Between the AI and Me: Analysing Listeners' Perspectives on AI- and Human-Composed Progressive Metal Music
Pedro Sarmento, Jackson Loth, Mathieu Barthet
TL;DR
This paper examines listener perceptions of AI-generated versus human-composed symbolic progressive metal using a mixed-methods listening and reflection approach. By comparing ProgGP-generated progressions and rock stimuli to human-composed counterparts, the study assesses genre congruence, preference, creativity, and perceived humanness, aided by qualitative thematic analysis. Findings indicate a clear listener preference for human-composed progressions, though AI samples—especially cherry-picked ones—can approach human judgments on several dimensions, and participants could often differentiate genres but not reliably distinguish AI from humans. The work contributes methodologically with a mixed evaluation framework, discusses ethical implications of data diversity in MIR, and advocates for genre-focused data provisioning and the use of guitar tablature to broaden representations in symbolic music generation.
Abstract
Generative AI models have recently blossomed, significantly impacting artistic and musical traditions. Research investigating how humans interact with and deem these models is therefore crucial. Through a listening and reflection study, we explore participants' perspectives on AI- vs human-generated progressive metal, in symbolic format, using rock music as a control group. AI-generated examples were produced by ProgGP, a Transformer-based model. We propose a mixed methods approach to assess the effects of generation type (human vs. AI), genre (progressive metal vs. rock), and curation process (random vs. cherry-picked). This combines quantitative feedback on genre congruence, preference, creativity, consistency, playability, humanness, and repeatability, and qualitative feedback to provide insights into listeners' experiences. A total of 32 progressive metal fans completed the study. Our findings validate the use of fine-tuning to achieve genre-specific specialization in AI music generation, as listeners could distinguish between AI-generated rock and progressive metal. Despite some AI-generated excerpts receiving similar ratings to human music, listeners exhibited a preference for human compositions. Thematic analysis identified key features for genre and AI vs. human distinctions. Finally, we consider the ethical implications of our work in promoting musical data diversity within MIR research by focusing on an under-explored genre.