These slides were presented at PyCon India, 2016. The slide deck mainly talks about adopting the appropriate Neural Network in order to make computers generate it's own music and focuses on LSTM (Long Short Time Memory - a special type of Recurrent Neural Network) architecture. It also talks about three prime phases involved in algorithmic music generation, which are as follows: 1. Preprocessing the input i.e. converting the MP3 files into np-tensors. 2. Training the model. 3. Generation phase. It also focusses on various challenges involved in doing so i.e. composing the music algorithmically. The slides contain the visualization of the input files given to the model as well as music generated by the neural network model. You can find the entire python code used for developing this model on GitHub: https://github.com/unnati-xyz/music-generation