Lineage & Freedom: A Quantum Computational Framework for Improvisation

Scott Oshiro, Center for Computer Research in Music and Acoustics (CCRMA) Stanford University

Abstract

Musical improvisation is the spontaneous act of musically communicating and composing with one or more musicians in the moment. However, in the current field of music information retrieval (MIR) there has not been a focus on developing algorithms to better understand the inner workings of improvisation, especially between musicians of different cultures and backgrounds. It is in part, due to the vast and complex space in which improvisation operates. A space with infinite outcomes that could possibly unfold. Because of this, classical computers are not efficient at modeling these types of interactions. Musical improvisation is more aligned with quantum computers, their properties allow us to simultaneously consider and handle this large number of musical possibilities. This dissertation proposes and outlines a new framework for developing music improvisation applications on quantum computers.

QuiKo: Quantum Beat Generation

Scott Oshiro, Center for Computer Research in Music and Acoustics (CCRMA) Stanford University

Abstract

In this chapter a quantum music generation application called QuiKo will be discussed. It combines existing quantum algorithms with data encoding methods from quantum machine learning to build drum and audio sample patterns from a database of audio tracks. QuiKo leverages the physical properties and characteristics of quantum computers to generate what can be referred to as Soft Rules proposed by Alexis Kirke. These rules take advantage of the noise produced by quantum devices to develop flexible rules and grammars for quantum music generation. These properties include qubit decoherence and phase kickback due controlled quantum gates within the quantum circuit. QuiKo builds upon the concept of soft rules in quantum music generation and takes it a step further. It attempts to mimic and react to an external musical inputs, similar to the way that human musicians play and compose with one another. Audio signals are used as inputs into the system. Feature extraction is then performed on the signal to identify the harmonic and percussive elements. This information is then encoded onto the quantum circuit. Measurements of the quantum circuit are then taken providing results in the form of probability distributions for external music applications to use to build the new drum patterns.

A QUANTUM-CLASSICAL NETWORK FOR BEAT-MAKING PERFORMANCE

Scott Oshiro, Center for Computer Research in Music and Acoustics (CCRMA) Stanford University

Omar Costa Hamido, University of California Irvine

This paper was published in the Journal of Network Music and Arts Vol. 2 (2020) Issue 1

Presented at the NOWNET Arts Conference 2020

Abstract

In recent years, quantum computing has emerged as the next frontier in computational and information technologies. Even though it has found potential applications in solving complex problems in fields such as chemistry, machine learning, and cryptography, among other fields, there has been little research conducted on its applications for music and acoustic technologies. This paper will discuss the use of a quantum internet protocol in the context of networked music performance in which quantum computing could play a role in processing musical data via a cloud-based music software application. We also propose an example model for a beat-making performance network using a smart music playlist application deployed on a simulated quantum internet. In the proposed system design and architecture, several beat-makers located remotely from each other are connected live over a simulated quantum internet in a distributed networked music performance. Each beat-maker node transmits and receives audio sample time slices of beat patterns from one another to use in their local performances. This model provides a proof of concept for implementing quantum algorithms, standards, and protocols in music software and network applications when a quantum internet becomes available.

JACKTRIP ON RASPBERRY PI

Chris Chafe, Scott Oshiro

Center for Computer Research in Music and Acoustics (CCRMA)

Stanford University, USA

This paper was presented at the 17th Linux Audio Conference 2019

Abstract

The jacktrip application for wide area network music performance has been ported to Raspberry Pi. The present setup runs Fedora 29 with the xfce desktop on a Model 3 B+ in conjunction with standard, low-cost stereo USB soundcards. We describe all the steps from initial OS installation through building and running jacktrip.