SchrödingeRNN: Generative Modeling of Raw Audio as a Continuously Observed Quantum State

Abstract

We introduce SchrödingeRNN, a quantum inspired generative model for raw audio. Audio data is wave-like and is sampled from a continuous signal. Although generative modelling of raw audio has made great strides lately, relational inductive biases relevant to these two characteristics are mostly absent from models explored to date. Quantum Mechanics is a natural source of probabilistic models of wave behaviour. Our model takes the form of a stochastic Schrödinger equation describing the continuous time measurement of a quantum system, and is equivalent to the continuous Matrix Product State (cMPS) representation of wavefunctions in one dimensional many-body systems. This constitutes a deep autoregressive architecture in which the systems state is a latent representation of the past observations. We test our model on synthetic data sets of stationary and non-stationary signals. This is the first time cMPS are used in machine learning.

arXiv comment32 pages, 20 figures, under review for MSML 2020

References in corpus (13)

• A Recurrent Latent Variable Model for Sequential Data
• SampleRNN: An Unconditional End-to-End Neural Audio Generation Model
• Sequential generation of entangled multi-qubit states
• Continuous Matrix Product States for Quantum Fields
• Unsupervised Generative Modeling Using Matrix Product States
• Tree Tensor Networks for Generative Modeling
• The challenge of realistic music generation: modelling raw audio at scale
• Holographic quantum states
• TensorNetwork for Machine Learning
• Probabilistic Modeling with Matrix Product States
• Expressive power of tensor-network factorizations for probabilistic modeling, with applications from hidden Markov models to quantum machine learning
• Efficient Learning of Harmonic Priors for Pitch Detection in Polyphonic Music
• Shortcut Matrix Product States and its applications