Deep Learning For Music Generation

The use of deep learning to mimics the working of the human brain by processing data for use in detecting objects, translating languages, recognizing speech and making decisions [4]. Recently, deep learning methods have achieved state-of-the-art results on examples of this problem and therefore, I think that Artificial Intelligence as a subject play an important role in this time. My colleague and I am a student of Computer Science and for the project we decided to build a Deep Learning for Music Generation. Our project deals with the generation of music using raw audio files (midi file) in the frequency domain relying on various architectures. Fully connected and convolutional layers are used along with LSTM's to capture rich features in the frequency domain and increase the quality of music generated. The hard amount of work can be reduced. We have tried to cover various aspects inside this project, and also tried to solve the major issues.