A comprehensive AI-powered music generation system that combines multiple models and approaches to generate music with control over melody, emotion, lyrics, and style.

• Multi-model music generation pipeline
• BERT-based music understanding and generation
• Emotion-aware music generation
• Lyrics to melody conversion
• Music style transfer
• Real-time music generation with agent-based system

• BERT Music Model: Music understanding and generation using BERT architecture
• GetMusic: Advanced music generation module
• Emotion: Emotion-aware music generation
• LLM: Large Language Model integration for music generation
• MelodyLogic: Melody generation and refinement
• SeederAgent: Agent-based real-time music generation

• Adding Generator: Audio processing and augmentation
• Lyric Checker: Lyrics processing and melody matching
• Recognition Transfer: Music style transfer
• Telemelody: Melody generation and processing

1. Python environment with dependencies:

   pip install -r requirements.txt

2. Install required system packages:

   # For audio processing
   apt-get install flac

1. Clone the repository
2. Install Python dependencies:

   pip install -r requirements.txt
   # Or using pipenv
   pipenv install

3. Download required model checkpoints (contact maintainers)

The project includes Docker support for containerized deployment:

docker build -t iah-music-generator .
docker run -it iah-music-generator

python getmusic/generate.py --config configs/default.yaml

bash emotion/Piano_gen.sh

python lyrichecker/lyrics_to_melody.py --input "your lyrics here"

The project is organized into multiple specialized components, each handling different aspects of music generation. Key directories:

• /bert: BERT-based music understanding
• /emotion: Emotion-aware generation
• /getmusic: Core music generation
• /llm: Language model integration
• /seederagent: Real-time generation agent

Contact maintainers for licensing information

Daniel Lee, Hoang Le

For questions and support, please contact us