https://scholars.lib.ntu.edu.tw/handle/123456789/607419
標題: | On the preparation and validation of a large-scale dataset of singing transcription | 作者: | Wang J.-Y JYH-SHING JANG |
關鍵字: | Automatic singing transcription;Dataset preparation;Dataset validation;Music information retrieval;Signal processing;Fine tuning;Large-scale dataset;Large dataset | 公開日期: | 2021 | 卷: | 2021-June | 起(迄)頁: | 276-280 | 來源出版物: | ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings | 會議論文: | 2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021 | 摘要: | This paper proposes a large-scale dataset for singing transcription, along with some methods for fine-tuning and validating its contents. The dataset is named MIR-ST500, which consists of more than 160,000 notes from 500 pop songs. To create this large-scale dataset, we set some labeling criteria and ask non-experts to label notes. We also perform some adjustments on the annotation to correct minor errors. Finally, to validate the dataset, we train a singing transcription model on MIR-ST500 dataset and evaluate it on various datasets. The result shows that we can certainly construct a better singing transcription model for various purposes using MIR-ST500, which is properly labeled and validated. ? 2021 IEEE |
URI: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85115050788&doi=10.1109%2fICASSP39728.2021.9414601&partnerID=40&md5=70bedf26b4c297347e361f22fd12f43a https://scholars.lib.ntu.edu.tw/handle/123456789/607419 |
ISSN: | 15206149 | DOI: | 10.1109/ICASSP39728.2021.9414601 |
顯示於: | 資訊工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。