A-Muze-Net: Music Generation by Composing the Harmony based on the Generated Melody
This work addresses music generation for AI applications, but it is incremental as it builds on existing neural network approaches with specific conditioning techniques.
The paper tackles the problem of generating piano music MIDI files by modeling melody and harmony separately, with the left hand conditioned on the right hand, resulting in a significant improvement over state-of-the-art methods for training on such datasets.
We present a method for the generation of Midi files of piano music. The method models the right and left hands using two networks, where the left hand is conditioned on the right hand. This way, the melody is generated before the harmony. The Midi is represented in a way that is invariant to the musical scale, and the melody is represented, for the purpose of conditioning the harmony, by the content of each bar, viewed as a chord. Finally, notes are added randomly, based on this chord representation, in order to enrich the generated audio. Our experiments show a significant improvement over the state of the art for training on such datasets, and demonstrate the contribution of each of the novel components.