Musify - Music Composition using AI

  • Machine Learning

Mentors :

  • Abhijit Kumar

  • Amit Joshi

  • Soumya Sharma

Mentees :

  • 10

"“Where words fail, music speaks.”

In the revolutionary age of Artificial Intelligence and Machine Learning, let us compose our own original music using Deep Learning without really knowing any music theory.
The project is aimed to combine the area of music and Deep Learning to create an automatic music generation model. A variant of Recurrent Neural Networks i.e. LSTM is used to capture the long term dependencies in the input sequence.
Who this project is for:

  • One who is keen to dive in the interesting world of Machine Learning and Artificial Intelligence and want to learn it from scratch.
  • The project is ideal for beginners, as it starts from the fundamentals and gradually builds up your skills in Python, Machine Learning.

Project Outcomes:

  • Introduction to the whole new world of machine Learning and Deep learning
  • Application of Deep Learning in Music Composition
  • Master the Deep Learning package: TensorFlow
  • Last but not the least, apply your coding skills to build an entire Project from scratch

Python Tutorial:

Selection Procedure:

  • Selection would be based on your SOP. Your SOP should include the motivation behind selection of this project or any particular inclination towards this topic.
  • You are advised to include any previous works done in Python, Machine Learning etc in your SOP, if any (not mandatory).


  • Basic Knowledge of any programming language will be appreciated, but no prior experience is required. We will start from scratch.
  • Your enthusiasm, involvement and interest to learn new things are required.

Tentative Timeline :

Week Number Tasks to be Completed
Week 1 Learning Python Basics
Week 2 Complete Walkthrough of Numpy and Pandas
Week 3 Explore Neural Networks and Deep Learning
Week 4 Making Basic Models using TensorFlow
Week 5 Development of Deep Learning model using LSTM
Week 6 Model Implementation and Testing