臺灣博碩士論文加值系統

由於多媒體解碼晶片的進步，以及嵌入式系統的快速發展，具有大容量的MP3音樂硬體播放器已成為流行，而且其產品隨處可見。由於MP3硬體儲存空間的不斷提高，要在大量MP3歌曲中，點選歌曲來播放成為一項新需求。本文提出一種在嵌入式系統上使用哼唱的方式，來對MP3音樂做查詢的系統實作架構。首先我們分析MP3音樂的原始資料串流，由壓縮領域中擷取出MP3特徵值，接著我們考慮嵌入式系統運算的特性，以及樂理的基礎來訂定特徵向量，盡量去除一些不必要的特徵值來減少特徵值維度，此外我們提出一個特徵快速過濾法，使得在查詢時只需跟相近的特徵索引做比對，以降低查詢比對效能的耗費，使其能在嵌入式系統硬體要求限制條件下，能有效地完成MP3音樂相似性查詢之執行。

The advancement of multimedia SOC (system on chip) is due to embedded system in progress. Now MP3 hardware player with large capacity has already become popular and its product is seen everywhere. The storage space of MP3 player is improvement so the easy way to select the songs which we want to listen is very important. In this paper, we propose the implementation of a humming-base MP3 music retrieving system on the embedded platform. In our approach, extracting the MDCT (modified discrete cosine transform coefficients) vector coefficients from the MP3 decoder are used to compute MP3 features. Then we consider the operation of the embedded system and the foundation of the music theory to reduce MP3 feature dimensions. In addition the fast vector filter makes retrieving MP3 music only with the closer MP3 features. For this reason we implement a humming-base MP3 music retrieving system under the hardware limiting conditions in the embedded platform.

摘要 I
ABSTRACT II
誌謝 III
內文目錄 IV
圖目錄 VI
表目錄 VIII
1. 序論 1
2. MP3音樂哼唱式查詢系統架構 4
2.1. MP3音樂索引 4
2.2. 特徵值計算 5
2.3. 哼唱查詢比對 5
3. MP3樂句分段 6
4. MP3相似性比對特徵值的擷取 7
4.1. MP3音訊編碼 7
4.2. 特徵向量擷取 7
4.3. MP3編碼結構與程式流程 9
4.4. MP3音訊解碼程式流程 10
5. 特徵值計算 12
5.1. 特徵擷取頻率範圍 12
5.2. 樂音特徵計算 13
6. 以音高分布特徵進行MP3樂句快速過濾比對的方法 15
6.1. 樂音音高分布的特徵向量的設計 15
6.2. 樂句音高分布向量的編碼 16
6.3. 相似性比對 17
7. 系統實作 18
7.1. 嵌入式平台硬體環境 19
7.2. 系統運作 20
8. 實驗 22
8.1. 實驗樣本 22
8.2. 影響準確率因素實驗 23
8.3. 樂音個數實驗 23
8.4. 低頻截止頻帶實驗 24
8.5. 高頻載止頻帶實驗 25
8.6. 嵌入式系統特徵值比對準確率實驗 25
8.7. 嵌入式系統特徵值比對效能實驗 26
8.8. 實驗結果綜合討論 28
9. 總結 29
10. 參考文獻 30
附錄A 使用PLATFORM BUILDER建制WINCE OS說明 35
附錄B MP3音樂哼唱查詢系統在嵌入式平台之使用方式 38

