Toward Multi-modal Music Emotion Classification

Yang, Yi-Hsuan; Lin, Yu‐Ching; Cheng, Heng-Tze; Liao, I-Bin; Ho, Yeh-Chin; Chen, Homer H.

doi:10.1007/978-3-540-89796-5_8

Cited by 65 publications

(38 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…by Wang, Zhang, and Zhu (2004), hand-selection of prototypical excerpts, e.g. by Liu et al (2006), limitation to one language-thus simplifying lyrics analysis (as done in existing work on lyrics mood analysis (Laurier, Grivolla, & Herrera, 2008;Yang, Lin, Cheng, et al 2008), or limitation to songs where existence of all on-line information was ensured ahead, e.g. by Laurier et al (2008), which all could not be made in a real-life application.…”

Section: Innovations In This Workmentioning

confidence: 99%

‘Mister D.J., Cheer Me Up!’:Musical and Textual Features for Automatic Mood Classification

Schuller

Hage

Schuller³

et al. 2010

Journal of New Music Research

View full text Add to dashboard Cite

Mass consumption of large collections of digital music asks for efficient and intuitive ways of organization. In this article, a system is presented which recognizes the evoked music mood on the basis of a wide variety of features, closely sticking to real world conditions. A twodimensional mood model is discussed in which moods resemble binary values for arousal and valence and an easy and thus user-friendly method is presented through which a fuzzy seven-class mood cluster is deducted. The songs of the 'Twenty Years of MTV Europe Most Wanted' music database consisting of recorded pop music tracks serve for evaluation of three groups of features: firstly, traditional features such as rhythm and tonal features, zero crossing rate, cepstral, and MPEG-7 Low Level Descriptors for audio content are extracted. Secondly, lyrics, chord sequences, and genre data are obtained from on-line sources. Thirdly, from all these, the high-level features, musical mode, and as a novel feature, the suited ballroom dance style, are created automatically. The features selected are data-driven, and Support Vector Machines are used for classification. Prediction accuracies of 77.4% for arousal and 72.9% for valence as well as 71.8% (including neighbours) for the seven-class cluster model are obtained pre-serving realism in terms of non-prototypical music selection and feature extraction throughout.

show abstract

Section: Innovations In This Workmentioning

confidence: 99%

‘Mister D.J., Cheer Me Up!’:Musical and Textual Features for Automatic Mood Classification

Schuller

Hage

Schuller³

et al. 2010

Journal of New Music Research

View full text Add to dashboard Cite

show abstract

“…Classification of music with the intention to facilitate its retrieval from large collections is presented in [37]. They employ audio feature extraction to analyze the contents of musical clips, and process those features by a multimodal fusion process.…”

Section: Classification Of Music Clips According To Emotionsmentioning

confidence: 99%

Analysis of psychoacoustic responses to digital music for enhancing autonomous creative systems

López-Ortega

Franco-Arcega

2015

Applied Acoustics

View full text Add to dashboard Cite

“…Accordingly, this approach is adopted by numerous works [13,14,11,2,8]. Yang and Lee [13], in one of the earlier works in the field, proposed the combination of lyrics and a number of audio features in order to maximise the classification accuracy and minimise the mean error.…”

Section: Mood Classification Using Audio and Lyricsmentioning

confidence: 99%

“…Nevertheless, the significantly small data corpus (145 songs with lyrics) made the work exploratory to make safe conclusions. In their work, Yang et al [14], extracted a number of low-level acoustic features from a 30 second part of the song that in addition to the lyrics features produced by 3 different approaches (Uni-gram, Probabilistic Latent Semantic Analysis & Bi-gram) are combined by 3 fusion methods. Therein, songs are classified in four categories following the Russell model [15] to conclude that the use of textual features offers a significant accuracy amelioration of the methods examined.…”

Section: Mood Classification Using Audio and Lyricsmentioning

confidence: 99%

Mood Classification Using Lyrics and Audio: A Case-Study in Greek Music

Brilis

Gkatzou

Koursoumis

et al. 2012

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Abstract. This paper presents a case-study of the effectiveness of a trained system into classifying Greek songs according to their audio characteristics or/and their lyrics into moods. We examine how the usage of different algorithms, featureset combinations and pre-processing parameters affect the precision and recall percentages of the classification process for each mood model characteristic. Experimental results indicate that the current selection of features offers accuracy results, the superiority of lyrics content over generic audio features as well as potential caveats with current research in Greek language stemming pre-processing methods.

show abstract

Toward Multi-modal Music Emotion Classification

Cited by 65 publications

References 18 publications

‘Mister D.J., Cheer Me Up!’:Musical and Textual Features for Automatic Mood Classification

‘Mister D.J., Cheer Me Up!’:Musical and Textual Features for Automatic Mood Classification

Analysis of psychoacoustic responses to digital music for enhancing autonomous creative systems

Mood Classification Using Lyrics and Audio: A Case-Study in Greek Music

Contact Info

Product

Resources

About