Neural Music Synthesis for Flexible Timbre Control
This work addresses the need for more adaptable and high-quality music synthesis tools for musicians and audio engineers, though it is incremental as it builds on existing WaveNet and neural network techniques.
The paper tackles the problem of music synthesis with flexible timbre control by modeling the entire process using generative neural networks, resulting in a learned embedding space that captures timbre variations and enables control and morphing, with synthesis quality evaluated numerically and perceptually.
The recent success of raw audio waveform synthesis models like WaveNet motivates a new approach for music synthesis, in which the entire process --- creating audio samples from a score and instrument information --- is modeled using generative neural networks. This paper describes a neural music synthesis model with flexible timbre controls, which consists of a recurrent neural network conditioned on a learned instrument embedding followed by a WaveNet vocoder. The learned embedding space successfully captures the diverse variations in timbres within a large dataset and enables timbre control and morphing by interpolating between instruments in the embedding space. The synthesis quality is evaluated both numerically and perceptually, and an interactive web demo is presented.