构建钢琴演奏统一表征空间，覆盖从初学者到大师的全技能水平。
MAJEPPA: Morphing and Assessing in a Unified Piano Performance Space
- 用自监督学习融合乐谱与演奏的联合表征，统一生成与理解。
- 在4000段标注录音上训练，涵盖6个水平和6种场景，表现优于基线。
- 适用于音乐教育评估、比赛排名与技巧分析，适合音乐人工智能研究者。
我们提出MAJEPPA，一个自监督框架，用于学习从初学者练习到大师音乐会录音全技能水平的钢琴演奏表征。我们构建了包含约4000段标注录音的MAJEPPA数据集，覆盖六个技能等级和六种录音场景。通过联合优化目标，采用单一预训练的MIDI自回归模型：下一音符预测任务学习不同技能水平下基于乐谱的演奏生成；同时使用InfoNCE与监督对比损失，将抽象乐谱与演奏表征对齐于统一嵌入空间。所提模型可在统一框架中生成并理解演奏。通过引入EVPMR基准测试套件，涵盖质量评估、比赛排名、错误与技巧分类等下游任务，验证了学习表征的有效性，推动实现真实钢琴演奏空间建模的进展。
原文摘要 · Abstract (English)
We present MAJEPPA, a self-supervised framework to learn piano performance representations that span the full skill spectrum, from beginner practice sessions to virtuoso concert recordings. We curate the MAJEPPA dataset, comprising ~4,000 annotated recordings across six expertise levels and six recording contexts. We adapt a single pre-trained MIDI autoregressive model with a joint objective: next-token prediction learns score-conditioned performance generation at various skill levels, while InfoNCE and supervised contrastive losses align abstract score and performance representations in a joint embedding space. The proposed model both generates and understands performances in a unified framework. By introducing the EVPMR benchmark, a suite of downstream tasks spanning quality assessment, competition ranking, mistake and technique classification, we evaluate the learnt representations, demonstrating progress towards a real-world model for the piano performance space.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。