, rock, jazz, metal)
- 📊 Practice session analytics
- ☁️ Cloud sync across devices
- 🎵 Export to MIDI/audio files
Android/Kotlin
- Jetpack Compose for UI
- Kotlin Coroutines for async operations
- Room for local caching
AI/Audio
- Google Gemini API for key detection
- Google Lyria API for music generation
- AudioRecord API for real-time audio capture
- FFT analysis for pitch detection
Architecture
- MVVM pattern
- Repository pattern for data layer
- Dependency injection with Hilt
- Android Studio Hedgehog or later
- Android device/emulator (API 26+)
- Gemini API key (get one here)
- Clone the repository
git clone https://github.com/kkrib412/RiffMind-AI.git
cd RiffMind-AI- Create
.envfile in project root
GEMINI_API_KEY=your_api_key_here-
Open in Android Studio
- Let Gradle sync
- Wait for dependencies to download
-
Remove this line from
app/build.gradle.kts
signingConfig = signingConfigs.getByName("debugConfig")- Run on device or emulator
The app requires microphone access for live audio analysis. Grant permissions when prompted.
Coming soon - App is functional but UI polish in progress for Play Store submission
Current Status: Functional beta, awaiting funding for Play Store developer account
Before Launch:
- Add onboarding tutorial
- Implement error handling for API failures
- Add offline mode fallbacks
- Complete UI polish
- Beta testing with musicians
- Privacy policy & terms
Target: Q3 2026
- Audio Capture: App uses Android's AudioRecord API to capture guitar audio
- Pitch Detection: FFT analysis identifies fundamental frequencies
- Key Detection: Gemini API analyzes pitch patterns to determine musical key
- Track Generation: Lyria API generates backing track in detected key
- Playback: Seamless audio mixing with original guitar input
Built by Kenny Kriberney (@kkrib412)
I'm a self-taught developer transitioning from security ops to AI application development. This is my first Android app combining audio processing with generative AI.
Why This Project?
- Solve a real problem I face as a guitarist
- Learn Android/Kotlin
- Explore AI music generation APIs
- Build something musicians actually want
MIT License - Free to use, modify, and learn from
- ECHO - Family memory app with AI voice preservation
- Sales-Coach-AI - AI-powered sales call analysis
- Portfolio - Full project showcase
Have ideas? Find a bug? Want to collaborate?
- Open an issue on GitHub
- Email: kennykriberney@gmail.com
- LinkedIn: Kenny Kriberney
⭐ Star this repo if you're excited about AI-powered music tools!