A Case Study on LLM Code Generation in Sonic Pi and Its Impact on Student Attitudes towards Computer Science
Josch Wang
Affiliation: Manhasset High School
IJSCAR Vol. 1, Issue 1 (2024) · pp. 19–23
Abstract
Research has demonstrated that creating music and soundscapes with tools like Sonic Pi can enhance the appeal of computer science and music composition as fields of study. However with current advancements in artificial intelligence (AI) the nature of computer science is evolving rapidly particularly with AI’s ability to generate code from a single prompt. Similarly AI technologies have already begun to transform music composition by enabling the generation of music. This study investigates whether live coding with large language models (LLMs) such as ChatGPT and Copilot influences perceptions of entering the fields of music composition and computer science. Understanding the impact of AI is crucial as it increasingly shapes various domains of our lives. To explore this we instructed a random sample of high school students in the basics of live coding emphasizing the inclusion of AI tools. The results indicate a positive shift in students’ attitudes toward programming and music creation suggesting that AI serves as a beneficial tool rather than a detrimental one.
Keywords: LLM Code Generation, Live Coding, Sonic Pi, CS Education