Xenic

A modern, Material 3 local music player with AI-powered music classification

• 🤖 AI Music Classification: Automatically classify your music as Christian or Secular using machine learning
• 🎵 Local Music Player: Play your local music files with a beautiful Material 3 interface
• 🎧 Gapless Playback: Enjoy uninterrupted transitions between tracks
• 🎛️ Built-in Equalizer: Fine-tune your audio with customizable equalizer settings
• 📝 Automatic Lyrics: Download and sync lyrics with your music
• 🎨 Material You: Adaptive theming that matches your device's color scheme
• 📱 Modern UI: Clean, intuitive interface designed for Android

Xenic features an advanced AI-powered music classification system that can automatically categorize your music library:

• Christian Music Detection: Identifies Christian music based on audio characteristics
• Secular Music Classification: Distinguishes secular music genres
• Batch Processing: Classify your entire music library at once
• Smart Playlists: Automatically generate playlists based on classification results
• Local Processing: Fast classification using optimized machine learning models

The classification system uses a custom-trained machine learning model with:

• Audio-Based Analysis: 65+ audio features including tempo, harmony, rhythm, and spectral properties
• High Accuracy: 84.1% test accuracy with balanced Christian/Secular detection
• Offline Operation: No internet connection required for classification
• Fast Processing: ~4.9 files/second with parallel processing

Model Source: The AI model was trained using the Christian Music Classifier project, which includes comprehensive audio feature extraction and machine learning pipeline.

• Minimum Android Version: API 26 (Android 8.0)
• Target Android Version: API 35 (Android 15)
• Architecture: MVVM with Repository pattern
• UI Framework: Jetpack Compose with Material 3
• Database: Room for local data storage
• Audio Processing: Custom audio feature extraction and classification
• AI Model: Trained Random Forest classifier with 65+ audio features
• Machine Learning: Based on Christian Music Classifier research
• Foundation: Built upon Booming Music by mardous

• Android 8.0 (API 26) or higher
• Storage permission for music files
• Internet connection (for lyrics download and initial classification)

• Android Studio Arctic Fox or later
• JDK 11 or later
• Android SDK 26+

git clone https://github.com/xenhusk/XenhuskMusic.git
cd XenhuskMusic
./gradlew assembleDebug

This project is licensed under the GNU General Public License v3.0 - see the LICENSE.txt file for details.

The AI music classification system in Xenic is powered by custom machine learning research conducted in the Christian Music Classifier project. This research includes:

• Audio Feature Engineering: 65+ carefully selected audio features
• Model Training: Random Forest and SVM classifiers with cross-validation
• Performance Optimization: 84.1% accuracy with balanced class detection
• Offline Capability: Complete audio analysis without internet dependency
• Comprehensive Testing: Extensive validation on 531+ audio files

The research demonstrates how audio characteristics can effectively distinguish between Christian and secular music without relying on lyrics or metadata.

David Paul Desuyo (xenhusk)

• GitHub: @xenhusk
• LinkedIn: xenhusk
• Email: desuyodavidpaul@gmail.com

Xenic is built upon Booming Music by Christians Martínez Alvarado (mardous), a modern Material 3 music player for Android. This project provided the solid foundation including:

• Modern UI Framework: Jetpack Compose with Material 3 design
• Music Player Core: Gapless playback, equalizer, and audio processing
• Library Management: Song, album, artist, and playlist organization
• Android Integration: Android Auto support, widgets, and system integration
• MVVM Architecture: Clean, maintainable codebase structure

Special thanks to mardous and the contributors for creating such a robust and feature-rich music player that made Xenic possible.

The AI classification system builds upon machine learning research conducted in the Christian Music Classifier project, which developed the audio-based classification algorithms and trained models used in Xenic.