发表机构
The University of New South Wales; The University of Western Australia; Queen Mary University of London; Great Bay University（新南威尔士大学; 西澳大利亚大学; 伦敦玛丽女王大学; 大湾区大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出首个端到端声部感知合唱转录框架PawCT，联合建模音符转录与SATB声部分配，在YouChorale上显著优于现有基线，验证了联合建模的有效性。
AI 中文摘要
单乐器自动音乐转录（AMT）已取得显著进展，但合唱应用要求将女高音、女低音、男高音和男低音（SATB）转录为独立声部。现有的音符级合唱AMT方法仅生成单一的合并音符轨道，限制了排练、教学和乐谱重建。为解决此局限，我们提出了面向声部感知的合唱转录（PawCT），据我们所知，这是首个从合唱音频中识别活跃SATB声部并将其分别转录为独立音符级轨道的端到端神经框架。PawCT结合了声部特定的起音、偏移和帧预测，以及声部存在性估计、联合级监督和基于SATB音域的先验（RP）及其有序连续性（OC）扩展的结构化训练目标，后者增加了声部内旋律连续性和跨声部音高排序。在YouChorale上，PawCT-RP-OC在50毫秒起音容差下实现了0.225的宏观声部感知音符F1值，相对优于改编的合唱基线（0.165）36.4%，并优于两阶段后处理分配流程（0.175）。其声部无关变体PagCT实现了0.382的50毫秒起音F1值，而先前最先进的合唱AMT模型为0.237。在CSD和Cantoria上的跨数据集评估进一步检验了数据集偏移下的性能。这些结果证明了联合建模音符转录和声部分配的优势。代码和演示可在该https URL获取。
英文摘要
Single-instrument automatic music transcription (AMT) has advanced substantially, yet choral applications require soprano, alto, tenor, and bass (SATB) to be transcribed as separate parts. Recent note-level choral AMT instead produces a single merged note track, limiting rehearsal, education, and score reconstruction. To address this limitation, we introduce Part-aware Choral Transcription (PawCT), to our knowledge the first end-to-end neural framework that identifies active SATB parts from choral audio and transcribes each into a separate note-level track. PawCT combines part-specific onset, offset, and frame prediction with part-presence estimation, union-level supervision, and structured training targets using a range prior (RP) based on SATB pitch ranges and its ordered-continuity (OC) extension, which adds within-part melodic continuity and cross-part pitch ordering. On YouChorale, PawCT-RP-OC achieves a macro part-aware note F1 of 0.225 at a 50-ms onset tolerance, outperforming an adapted choral baseline (0.165) by 36.4% relative and a two-stage post-hoc assignment pipeline (0.175). Its part-agnostic variant, PagCT, achieves a 50-ms onset F1 of 0.382, compared with 0.237 for the previous state-of-the-art choral AMT model. Cross-dataset evaluations on CSD and Cantoria further assess performance under dataset shift. These results demonstrate the benefit of jointly modeling note transcription and vocal-part assignment. Code and demos are available at https://hanyu-meng.github.io/Paw_Choral_AMT_Demo/.
CommentsSubmitted to ICASSP2027
