Demonstration of PerformanceNet: A Convolutional Neural Network Model for Score-to-Audio Music Generation
This work addresses the challenge of creating expressive audio performances from symbolic music scores for musicians and AI applications, representing a novel method rather than an incremental improvement.
The authors tackled the problem of generating audio from musical scores by introducing PerformanceNet, a convolutional neural network that automatically assigns performance-level attributes like velocity changes and synthesizes audio, achieving a new approach to score-to-audio music generation.
We present in this paper PerformacnceNet, a neural network model we proposed recently to achieve score-to-audio music generation. The model learns to convert a music piece from the symbolic domain to the audio domain, assigning performance-level attributes such as changes in velocity automatically to the music and then synthesizing the audio. The model is therefore not just a neural audio synthesizer, but an AI performer that learns to interpret a musical score in its own way. The code and sample outputs of the model can be found online at https://github.com/bwang514/PerformanceNet.