About This Project
Implemented an LSTM-based recurrent neural network to generate symbolic music in ABC notation. Used PyTorch to train embeddings, LSTM layers, and softmax output to predict sequential notes. Demonstrated temporal dependency learning and sequence prediction accuracy through iterative sampling.
Key Features
- Character-level music modeling
- Sequential sampling procedure
- LSTM-based architecture
- Softmax output generation
