Machine learning research that matters for music creation: A case study

Abstract

* Corresponding author 1 Research applying machine learning to music modeling 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.

Exploring the Role of Deep Learning Technology in the Sustainable Development of the Music Production Industry

“…Deep learning refers to the machine learning of past musical features to simulate and generate the emotional expression required in a specific type of music. [11,13,18] SCT This term defines the customer's evaluation of the benefits and costs of switching to new technologies, services, or products. [54,55] Notes: MPQ: quality of music production, MPQ-EF: emotions and feelings for music production, MPQ-TC: techniques and capabilities for music production, MPQ-EI: external interacting factors affecting music production, SA: showing appropriateness, SMP: satisfaction of music products, DL: deep learning, SCT: switching costs of technology usage.…”

“…Due to the great efforts of worldwide scholars and professionals in recent years, deep learning has successfully mastered various features and styles of past music pieces and automatically created music to satisfy audiences [17]. Deep learning is recognized as an innovative field that could help the music production industry change and evolve in the future [18].…”

Quantum GestART: identifying and applying correlations between mathematics, art, and perceptual organization

“…Research between machine learning and meaningful artistic applications is a fertile field of research (Sturm et al 2018). Rather than adopting a one-to-one relationship between images and sounds, we propose a possible computational application of the method discussed in Section 3 using machine learning approaches.…”

Dynamic Procedural Music Generation from NPC Attributes

“…• Analytic algorithms reduce the potential data size and general perceived musical feel by distilling specific features from existing pieces of music. This technique is often seen in experimental or learning-based algorithms, in which a single monophonic line is extracted from a polyphonic piece [9], or when a basic musical structure is learned from a large database of pieces [27] [14] [15] [53].…”