情緒分類器形式 - 基於低音聲部特徵萃取的和弦行進與風格演算法之音樂情緒分類研究

Judith C. Brown[26]利用 DTW(dynamic time warping)的技術，以及基頻追蹤

(Fundamental frequency tracking)，並配合殺人鯨的聲音語料庫，加以運算之後，來辨認殺人鯨的情緒分類狀況。

針對以上的技術發展，可以應用在其他各種層面。例如：轉換到其他各類動物及昆蟲上，建置個別的聲音語料庫，以便讓人類更加了解動物的情緒，協助動物照顧人員與研究專家認知動物的需求，拉近人類與動物之間的距離。

現今除了傳統的醫學治療之外，又有著音樂治療此種新型態的治療方式。某些原始民族的巫醫會在宗教儀式當中使用音樂吟唱、或是咒語來治療，姑且不討論其有無科學依據，但可以見得音樂能對人類的心理造成一定程度的療效。根據腦神經學者研究發現，人類的大腦裡原本就存在著音樂元素，許多臨床研究證明了音樂用來成功治療的案例。在十九世紀初期，許多病人對種種刺激都沒有任何反應，但唯獨對音樂有感受力。

因此、搭配音樂情緒的研究，再配合音樂治療的方法，可以利用不同風格的音樂、產生不同類型的情緒，針對不同的病症實施相對的療程，將是未來新一波的治療方式。

傳統上中醫師除了「望診、聞診、問診、切診」另外也可配合中醫脈搏偵測，利用極細微的聲音接收裝置，搭配上述資料庫的應用，可以基本的建置出寸、關、尺，各部再有浮、中、沉，所以 3*3 等所謂 3 部 9 候，共約 28 種的脈象，進一步的幫助中醫師判斷病情。

五、結論與延伸應用

而美國研究發現，聽自己喜歡聽的歌可以擴張血管，對心臟有相當助益。本研究也能針對使用者的習性，提供適合的音樂，使之心情愉悅，積極進取。

音樂所隱含的情緒包羅萬象，往往一首歌曲之中夾雜了許多情緒的成分，目前很難以用四分法去精準的區分出一首歌真正的情緒。在加上聽者會隨著人、事、時、地、物等等各因素層面的影響，導致各聽眾對同一首歌的情緒判別不一致，使得出來的結果客觀性十分強烈。

本研究提供一項做為分析音樂情緒的參考方法，提出以Bass辨認和絃行進，並參照幾種常用的和絃進行，給出數值變化，建構一分析音樂情緒的系統介面。只是使用上仍有些許限制，例如只能針對以Bass Line為伴奏的音樂，無法針對以Bass為主奏樂器分析，另外低音聲部的特徵萃取、系統中的樣板可以改用HMM(Hidden Markov Model)或 GMM(Gaussian Mixture Model)等方式訓練，及和絃的辨識率也有改善的空間。希望將來能加強辨識率，提高分析結果，廣泛應用在其他層面，如音樂治療等。

六、參考文獻

[1] 方銘健（1997）。藝術、音樂情感與意義。台北：全音。

[2] Reimer, B. (2005). “New brain research on emotion and feeling: Dramatic implications for music education.” The International Journal of Arts Education, 3(1), 8-36.

[3] Radocy, R. E., & Boyle, J. D. (2003). “Psychological foundations of musical behavior(4th ed.)”. Springfield, IL: Charles C. Thomas.

[4] Thayer, R. E., “The Biopsychology of Mood and Arousal”, Oxford University Press, (1989).

[5] Lie Lu, Member, IEEE, Dan Liu, and Hong-Jiang Zhang, Fellow, IEEE, “Automatic Mood Detection and Tracking of Music Audio Signals”, IEEE TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING, VOL. 14, NO. 1, JANUARY (2006).

[6] Juo-Han Chen, “Content Base Music Emotion Analysis and Recognition”, National Tsing Hua University,(2006).

[7] 黃靜芳, 吳舜文, ” 大學生音樂選曲與情緒反應之相關研究”, 國際藝術教育學刊,(2007)

[8] Tzanetakis, G. Cook, P. “Musical genre classification of audio signals” , IEEE Transactions on Speech and Audio Processing, Jul (2002).

