By definition, traditional music is in a constant state of friction with innovation, exemplified by resistance to ‘outside’ influences such as different instruments, different ways of learning, and forces of commercialisation. An emerging external influence is artificial intelligence (AI), which is now capable of synthesising music collections at scales dwarfing those crafted by people and communities. In this paper, we examine the impact of research and development of AI on Irish traditional music through case studies of two generative AI systems: folk-rnn and Suno. How can researchers and engineers (academic or industrial) who develop and apply AI to specific practices of music make meaningful and non-harmful contributions to those practices? To answer this question, we critically reflect on the tensions that arise between tradition and innovation, how Irish traditional music becomes subject to data colonialism, and the interdisciplinary challenges of ethically engaging as researchers with a traditional music community. We ask what perspectives are needed to balance the interests of academic research and value systems in traditional music communities, and provide three ways forward for computer science to deepen the considerations of their impacts on communities of practice.

Explainable Artificial Intelligence (XAI) is a discipline concerned with understanding predictions of AI systems. What is ultimately desired from XAI methods is for an AI system to link its input and output in a way that is interpretable with reference to the environment in which it is applied. A variety of methods have been proposed, but we argue in this paper that what has yet to be considered is miscommunication: the failure to convey and/or interpret an explanation accurately. XAI can be seen as a communication process and thus looking at how humans explain things to each other can provide guidance to its application and evaluation. We motivate a specific model of communication to help identify essential components of the process, and show the critical importance for establishing common ground, i.e., shared mutual knowledge, beliefs, and assumptions of the participants communicating.

Several companies now offer platforms for users to create music at unprecedented scales by textual prompting. As the quality of this music rises, concern grows about how to differentiate AI-generated music from human-made music, with implications for content identification, copyright enforcement, and music recommendation systems. This article explores the detection of AI-generated music by assembling and studying a large dataset of music audio recordings (30,000 full tracks totaling 1,770 h, 33 m, and 31 s in duration), of which 10,000 are from the Million Song Dataset (Bertin-Mahieux et al., 2011) and 20,000 are generated and released by users of two popular AI music platforms: Suno and Udio. We build and evaluate several AI music detectors operating on Contrastive Language Audio Pretraining embeddings of the music audio, then compare them to a commercial baseline system as well as an open-source one. We applied various audio transformations to see their impacts on detector performance and found that the commercial baseline system is easily fooled by simply resampling audio to 22.05 kHz. We argue that careful consideration needs to be given to the experimental design underlying work in this area, as well as the very definition of "AI music." We release all our code at https://github.com/lcrosvila/ai-music-detection.

This paper surveys 560 publications about music genre recognition (MGR) published between 2013–2022, com- plementing the comprehensive survey of [474], which cov- ered the time frame 1995–2012 (467 publications). For each publication we determine its main functions: a review of research, a contribution to evaluation methodology, or an experimental work. For each experimental work we note the data, experimental approach, and figure of merit it ap- plies. We also note the extents to which any publication engages with work critical of MGR as a research problem, as well as genre theory. Our bibliographic analysis shows for MGR research: 1) it typically does not meaningfully engage with any critique of itself; and 2) it typically does not meaningfully engage with work in genre theory. 

This paper surveys 560 publications about music genre recognition (MGR) published between 2013–2022, complementing the comprehensive survey of [474], which covered the time frame 1995–2012 (467 publications). For each publication we determine its main functions: a review of research, a contribution to evaluation methodology, or an experimental work. For each experimental work we note the data, experimental approach, and figure of merit it applies. We also note the extents to which any publication engages with work critical of MGR as a research problem, as well as genre theory. Our bibliographic analysis shows for MGR research: 1) it typically does not meaningfully engage with any critique of itself; and 2) it typically does not meaningfully engage with work in genre theory.

In this paper, we use critical and agonistic modes of inquiry to analyse and critique a specific application of AI to music practice. It records a structured interdisciplinary dialogue between 1) a musicologist and social scientist and 2) an engineer in music and computer science, focusing on folk-rnn and Irish Traditional Music (ITM) as a case study. We debate the role of data ethics in AI music applications, the dynamics of inclusion and exclusion, and the nature of embedded value systems and power asymmetries inherent in applying AI to music. We discuss how identifying the value of AI music applications is critical for ensuring research efforts make musical contributions along with academic and technical ones. Overall, this agonistic dialogue exemplifies how questions of right and wrong — the core of ethics — can be examined as AI is applied more and more to music practice.

As music generated using artificial intelligence (AI music) becomes more prevalent — originating not only from individuals but also commercial services — the need to study it and its impacts becomes important. How can this material and its sources be meaningfully studied and critically engaged with, especially considering the unprecedented scales possible with generative AI? The paper begins to answer this question by considering AI music along seven aspects: 1) the company providing an AI music service; 2) its founders and employees; 3) the use of the service; 4) the users; 5) the algorithms; 6) the music; and 7) the sustainability. We make our discussion more concrete by considering the contemporary AI music service Boomy. While our investigations are preliminary and focused on a single AI music service, we argue that they open several interesting avenues of exploration for many disciplines and their intersections to help prepare for the coming flood of AI music. This paper asks many more questions than it answers, which is a feature (not a bug) of it advocating for a new domain of study: AI Music Studies.

A text-to-music (TTM) model should synthesize audio that reflects the concepts in a given prompt as long as it has been trained on those concepts. If a prompt references concepts that the TTM model has not been trained on then the audio it synthesizes will likely not match. This paper investigates the application of a simple gradient-based approach called textual inversion (TI) to expand the concept vocabulary of a trained TTM model without compromising the fidelity of concepts on which it has already been trained. We apply this technique to MusicGen and measure its reconstruction and editability quality, as well as its subjective quality. We see TI can expand the concept vocabulary of a pretrained TTM model, thus making it personalized and more controllable without having to finetune the entire model. 

Modeling harmonic progressions in symbolic music is a complex task that requires generating musically coherent and varied chord sequences. In this study, we employ a transformer-based architecture trained on a comprehensive dataset of 48,072 songs, which includes an augmented set of 4,300 original pieces from the iReal Pro application transposed across all chromatic keys. We introduce a novel tokenization and voicing encoding strategy designed to enhance the musicality of the generated chord progressions. Our approach not only generates chord progression suggestions but also provides corresponding voicings tailored for instruments such as piano and guitar. To evaluate the effectiveness of our model, we conducted a listening test comparing the harmonic progressions produced by our approach against those from a baseline model. The results indicate that our model generates progressions with more fluid voicings, coherent harmonic motion, and plausible chord suggestions, effectively utilizing repetition and variation to enhance musicality.