Researchers have developed ScorePrompts, an interactive system that allows users to explore symbolic music scores using natural language. Users can upload a score, receive descriptions of its musical structure, ask questions about specific sections, and view the analysis results directly on the notation. The system integrates various music information retrieval (MIR) components to analyze harmony, tonality, form, and texture, then uses a schema-constrained language model to generate natural-language explanations. This approach enables users to access and inspect detailed musical analysis without needing to infer it from raw data, facilitating exploratory score analysis. AI
IMPACT Enables new forms of AI-assisted musicological research and education.
RANK_REASON The item describes a new research paper detailing a novel system for music score analysis. [lever_c_demoted from research: ic=1 ai=1.0]
- alphaXiv
- arXiv
- CatalyzeX
- DagsHub
- Emmanouil Karystinaios
- Gotit.pub
- Hugging Face
- MusicXML
- ScienceCast
- ScorePrompts
- Verovio
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →