NN Music: Improvising with a ‘Living’ Computer

Abstract

This paper proposes attributes of a living computer music, the product of a live algorithm. It illustrates how these attributes can inform creative design with reference to a real-time system for solo performer-machine collaboration, Neural Network Music, and the PQƒ framework proposed for live algorithms. Improvisation is treated as a classification problem at a high level of musical behaviour which can be measured statistically and train a multilayer perceptron neural network. Network outputs shape a stochastic-based synthesis engine. Mappings are covertly assigned, revisited by both player and machine as a performance develops. As the timing and choice of mapping is unknown, both participants are invited to learn and adapt to a responsive sonic environment which is created afresh on each performance. This offers a novel real-time application of feed-forward neural networks and a challenging, creative technological platform for freely improvised music.

Maxine’s Turing Test – A Player-Program as Co-Ethnographer of Socio-Aesthetic Interaction in Improvised Music

“…Rowe introduced a useful binary between player and instrument programs, drawing this distinction between systems treated as semi-autonomous agents in performance versus those which remain the object of control of a human performer (Rowe 1993). Building on the player program concept, Blackwell and Young develop the concept of live algorithm, and discuss the ideals for how such systems should be designed (Young 2008). As noted by several researchers (Eigenfeldt et.…”

“…Maxine is an interactive musical agent based in Max/MSP, designed create music in real-time based on audio input in dialogue with a human performer, in a manner reflective of freer practices of improvisation (Banerji 2010). Within the taxonomy of interactive music systems, Maxine can be classified as either a "player program" or "Live Algorithm for Music" (Young 2008), interacting with audio input from the environment and producing her own contributions in the form of live audio processing and control of MIDIbased musical events. On one level, the goal of this research is the development of a machine capable of realtime improvisative creativity reflective of practices developed by improvisers over the past 50 years.…”

Towards a Live Dance Improvisation between an Avatar and a Human Dancer

“…For example, [15] describes a system that generates novel music and adapts to the reaction of a human co-performer in real time. Much fewer examples of artificial improvisation have been reported within dance.…”

Crafting Creative Melodies: A User-Centric Approach for Symbolic Music Generation

“…Each agent optimizes its musical task by utilizing its specific set of rules or learned behaviors. NN music [26] is a musical agent that uses multi-layer perceptron (MLP) neural networks with PQF modular architecture [32]. The PQF architecture includes three modules: listening and analysis (P), performing/synthesis (Q), and generation (F).…”