MTG-QBH: Query By Humming dataset

This dataset includes 118 recordings of sung melodies. The recordings were made as part of the experiments on Query-by-Humming (QBH) reported in the following article: J. Salamon, J. Serrà and E. Gómez, "Tonal Representations for Music Retrieval: From Version Identification to Query-by-Humming", International Journal of Multimedia Information Retrieval, special issue on Hybrid Music Information Retrieval, In Press (accepted Nov. 2012). The recordings were made by 17 different subjects, 9 female and 8 male, whose musical experience ranged from none at all to amateur musicians. Subjects were presented with a list of songs out of which they were asked to select the ones they knew and sing part of the melody. The subjects were aware that the recordings will be used as queries in an experiment on QBH. There was no restriction as to how much of the melody should be sung nor which part of the melody should be sung, and the subjects were allowed to sing the melody with or without lyrics. The subjects did not listen to the original songs before recording the queries, and the recordings were all sung a capella without any accompaniment nor reference tone. To simulate a realistic QBH scenario, all recordings were done using a basic laptop microphone and no post-processing was applied. The duration of the recordings ranges from 11 to 98 seconds, with an average recording length of 26.8 seconds. In addition to the query recordings, three meta-data files are included, one describing the queries and two describing the music collections against which the queries were tested in the experiments described in the aforementioned article. Whilst the query recordings are included in this dataset, audio files for the music collections listed in the meta-data files are NOT included in this dataset, as they are protected by copyright law. If you wish to reproduce the experiments reported in the aforementioned paper, it is up to you to obtain the original audio files of these songs. All subjects have given their explicit approval for this dataset to be made public. Please Acknowledge MTG-QBH in Academic Research Using this dataset When the MTG-QBH dataset is used for academic research, we would highly appreciate if scientific publications of works partly based on the MTG-QBH dataset cite the above publication. We are interested in knowing if you find our datasets useful! If you use our dataset please email us at [e-mail] and tell us about your research. https://www.upf.edu/web/mtg/mtg-qbh

本数据集包含118段人声演唱旋律录音。这些录音是下述论文中报道的哼唱检索（Query-by-Humming, QBH）实验的一部分：J. Salamon、J. Serrà与E. Gómez所著《面向音乐检索的音调表示：从版本识别到哼唱检索》，发表于《国际多媒体信息检索期刊》（International Journal of Multimedia Information Retrieval）混合音乐信息检索特刊，已录用（2012年11月录用）。

录音由17名不同受试者完成，其中9名女性、8名男性，受试者的音乐素养从完全无基础到业余音乐人不等。研究人员向受试者提供了一份歌曲列表，要求受试者从中挑选自己熟悉的歌曲，并演唱其中的部分旋律。受试者知晓这些录音将被用作哼唱检索实验的查询样本。对于演唱的旋律长度与段落均无限制，受试者可选择带歌词或不带歌词演唱旋律。受试者在录制查询样本前并未聆听原曲，所有录音均为无伴奏清唱（a capella），未添加任何伴奏或参考音高。

为模拟真实的哼唱检索场景，所有录音均使用普通笔记本电脑麦克风录制，且未进行任何后期处理。录音时长介于11秒至98秒之间，平均时长为26.8秒。

除查询样本录音外，本数据集还包含3个元数据（meta-data）文件：1个用于描述查询样本，另外2个用于描述上述论文实验中用于测试查询样本的音乐库。尽管本数据集包含查询样本录音，但元数据文件中列出的音乐库音频文件并未包含在内，原因是这些音频受版权法保护。若您希望复现上述论文中报道的实验，需自行获取这些歌曲的原始音频文件。

所有受试者均已明确同意本数据集公开。

学术研究使用本数据集的致谢说明：若基于MTG-QBH数据集开展研究并撰写学术成果，恳请引用上述发表的论文，我们将不胜感激。

我们期待了解您是否认为本数据集具有实用价值！若您使用本数据集，请发送邮件至[e-mail]，与我们分享您的研究工作。更多信息请访问：https://www.upf.edu/web/mtg/mtg-qbh

搜集汇总

背景与挑战

背景概述

MTG-QBH是一个用于查询哼唱（QBH）研究的公开数据集，包含118段由17名不同音乐背景的受试者录制的哼唱旋律录音，录音时长平均26.8秒，模拟了真实场景下的无伴奏录制。数据集还提供了查询和音乐集合的元数据文件，但音乐集合的原始音频文件因版权问题未包含，主要用于音乐信息检索领域的实验和学术研究。