상황인지 음악추천을 위한 음악 분위기 검출 = Detection of Music Mood for Context-aware Music Recommendation

다국어 초록 (Multilingual Abstract)

To provide context-aware music recommendation service, first of all, we need to catch music mood that a user prefers depending on his situation or context. Among various music characteristics, music mood has a close relation with people’s emotion. Based on this relationship, some researchers have studied on music mood detection, where they manually select a representative segment of music and classify its mood. Although such approaches show good performance on music mood classification, it's difficult to apply them to new music due to the manual intervention. Moreover, it is more difficult to detect music mood because the mood usually varies with time.
To cope with these problems, this paper presents an automatic method to classify the music mood. First, a whole music is segmented into several groups that have similar characteristics by structural information. Then, the mood of each segments is detected, where each individual's preference on mood is modelled by regression based on Thayer's two-dimensional mood model. Experimental results show that the proposed method achieves 80% or higher accuracy.

국문 초록 (Abstract)

상황인지 음악추천 서비스를 제공하기 위해서는 무엇보다 상황 또는 문맥에 따라 사용자가 선호하는 음악의 분위기를 파악할 필요가 있다. 음악 분위기 검출에 대한 기존 연구의 대부분은 수작업으로 대표구간을 선정하고, 그 구간의 특징을 이용하여 분위기를 판별한다. 이러한 접근 방법은 분류 성능이 좋은 반면 전문가의 간섭을 요구하기 때문에 새로운 음악에 대해서는 적용하기 어렵다. 더욱이, 곡의 진행에 따라 음악 분위기가 달라지기 때문에 음악의 대표 분위기를 검출하는 것이 더욱 어려워진다.
본 논문에서는 이러한 문제점들을 보완하기 위해 음악 분위기를 자동으로 판별하는 새로운 방법을 제안하였다. 먼저 곡 전체를 구조적 분석 방법을 통하여 비슷한 특성을 갖는 세그먼트들로 분리한 후 각각에 대해 분위기를 판별한다. 그리고 세그먼트별 분위기 파악 시 Thayer 의 2 차원 분위기 모델에 기초한 회귀분석 방법으로 개인별 주관적 분위기 성향을 모델링하였다. 실험결과, 제안된 방법이 80% 이상의 정확도를 보였다.

참고문헌 (Reference)

1 S. Abdallah, "Theory and Evaluation of a Bayesian Music Structure Extractor" 2005

2 P.R. Farnsworth, "The Social Psychology of Music" The Dryden Press 1958

3 R.E. Thayer, "The Biopsychology of Mood and Arousal" Oxford University Press 1989

4 M. Levy, "Structural Segmentation of Musical Audio by Constrained Clustering" 16 : 318-326, 2008

5 M. Goto, "SmartMusicKIOSK:Music Listening station with Chorus-Search Function" 31-40, 2003

6 Y. K. Kim, "Singer Identification in Popular Music Recordings Using Voice Coding Features" 2002

7 H. Katayose, "Sentiment Extraction in Music" 2 : 1083-1087, 1998

8 Y. Feng, "Popular Music Retrieval by Detecting Mood"

9 Y. Shiu, "Musical Structure Analysis Using Similarity Matrix and Dynamic Programming" 3 : 398-409, 2005

10 G. Tzanetakis, "Musical Genre Classification of Audio Signals" 10 (10): 293-302, 2002

11 D. Jiang, "Music Type Classification by Spectral Contrast Feature" 1 : 113-116, 2002

12 J. Paulus, "Music Structure Analysis by Finding Repeated Parts"

13 Y.H. Yang, "Music Emotion Recognition:the Role of Individuality" 13-21, 2007

14 Y.H. Yang, "Music Emotion Classification:a Fuzzy Approach"

15 J. J. Burred, "Hierarchical Automatic Audio Signal Classification" 52 (52): 357-365, 2004

16 D. Liu, "Form and Mood Recognition of Johann Strauss's Waltz Centos" 12 (12): 587-593, 2003

17 M. Levy, "Extraction of High-Level Musical Structure From Audio Data and Its Application to Thumbnail Generation" 5 : 13-16, 2006

18 D. Hevner, "Experimental Studies of the Elements of Expression in Music" 48 : 246-268, 1936

19 T. Li, "Detecting Emotion in Music" 239-240, 2003

20 G. Peeters, "Deriving Musical Structure from Signal Analysis for Music Audio Summary Generation: "Sequence" and "State" Approach" Springer-Verlag 2771 : 143-166, 2004

21 T. Li, "Content-based Music Similarity Search and Emotion Detection" 5 : 705-708, 2004

22 T. Zhang, "Automatic Singer Identification" IEEE CS Press 2003

23 X. Shao, "Automatic Music Summarization Based on Music Structure Analysis" 2 : 1169-1172, 2005

24 L. Lu, "Automatic Mood Detection and Tracking of Music Audio Signals" 14 : 5-18, 2006

25 T. Zhang, "Automatic Generation of Music Thumbnails" 228-231, 2007

26 L. Lu, "Automated Extraction of Music Snippets" 140-147, 2003

27 Y.H. Yang, "A Regression Approach to Music Emotion Recognition" 16 : 448-457, 2008

28 J. J. Burred, "A Hierarchical Approach to Automatic Musical Genre Classification" 2003

29 T. Tolonen, "A Computationally Efficient Multipitch Analysis Model" 8 : 708-716, 2000

30 T. Li, "A Comparative Study on Content-based Music Genre Classification" 282-289, 2003