Jingju Music Scores Collection

The Jingju Music Scores Collection (JMSC) is part of the Jingju Music Corpus created in the CompMusic project. Created with the purpose of the melodic research of jingju singing lines, it contains 108 MusicXMLscores, covering 1084 melodic lines. In terms of the jingju musical system, it covers the following elements: role type: laosheng, dan, laodan shengqiang: erhuang, xipi, nanbangzi, sipingdiao banshi: manban, sanyan, zhongsanyan, kuaisanyan, yuanban, erliu, kuai'erliu, liushui, kuaiban The provided README.md file contains details about the content of the JMSC in terms of scores, and in terms of lines. The scores are sourced from printed editions and created with MuseScore 2.1.0, including the lyrics. The original jianpu notation is transnotated into staff notation. All the scores contain a separated staff for the accompaniment (jinghu) line. Content of the JMSC The JMSC contains the following three folders: The MusicXML folder contains the MusicXML scores and the scores_data.csv and lines_data.csv files containing metadata and annotations. The MuseScore folder contains the scores created with MusicScore in its original format, from which the MusicXML scores were exported. Annotations and metadata The MusicXML folder of the JMSC contains the following two files with annotations and metadata. The scores_data.csv file contains metadata and annotations for each score. The included information consists of the following data: title of the aria (in Chinese), role type, shengqiang, banshi, 'yes' or 'no' the original score contained a separated line for the accompaniment, the reference to the printed source, and the MusicBrainz ID of related recordings. The lines_data.csv file contains annotations for each line in the JMSC. The included information consists of the following data: role type, shengqiang, banshi, line type, lyrics of the line, starting offset of the line, ending offset of the line, linguistic tones, lyrics of the first line section, starting offset of the first line section, ending offset of the first line section, lyrics of the second line section, starting offset of the second line section, ending offset of the second line section, lyrics of the third line section, starting offset of the third line section, and ending offset of the third line section. Regarding line types, 's' stands for opening line in xipi, 's1' for the long opening line type, and 's2' for the short opening line type in erhuang, and 'x' for closing line. Using the JMSC Since the scores are sourced from copyrighted printed editions, they can be only shared for non commercial research purposes (see below). The metadata and annotations files are released under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license. Different datasets from the JMSC have been used in the following publications: Gong, Rong, Nicolas Obin, Georgi Dzhambazov, and Xavier Serra (2017) "Score-informed syllable segmentation for jingju a cappella singing voice with mel-frequency intensity profiles." In Proceedings of the 7th International Workshop on Folk Music Analysis (FMA 2017), Málaga, Spain, June 14–16, pp. 107–113. https://doi.org/10.5281/zenodo.556820 Gong, Rong, Jordi Pons and Xavier Serra (2017) "Audio to Score Matching by Combining Phonetic and Duration Information." In Proceedings of the 18th International Society for Music Information Retrieval Conference (ISMIR 2017), Suzhou, China, October 23–27, 428–434. https://arxiv.org/abs/1707.03547 Gong, Rong, and Xavier Serra (2018) "Singing voice phoneme segmentation by hierarchically inferring syllable and phoneme onset positions." In Interspeech 2018, Hyderabad, India, September 2–6, 716–720. https://arxiv.org/abs/1806.01665 Caro Repetto, Rafael (2018) The musical dimension of Chinese traditional theatre: An analysis from computer aided musicology. PhD thesis, Universitat Pompeu Fabra, Barcelona, Spain. http://hdl.handle.net/10803/665357 Contact For any questions or comments about the JMSC, please contact Rafael Caro Repetto ([e-mail]) Acknowledgements The creation of the JMSC is funded by the European Research Council under the European Union’s Seventh Framework Program (FP7/2007-2013), as part of the CompMusic project (ERC grant agreement 267583).