Enhanced Feature Summarizing for Effective Cover Song Identification
Enhanced Feature Summarizing for Effective Cover Song Identification
复制标题
增强的特征总结可有效识别翻唱歌曲
DOI：
10.1109/taslp.2019.2942157
复制
发表时间：
2019-09
期刊：
影响因子：
--
通讯作者：
Chen Ning
中科院分区：
文献类型：
--
作者：
Hu Jingyi;Chen Ning
最近提出了基于自相似性分析的特征摘要技术（SuCo）来提高封面歌曲识别（CSI）的时间和存储效率。本文对SuCo模型的特征总结和交叉相似度计算策略进行如下修改，以提高其识别精度。在特征总结阶段，首先采用Hubness Reduction（HR）策略来减少特征子序列社区中可能存在的“Hubness”现象，从而影响检索效果。然后，引入最初在生物学中提出的用于提高基因功能预测准确性的网络增强（NE）技术，以减少由于特征提取和相似性测量的限制以及固有的音乐和声学变化而引起的自相似性网络中的噪声。在交叉相似度计算阶段，首先将参考文献的汇总代表特征子序列进行级联以获得其组合代表特征。然后，考虑到非线性递推性质对于描述基于旋律感知的相似度很重要，采用Qmax来衡量参考的组合代表特征与查询的未概括特征序列之间的相似度。在具有 5 种特征和 2 种代表性特征子序列选择方法的 4 个开放 CSI 数据集上进行的大量实验验证了： i）所提出的方案在检索效果上优于 SuCo 模型。 ii）上述每项修改都有助于提高所提出方案的性能。 iii）所提出的方案实现了高度泛化。
Self-similarity analysis-based feature summarizing technique (SuCo) was proposed recently to improve the time and memory efficiency of Cover Song Identification (CSI). In this paper, both the feature summarizing and the cross-similarity calculating strategies of the SuCo model are modified as follows to enhance its identification accuracy. At the feature summarizing stage, first, the Hubness Reduction (HR) strategy is adopted to reduce the possible ‘Hubness’ phenomenon existing in the feature subsequence community, which may affect the retrieval effectiveness. Then, the Network Enhancement (NE) technique, which was originally proposed in biology to improve gene-function prediction accuracy, is introduced to reduce the noise in the self-similarity network caused by the limitation of feature extraction and similarity measuring, and the inherent musical and acoustic variations. At the cross-similarity calculating stage, first, the summarized representative feature subsequences of the reference are concatenated to obtain its combined representative feature. Then, considering that the nonlinear recurrence property is important for describing the melody perception-based similarity, Qmax is adopted to measure the similarity between the combined representative feature of the reference and the unsummarized feature sequence of the query. Extensive experiments carried out on four open CSI datasets with 5 types of features and 2 kinds of representative feature subsequence choosing methods verify that: i) The proposed scheme outperforms the SuCo model in retrieval effectiveness. ii) Each of the above modifications contributes to the performance enhancement of the proposed scheme. iii) The proposed scheme achieves high generalization.
登录
查看更多内容
翻唱歌曲识别的相似度融合方案
影响因子：
1.1
作者：
Chen, Ning;Xiao, Hai-dong
通讯作者：
Xiao, Hai-dong
CI-SNF：利用上下文信息改进基于 SNF 的信息检索
影响因子：
18.6
作者：
Chen Ning
通讯作者：
Chen Ning
DOI：
10.1109/taslp.2015.2409735
发表时间：
2015-04
期刊：
IEEE/ACM Transactions on Audio, Speech, and Language Processing
影响因子：
--
作者：
Chung-Che Wang;J. Jang
通讯作者：
Chung-Che Wang;J. Jang
影响因子：
1.1
作者：
Flexer A;Stevens J
通讯作者：
Stevens J
DOI：
--
发表时间：
2016-12
期刊：
ArXiv
影响因子：
--
作者：
Filip Korzeniowski;G. Widmer
通讯作者：
Filip Korzeniowski;G. Widmer