Sitar Script is an advanced audio processing and note classification system designed specifically for sitar music. This project uses machine learning and deep learning techniques, including CNN, CNN-LSTM, and CNN-GRU architectures, to analyze audio files, extract features, and classify sitar notes accurately.
- Audio Feature Extraction: Extracts MFCC, Chroma features, and detects onsets for precise note analysis.
- Deep Learning Models: Utilizes various architectures such as CNN, CNN-LSTM, and CNN-GRU for note classification, allowing flexibility and enhanced performance.
- Preprocessing Tools: Handles non-silent segment detection and feature scaling for efficient processing.
- Real-Time Prediction: Accepts audio files as input and outputs the sequence of predicted sitar notes.