Digital advances have transformed the face of automatic music generation since its beginnings at the dawn of computing. Despite the many breakthroughs, issues such as the musical tasks targeted by different machines and the degree to which they succeed remain open questions. We present a functional taxonomy for music generation systems with reference to existing systems. The taxonomy organizes systems according to the purposes for which they were designed. It also reveals the inter-relatedness amongst the systems. This design-centered approach contrasts with predominant methods-based surveys and facilitates the identification of grand challenges to set the stage for new breakthroughs.

Research applying machine learning to music modelling and generation typically proposes model architectures, training methods and datasets, and gauges system performance using quantitative measures like sequence likelihoods and/or qualitative listening tests. Rarely does such work explicitly question and analyse its usefulness for and impact on real-world practitioners, and then build on those outcomes to inform the development and application of machine learning. This article attempts to do these things for machine learning applied to music creation. Together with practitioners, we develop and use several applications of machine learning for music creation, and present a public concert of the results. We reflect on the entire experience to arrive at several ways of advancing these and similar applications of machine learning to music creation.

Automatic music generation systems have gained in popularity and sophistication as advances in cloud computing have enabled large-scale complex computations such as deep models and optimization algorithms on personal devices. Yet, they still face an important challenge, that of long-term structure, which is key to conveying a sense of musical coherence. We present the MorpheuS music generation system designed to tackle this problem. MorpheuS' novel framework has the ability to generate polyphonic pieces with a given tension profile and long- and short-term repeated pattern structures. A mathematical model for tonal tension quantifies the tension profile and state-of-the-art pattern detection algorithms extract repeated patterns in a template piece. An efficient optimization metaheuristic, variable neighborhood search, generates music by assigning pitches that best fit the prescribed tension profile to the template rhythm while hard constraining long-term structure through the detected patterns. This ability to generate affective music with specific tension profile and long-term structure is particularly useful in a game or film music context. Music generated by the MorpheuS system has been performed live in concerts.

We present the architecture, technology and experimental applications of a real-time, multi-site, interactive and collaborative environment called Distributed Immersive Performance (DIP). The objective of DIP is to develop the technology for live, interactive musical performances in which the participants - subsets of musicians, the conductor and the audience - are in different physical locations and are interconnected by very high fidelity multichannel audio and video links. DIP is a specific realization of broader immersive technology - the creation of the complete aural and visual ambience that places a person or a group of people in a virtual space where they can experience events occurring at a remote site or communicate naturally regardless of their location. The DIP experimental system has interaction sites and servers in different locations on the USC campus and at several partners, including the New World Symphony of Miami Beach, FL. The sites have different types of equipment to test the effects of video and audio fidelity on the ease of use and functionality for different applications. Many sites have high-definition (HD) video or digital video (DV) quality images projected onto wide screen wall displays completely integrated with an immersive audio reproduction system for a seamless, fully three-dimensional aural environment with the correct spatial sound localization for participants. The system is capable of storage and playback of the many streams of synchronized audio and video data (immersidata), and utilizes novel protocols for the low-latency, seamless, synchronized real-time delivery of immersidata over local area networks and wide-area networks such as Internet2. We discuss several recent interactive experiments using the system and many technical challenges common to the DIP scenario and a broader range of applications. These challenges include: (1). low latency continuous media (CM) stream transmission, synchronization and data loss management; (2). low latency, real-time video and multichannel immersive audio acquisition and rendering; (3). real-time continuous media stream recording, storage, playback; (4). human factors studies: psychophysical, perceptual, artistic, performance evaluation; (5). robust integration of all these technical areas into a seamless presentation to the participants.

Robust data retrieval in the presence of uncertainty is a challenging problem in multimedia information retrieval. In query-by-humming (QBH) systems, uncertainty can arise in query formulation due to user-dependent variability, such as incorrectly hummed notes, and in query transcription due to machine-based errors, such as insertions and deletions. We propose a fingerprinting (FP) algorithm for representing salient melodic information so as to better compare potentially noisy voice queries with target melodies in a database. The FP technique is employed in the QBH system back end; a hidden Markov model (HMM) front end segments and transcribes the hummed audio input into a symbolic representation. The performance of the FP search algorithm is compared to the conventional edit distance (ED) technique. Our retrieval database is built on 1500 MIDI files and evaluated using 400 hummed samples from 80 people with different musical backgrounds. A melody retrieval accuracy of 88% is demonstrated for humming samples from musically trained subjects, and 70% for samples from untrained subjects, for the FP algorithm. In contrast, the widely used ED method achieves 86% and 62% accuracy rates, respectively, for the same samples, thus suggesting that the proposed FP technique is more robust under uncertainty, particularly for queries by musically untrained users.