Learning Low-Dimensional Embeddings of Audio Shingles for Cross-Version Retrieval of Classical Music
Learning Low-Dimensional Embeddings of Audio Shingles for Cross-Version Retrieval of Classical Music
复制标题
DOI：
10.3390/app10010019
复制
发表时间：
2019-12
期刊：
影响因子：
--
通讯作者：
Frank Zalkow;Meinard Müller
中科院分区：
文献类型：
--
作者：
Frank Zalkow;Meinard Müller
跨版本音乐检索的目标是使用一个简短的查询音频片段来识别给定音乐片段的所有版本。一种特别适用于西方古典音乐的先前方法是基于使用色度特征的短序列的最近邻搜索，也称为音频带状疱疹。从效率的角度来看，索引和降维是重要的方面。在本文中，我们扩展了以前的工作，采用两种嵌入技术，一种是基于经典的主成分分析，另一种是基于神经网络与三重损失。此外，我们报告了系统进行的实验与西方古典音乐录音，并讨论了检索质量和嵌入维数之间的权衡。作为一个主要的结果，我们表明，使用神经网络，可以减少音频瓦片从240到不到8个维度，只有适度的损失在检索精度。此外，我们提出了扩展的实验与不同大小的数据库和不同的查询长度来测试的可扩展性和推广的降维方法。我们还提供了一个更详细的视图到检索问题，通过分析出现在最近邻搜索的距离。
Cross-version music retrieval aims at identifying all versions of a given piece of music using a short query audio fragment. One previous approach, which is particularly suited for Western classical music, is based on a nearest neighbor search using short sequences of chroma features, also referred to as audio shingles. From the viewpoint of efficiency, indexing and dimensionality reduction are important aspects. In this paper, we extend previous work by adapting two embedding techniques; one is based on classical principle component analysis, and the other is based on neural networks with triplet loss. Furthermore, we report on systematically conducted experiments with Western classical music recordings and discuss the trade-off between retrieval quality and embedding dimensionality. As one main result, we show that, using neural networks, one can reduce the audio shingles from 240 to fewer than 8 dimensions with only a moderate loss in retrieval accuracy. In addition, we present extended experiments with databases of different sizes and different query lengths to test the scalability and generalizability of the dimensionality reduction methods. We also provide a more detailed view into the retrieval problem by analyzing the distances that appear in the nearest neighbor search.