[1]Akinori Ito, et al. “Comparison of Features for DP-Matching Based Query-by-Humming System,” In International Symposium on Music Information Retrieval 2004.
[2]Anna Lubiw and Luke Tanur “Pattern Matching in Polyphonic Music as A Weighted Geometric Translation Problem,” In International Symposium on Music Information Retrieval 2004.
[3]Batke, J.-M, et al. “Evaluation of Distance Measures for MPEG-7 Melody Contours,” IEEE 6th Workshpo on Multimedia Signal Processing 2004.
[4]Brandenburg, K. and G. Stoll, “ISO-MPEG-1 Audio: A Generic Standard for Coding of High Quality Digital Audio,” Journal of the Audio Engineering Society, Vol. 42, No. 10, Oct 1994, pp. 780-792.
[5]Christopher Raphael “A Hybrid Graphical Model for Aligning Polyphonic Audio with Musical Scores,” In International Symposium on Music Information Retrieval 2004.
[6]Craig Stuart Sapp, Yi-Wen Liu and Eleanor Selfridge-Field “Search-Effectiveness Measures for Symbolic Music Queries in Very Large Databases,” In International Symposium on Music Information Retrieval 2004.
[7]Daniel Mullensiefen and Klaus Frieler “Optimizing Measures of Melodic Similarity for The Exploration of A Large Folk Song Database,” In International Symposium on Music Information Retrieval 2004.
[8]Roger B. Dannenberg, et al. ”The MUSART Testbed for Query-by- Humming Evaluation,” In International Symposium on Music Information Retrieval 2003.
[9]J. Stephen Downie “Toward the Scientific Evaluation of Music Information Retrieval Systems,” In International Symposium on Music Information Retrieval 2003.
[10]J. Stephen Downie, Joe Futrelle and David Tcheng“The International Music Information Retrieval Systems Evaluation Laboratory: governance, Access and Security,” In International Symposium on Music Information Retrieval 2004.
[11]Erdem Unal, Shrikanth S. Narayanan and Elaine Chew“A Statistical Approach to Retrieval under User-Dependent Uncertainty in Query-by-Humming Systems,” In Proceedings of the 12th Annual ACM International Conference on Multimedia 2004.
[12]Foote, J. “Content-Based Retrieval of Music and Audio,” Multimedia Storage and Archiving Systems II, Proc. SPIE, Vol. 3229, pp 138-147.
[13]Foote, J. “An Overview of Audio Information Retrieval,” ACM Multimedia Systems, Vol. 7, pp 2-10, Jan 1999.
[14]Ghias, A., Logan, H., Chamberlin, D., And Smith, B. C., “Query by Humming: Musical Information Retrieval in an Audio database,” in Proc. Of Third ACM International Conference on Multimedia, 1995, pages 231~236.
[15]Goto, M. “Music Scene Description Project: Toward Audio-Based Real-Time Music Understanding,” In International Symposium on Music Information Retrieval 2003.
[16]ISO/IEC 11172-3:1993, “Information Technology – Coding of Moving Pictures and Associated Audio for Digital Storage Media at up to about 1.5 Mbit/s Part 3: Audio.”
[17]Jana Eggink and Guy J. Brown “Extracting Melody Lines from Complex Audio,” In International Symposium on Music Information Retrieval 2004.
[18]Roger Jang, J. S. and H. R. Lee, “Hierarchical Filtering Method for Content-based Music Retrieval via Acoustic Input,” In Proc. ACM Multimedia 2001.
[19]J.S.R. Jang and Y.S. Jang, “On the implementation of melody recognition on 8-bit and 16-bit microcontrollers,” IEEE Multimedia Conference, Vol. 2 2003 pp.704-708.
[20]Jyh-Shing Roger Jang and Yung-Sen Jang“Microcontroller Implementation of Melody Recognition: a Prototype,” In Proceedings of the Eleventh ACM International Conference on Multimedia 2003.
[21]Jin Ha Lee and J, Stephen Downie“Survey of Music Information Needs, Uses and Seeking Behaviors: Preliminary Findings,” In International Symposium on Music Information Retrieval 2004.
[22]Kosugi, N., Y. Nishihara, S. Kon’ya, M. Yamamuro, and K. Kushima, “Music Retrieval by Humming,” In Proceedings of PACRIM’99 , pages 404-407, IEEE, August 1999.
[23]Kosugi, N., Y. Nishihara, S. Kon’ya, M. Yamamuro, and K. Kushima, “A Practical Query-By-Humming System for a Large Music Database,” In Proc. ACM Multimedia 2000.
[24]Liu, Chih-Chin and P. J. Tsai, “Content-Based Retrieval of MP3 Music Object,” to appear in Proc. Of the ACM Intl. Conf. On Information and Knowledge Management, 2001.
[25]Liu, Chih-Chin and Hung-Min Yu, “Effective Retrieval of MP3 Music Objects WITH Background Reduction Technique,” in Proc. of Third International Workshop on Content-Based Multimedia Indexing (CBMI 2003), 2003, IRISA, Rennes, France.
[26]Liu, Chih-Chin and Wei-Yi Kuo, “Content-Based Segmentation of MP3 Music Objects,” In Proc. of Workshop on Computer Music and Audio Technologies (WOCMAT), 2005, Taipei, Taiwan.
[27]Martin F. McKinney and Jeroen Breebaart “Features for Audio and Music Classification,” In International Symposium on Music Information Retrieval 2003.
[28]Maarten Grachten, Josep-Lluis Arcos and Ramon Lopez de Mantaras “Melodic Similarity: Looking for A Good Abstraction Level,” In International Symposium on Music Information Retrieval 2004.
[29]Micheline Lesaffre, et al. “The MAMI Query-by-Voice Experiment: Collecting and Annotating Vocal Queries for Music Information Retrieval,” In International Symposium on Music Information Retrieval 2003.
[30]Norman H. Adams, et al.“Time Series Alignment for Music Information Retrieval,” In International Symposium on Music Information Retrieval 2004.
[31]Pan, D., “A Tutorial on MPEG/Audio Compression,” IEEE Multimedia Magazine, Summer 1995, pp. 60-74
[32]Rainer Typke, Frans Wiering and Remco C. Veltkamp “A Search Method for Notated Polyphonic Music with Pitch and Tempo Fluctuations,” In International Symposium on Music Information Retrieval 2004.
[33]Richard L. Kline and Ephraim P. Glinert“Approximate Matching Algorithms for Music Information Retrieval Using Vocal Input,” In Proceedings of the eleventh ACM International Conference on Multimedia 2003.
[34]Roger B. Dannenberg and Ning Hu“Understanding Search Performance in Query-by-Humming Systems,” In International Symposium on Music Information Retrieval 2004.
[35]Rolland, P. Y., G Raskinis, and J. G. Ganascia, “Musical Content-based Retrieval: an Overview of the Melodiscov Approach and System,” In Proc. ACM Multimedia 99, pages 81-84, November 1999.
[36]Shai Shalev-Shwartz, Joseph Keshet and Yoram Singer “Learning to Align Polyphonic Music,” In International Symposium on Music Information Retrieval 2004.
[37]Johah Shifrin and William P.irmingham “Effectiveness of HMM-Based Retrieval on Large Databases” In International Symposium on Music Information Retrieval 2003.
[38]Shyamala Doraisamy and Stefan Riiger “A Polyphonic Music Retrieval System Using N-Grams,” In International Symposium on Music Information Retrieval 2004.
[39]Masataka Goto “A Chorus Section Detection Method for Musical Audio Signals and Its Application to a Music Listening Station,” IEEE Transactions onAudio, Speech and Language Processing 2006.
[40]Wei-Ho Tsai, et al. “Blind Clustering of Popular Music Recordings Based on Singer Voice Characteristics,” In International Symposium on Music Information Retrieval 2003.
[41]Yongwei Zhu and Mohan S Kankanhalli “A Robust Music retrieval Method for Query-by-Humming,” InIEEE International Conference on Information Technology 2003.