Artificial intelligence and digital technologies for piano sight-reading skill development: a scoping review

Piano sight-reading is a complex cognitive activity that many pupils remain unable to perform despite sustained educational efforts. AI and digital technology have revolutionized numerous educational fields; however, their integration with educational technology for sight-reading piano remains diffuse and concerning to experts due to a lack of coherence across AI-related investigations. This study aims to systematize knowledge on the application of AI and digital technologies in educational technology for sight-reading piano, following the PRISMA-ScR guidelines. A search of four main databases (Web of Science, IEEE Xplore, Scopus, ACM Digital Library) was conducted for papers on AI-related technology for sight-reading piano from 2014 to 2024. This resulted in screening 368 entries to select 33 relevant to the study objective. Five types of technology exist: AI-related intelligent tutoring systems, computer vision and optical music recognition, pattern recognition with deep learning, applications of virtual reality and augmented reality, and mobile and IoT. The study demonstrates a discrepancy between the complexity of AI and accessibility for pupils. AI-powered tutoring systems and deep learning approaches are showing promising results in controlled settings, but evidence on long-term effectiveness remains limited. A fundamental tension exists between analytical sophistication and accessibility: high-performing systems require substantial computational resources, while accessible mobile solutions provide much weaker analytical capabilities. On the other hand, accessibility for pupils remains a top priority, including the use of IoT technology for educational sight-reading piano.