Artificial-Music-Synthesis

This repository contains the iPython Notebook of the implementation of Artificial Music Generation using Recurrent Neural Network.

Important Points about the Project:

• A Deep RNN model is used to develop the model.
• A series of LSTM(Long-Short Term Memory) and GRU(Gated Recurrent Unit) cells are used as the basic units.
• The model is similar to a Language Predictive model.
• The frequency pattern of the musical notes is analyzed.
• Musical notes & tones in fixed intervals of time are represented in the form of One-hot encoding.
• The model is trained on several epochs using the generated one-hot encoding.
• The trained model is finally used to predict the time-sequence frequency pattern of musical notes.
• These generated encodings are finally Post-processed to get the melodious musical sound.
• The generated Music is a combination of sounds produced from various musical instruments.

Frameworks and Libraries Used:

• Model is developed on Keras Framework.
• For Pre-processing and Post-processing of Musical data, IPython library is used.

Note : The Music Files that are synthesised by the network are also uploaded.