Character-Level Singing Technique Detection and Evaluation: A Two-Stage Approach
IEEE Signal Processing Letters · 2026 · 33 · 3172-3176
DOI: 10.1109/lsp.2026.3716909 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Signal Processing Letters, 2026. Verfasst von Qi Xue und 4 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben.
Abstract (Originalsprache)
We address automated singing voice evaluation: given a teacher's demonstration and a student's recording of the same lyrics, the system produces character-level comparisons of pitch, rhythm, and six singing techniques, together with natural-language pedagogical feedback. We construct a dataset of 870 real-world teacher–student pairs and obtain expert-verified technique labels on a 246-song subset. We propose a two-stage system: the first stage extracts character-level timestamps, pitch contours, and technique labels via forced alignment, fundamental frequency estimation, and a technique extractor built on self-supervised music representations; the second stage feeds the structured representation into a fine-tuned large language model to generate diagnostic feedback. On expert-annotated data, the technique extractor achieves a character-level average F-score of 0.82, versus 0.01–0.10 for three state-of-the-art audio multimodal large language models; it is independently validated on an open-source benchmark with an event-level macro F-score of 0.83. On the feedback task, these models further exhibit confident hallucination—producing detailed, well-structured suggestions that are grounded in incorrectly identified problems—highlighting the limitations of end-to-end audio models for fine-grained singing evaluation.
Zitieren
APA
Xue, Q., Wang, Y., Zhou, Y., Duan, H., & Gao, X. (2026). Character-Level Singing Technique Detection and Evaluation: A Two-Stage Approach. IEEE Signal Processing Letters, 33, 3172-3176. https://doi.org/10.1109/lsp.2026.3716909
Vancouver
Xue Q, Wang Y, Zhou Y, Duan H, Gao X. Character-Level Singing Technique Detection and Evaluation: A Two-Stage Approach. IEEE Signal Processing Letters. 2026;33:3172-3176. doi:10.1109/lsp.2026.3716909
BibTeX
@article{Xue2026,
author = {Xue, Qi and Wang, Yixuan and Zhou, Yiquan and Duan, Haijun and Gao, Xin},
title = {Character-Level Singing Technique Detection and Evaluation: A Two-Stage Approach},
journal = {IEEE Signal Processing Letters},
year = {2026},
volume = {33},
pages = {3172--3176},
doi = {10.1109/lsp.2026.3716909}
}
EUVIMED