Chroma特征的鲁棒性验证

计算机科学 ›› 2014, Vol. 41 ›› Issue (Z6): 24-28.

Chroma特征的鲁棒性验证

张秀,李念祖,李伟

复旦大学计算机学院上海201203;复旦大学计算机学院上海201203;复旦大学计算机学院上海201203

出版日期:2018-11-14 发布日期:2018-11-14
基金资助:
本文受863国家自然科学基金项目(2011AA01A109),国家自然科学基金(61171128)资助

Verification for Robustness of Chroma Feature

ZHANG Xiu,LI Nian-zu and LI Wei

Online:2018-11-14 Published:2018-11-14

摘要/Abstract

摘要： 基于内容的多版本音乐识别是近些年来音乐信息检索领域一个比较热门的研究课题。考虑到多版本音乐可能在节奏、速度、音调、音色以及结构等方面的变化,该研究的关键在于选取能反映音乐主要旋律走向的相对稳定的音频特征,在不同的音乐版本之间进行相似度的比较。Chroma特征反映了音频能量在各个音调类间的相对分布,考虑了和声信息、与音色无关、对噪声鲁棒,所以成为多数多版本音乐识别算法使用的特征。通过设计和实验,探究不同的音频干扰形式对Chroma特征的影响,就Chroma特征对音调无关因素的鲁棒性进行验证。

关键词: Chroma特征,鲁棒性,音调不变性,多版本音乐识别中图法分类号TP399文献标识码A

Abstract: Content-based cover song detection is a hot research program in the field of music information retrieval in recent years．With the consideration that the different versions of music may change in many aspects,such as tempo,speed,pitch,timbre and structure,the key factor in this research is to extract robust audio feature,which can represent songs’ main melody progress,and compare their similarities between cover songs．Chroma features express the distribution of audio energy among different pitch class,consider presence of harmonics ,timbre independence and robustness to noise,so are used by a lot of cover song detection systems．Based on the theory,we designed and conducted the experiment to explore the effect of different forms of audio interference,and verified Chroma’s robustness in the aspect of pitch invariance.

Key words: Chroma feature,Robustness,Pitch invariance,Cover song detection

张秀,李念祖,李伟. Chroma特征的鲁棒性验证[J]. 计算机科学, 2014, 41(Z6): 24-28. https://doi.org/

ZHANG Xiu,LI Nian-zu and LI Wei. Verification for Robustness of Chroma Feature[J]. Computer Science, 2014, 41(Z6): 24-28. https://doi.org/

参考文献

[1] 李伟,汪竹蓉,李晓强,等．数字音频认证研究综述[J]．计算机科学,2009,6(10):21-24
[2] Ahonen T E．Combining Chroma Features For Cover Version Identification[C]∥Proc．International Symposium on Music Information Retrieval (ISMIR)．Utrecht,Netherlands,2010:165-170
[3] Bartsch M A,Wakefield G H．Audio thumbnailing of popularmusic using chroma-based representations[J]．IEEE Transactions on Multimedia,2005,7(1):96-104
[4] Brown J C．Calculation of a constant Q spectral transform[J]．The Journal of the Acoustical Society of America,1991,89:425
[5] Cho T,Bello J P．A feature smoothing method for chord recognition using recurrence plots[C]∥Proceedings of the 12th International Society for Music Information Retrieval Conference．2011:651-656
[6] Csurka G,Dance C,Fan L,et al．Visual categorization with bags of keypoints[C]∥Workshop on statistical learning in computer vision,ECCV．2004,1:22
[7] Ellis D P W,Cotton C V．The 2007labrosa cover song detection system[C]∥MIREX 2007．2007
[8] Ellis D P W,Poliner G E．Identifying ‘cover songs' with chroma features and dynamic programming beat tracking[C]∥IEEE International Conference on Acoustics,Speech and Signal Proces-sing,2007(ICASSP 2007)．IEEE,2007,4
[9] Foote J．Arthur:Retrieving orchestral music by long-term struc-ture[C]∥International Symposium on Music Information Retrieval．2000,1
[10] Fujishima T．Apparatus and method for recognizing musicalchords:U.S．Patent 6,7,502[P]．2000-5-2
[11] Gómez E．Tonal description of music audio signals[C]∥Unpublished doctoral dissertation,Universitat Pompeu Fabra．Barcelona,Spain,2006
[12] Gómez E,Herrera P．The song remains the same:Identifyingversions of the same piece using tonal descriptors[C]∥Procee-dings of 7th Intl．Conference on Music Information Retrieval．2006
[13] Muller M,Ewert S．Towards timbre-invariant audio features for harmony-based music[J]．IEEE Transactions on Audio,Speech,and Language Processing,2010,18(3):649-662
[14] Yang C．Music database retrieval based on spectral similarity[R]．Stanford,2001

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed