Creativity in Generative Musical Networks: Evidence From Two Case Studies

Abstract

Deep learning, one of the fastest-growing branches of artificial intelligence, has become one of the most relevant research and development areas of the last years, especially since 2012, when a neural network surpassed the most advanced image classification techniques of the time. This spectacular development has not been alien to the world of the arts, as recent advances in generative networks have made possible the artificial creation of high-quality content such as images, movies or music. We believe that these novel generative models propose a great challenge to our current understanding of computational creativity. If a robot can now create music that an expert cannot distinguish from music composed by a human, or create novel musical entities that were not known at training time, or exhibit conceptual leaps, does it mean that the machine is then creative? We believe that the emergence of these generative models clearly signals that much more research needs to be done in this area. We would like to contribute to this debate with two case studies of our own: TimbreNet, a variational auto-encoder network trained to generate audio-based musical chords, and StyleGAN Pianorolls, a generative adversarial network capable of creating short musical excerpts, despite the fact that it was trained with images and not musical data. We discuss and assess these generative models in terms of their creativity and we show that they are in practice capable of learning musical concepts that are not obvious based on the training data, and we hypothesize that these deep models, based on our current understanding of creativity in robots and machines, can be considered, in fact, creative.

Social robotics in music education: A systematic review

“…Finally, we would like to highlight that the studies under examination provided evidence of positive effects on students' learning in the educational context related to music. This is also noted in previous studies (Cádiz et al, 2021;Chakraborty et al, 2021;Savery et al, 2021). Nonetheless, these results must be interpreted very cautiously, due to the limitations to generalize them because the sample used is so small.…”

“…Research into robotics and music is enjoying a boom (Cádiz et al, 2021) in the broad field of music research (Ilari, 2020). "Robotic music" focuses on developing the intelligence of machines, in terms of algorithms and cognitive models, with the aim of capturing the underlying principles of music perception, composition, and playing (Chakraborty et al, 2021).…”

Transparency in Music-Generative AI: A Systematic Literature Review

“…They expose the stated issue of creativity, review three popular approaches to evaluate computational creativity, observe the main approaches to music generation and identify its open challenges. Then, Cádiz et al (2021) points out the need to challenge the understanding of computational creativity and stresses the need for more research in this area. In their work, they present a creativity assessment of two generative models (TimberNet 17 and StyleGAN Pianorrolls 18 ) and demonstrate they were able to learn musical concepts not straightforward in the training data.…”

“…A generation algorithm is expected to produce similar outputs to the input or reference data, balancing between differing and not replicating the training data (Cádiz et al, 2021). This expectation arises critical questions about whether the generation model is extracting or copying fragments from the input data, and whether its generations can be considered original.…”

Examining the Relationships between High School Students' Interests in Music and Their Creativity

“…But as Park (2019) asks: When we see something as a work of art, should it just be created, chosen and combined by humans? We are used to the idea that people can create things or ideas that other people consider "new"this occurs in almost every field every day (Cádiz et al, 2021). If art education is personal, it improves creativity.…”