[9] Yusuke Tsuchihashi,Tetsuro Kitahara, Haruhiro Katayose. “USING BASS-LINE FEATURES FOR CONTENT-BASED MIR”, Kwansei Gakuin

University,Japan,CrestMuse Project,CREST,JST,Japan, ISMIR (2008) – Session 5b – Feature Representation

[10] Kouhei Sumi,Katsutoshi Itoyama,Kazuyoshi Yoshii,Kazunori Komatani,Tetsuya Ogata,and Hiroshi G. Okuno, “AUTOMATIC CHORD RECOGNITION BASED ON PROBABILISTIC INTEGRATION OF CHORD TRANSITION AND BASS PITCH

ESTIMATION”, Dept. of Intelligence Science and Technology Graduate School of Informatics, Kyoto University, National Institute of Advanced Industrial Science and Technology (AIST),ISMIR (2008) – Session 1a – Harmony

[11] Masataka Goto. “A real-time music-scene-description system:predominant-F0 estimation for detecting melody and bass lines in real-world audio signals”, National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Umezono, Tsukuba, Ibaraki 305-8568, Japan Received 22 May 2002;received in revised form 9 May 2003;accepted 13 March 2004

[12] Jakob Abeßer, Hanna Lukashevich, Christian Dittmar, Gerald Schuller, “GENRE CLASSIFICATION USING BASS-RELATED HIGH-LEVEL FEATURES AND PLAYING STYLES”, Jakob Abeßer, Hanna Lukashevich, Christian Dittmar, Gerald Schuller, 10th International Society for Music Information Retrieval Conference (ISMIR 2009) [13] Emiru Tsunooy, Geroge Tzanetakisyy, Nobutaka Onoy, Shigeki Sagayama, “AUDIO

GENRE CLASSIFICATION USING RHYTHM AND BASS-LINE PATTERN

INFORMATION”, Graduate School of Information Science and Technology University of Tokyo, Japan, Computer Science Department University of Victoria, Canada(2009) [14] Emiru Tsunoo, Nobutaka Ono, Shigeki Sagayama. “MUSICAL BASS-LINE PATTERN

CLUSTERING AND ITS APPLICATION TO AUDIO GENRE CLASSIFICATION”

Graduate School of Information Science and Technology The University of Tokyo, Japan, 10th International Society for Music Information Retrieval Conference (ISMIR 2009)

[15] Cyril Laurier and Perfecto Herrera, “Mood Cloud : A Real-Time Music Mood Visualization Tool”, Music Technology Group Universitat Pompeu Fabra, Ocata 1, 08003 Barcelona, Spain,(2007)

[16] Yi-Hsuan Yang, Yu-Ching Lin, Heng-Tze Cheng, and Homer Chen. “Mr. Emo: Music Retrieval in the Emotion Plane” , National Taiwan University, International Multimedia

Conference, Proceeding of the 16th ACM international conference on Multimedia,(2008) [17] http://blogs.msdn.com/blogfiles/audiofool/WindowsLiveWriter/LouderSoundsBetter_12

855/FletcherMunson_EqualLoudness2.jpg

[18] Jouni Paulus, Anssi Klapuri,” Measuring the Similarity of Rhythmic Patterns”, Tampere University of Technology Institute of Signal Processing P.O. Box 553, FIN-33101 Tampere, Finland,(2002)

[19] Perfecto Herrera-Boyer, Geoffroy Peeters and Shlomo Dubnov, “Automatic

Classification of Musical Instrument Sounds”, 1MTG-IUA, Universitat Pompeu Fabra, Barcelona, Spain,IRCAM, Paris, France and The Hebrew University,Jerusalem, Israel, Journal of New Music Research, 2003

[20] J. Foote , ”Visualizing Music and Audio using Self-Similarity”, ACM Multimedia Conference,(1999).

[21] http://neural.cs.nthu.edu.tw/jang/books/audioSignalProcessing/dpDtw_chinese.asp [22] Donald J. Bemdt,JameCs lifford, “Using Dynamic Time Warping to Find Patterns in

Time Series”, Information Systems Department,Stern School of Business,New York University,44 West 4th Street,New York, New York 10012-1126, AAAI Technical Report WS-94-03. Compilation copyright © (1994)

[23] Hagen Kaprykowsky, Xavier Rodet, ” GLOBALLY OPTIMAL SHORT-TIME DYNAMIC TIME WARPING APPLICATION TO SCORE TO AUDIO ALIGNMENT”, Ircam – Centre Pompidou – Analyse Synthèse

[24] 浦田泰宏著，(2006)，和弦進行活用與演奏秘笈，孫孙虹、蕭良悌譯，典絃音樂文化國際事業有限公司

[25] http://allmusic.com/

[26] Judith C. Brown, “Automatic classification of killer whale vocalizations using dynamic time warping”, Physics Department, Wellesley College, Wellesley, Media Lab,

Massachusetts, Institute of Technology, (2006) Acoustical Society of America

在文檔中基於低音聲部特徵萃取的和弦行進與風格演算法之音樂情緒分類研究 (頁 45-50)