在本論文中，我們提出了一個名為『虛擬頻道（virtual channel）』的新型態線上廣播電台（online radio）之資訊服務系統。在這個系統中，它最主要的功能是讓我們透過統合與分析各個網路電台的播放內涵後，經由自動選取並轉換到合適的電台的機制，讓我們享受到一個持續撥放著音樂的環境，也由於我們不需要手動地尋找目前正在撥放著有音樂內容的電台（就如同擁有一個一直撥放音樂的電台），所以我們的系統才名為虛擬音樂頻道系統；此外在本系統中，我們亦分析並判斷出每一個音樂頻道所屬的音樂心情（music mood），也就是在已經判斷出目前正在撥放著音樂的電台中，再進一步地判別出此撥放的音樂所帶給人的感覺是被歸類為悲傷、滿足、急躁或是熱鬧的機制；在經過以上兩層的判斷程序後，我們可以根據使用者的習慣或是外在所需的音樂環境（例如在輕鬆時想要有滿足風格的音樂氣氛等）來建立起一個可動態調整的音樂電台選擇清單（playbill）。綜合以上的系統服務需求，我們首先要研究的是：如何有效率的擷取出每一個網路電台資料串流中的特徵值，並且透過有效率的分析與判斷機制來找出正在撥放著含有音樂內涵的網路電台；緊接著是如何透過特徵值的分析來判斷出那些正被撥放的音樂片段們所代表的音樂心情。由於在資料串流的環境下有著反應時間的限制，所以我們必須找出兼具有特徵值組合精簡以及判斷時間迅速的分類機制的來建構此系統。最後我們將展示一系列的實驗來驗證系統中分類機制的判斷正確率以及評估系統整體的表現。

In this thesis, an online radio service called virtual channel is proposed. This service provides a continuous musical environment where the playbill is made according to the atmosphere or the user’s habit. To achieve this goal, we first extract and analyze the audio features of multiple data streams from the online radio servers. Through this step, we classify all radio channels and determine whether the content contains musical intension. After that, we continue to analyze the remaining channels which are playing music and detect the musical mood of the music slices by music psychological theory and heuristic rules. The playbill according to the user’s preference can then be arranged. Due to the constraint of the processing time in the streaming environment, we have to process all schemes mentioned above in real time. Therefore, we have to shorten the judging time by extracting few features and using efficient classifiers to satisfy the limitation. Finally, we perform a series of experiments to evaluate the performance of the proposed framework, and the discrimination rate of the schemes. The results show that the mechanism of virtual channel is workable in the streaming environment.

[Rose00] B. Rose and L. Rossin. Internet study v: “Startling new insights about the internet and streaming,” Sep. 2000 http://www.arbitron.com/downloads/internet_study_v.pdf.
[Davi01] David B. Hauver, and James C. French, “Flycasting: Using Collaborative Filtering to Generate a Playlist for Online Radio,” WEDELMUSIC ‘2001 (International Conference on Web Delivering of Music)
[Luka03] Lukasz Golab and M. Tamer Ozsu, “Issue in Data Stream Management,” In ACM SIGMOD (Special Interest Group On Management Of Data) Record, Volume 32, Number 2, June 2003, pp.5—14.
[Zhu02] Yunyue Zhu, Dennis Shasha, “StatStream: Statistical Monitoring of Thousands of Data Streams in Real Time,” VLDB (Very Large DataBase) 2002: 358-369
[Clau01] Claudine Santos Badue, Ricardo A. Baeza-Yates, Berthier A. Ribeiro-Neto, Nivio Ziviani, “Distributed Query Processing Using Partitioned Inverted Files,” SPIRE (String Processing and Information REtrieval) 2001: 10-20
[Anna01] Anna C. Gilbert, Yannis Kotidis, S. Muthukrishnan, Martin Strauss, “Surfing Wavelets on Streams: One-Pass Summaries for Approximate Aggregate Queries,” VLDB 2001: 79-88
[Char03] Charu C. Aggarwal, Jiawei Han, Jianyong Wang, Philip S. Yu, “A Framework for Clustering Evolving Data Streams,” VLDB (Very Large DataBase) 2003: 81-92
[Savi99] Savitha Srinivasan, Dragutin Petkovic, Dulce B. Ponceleon, “Towards robust features for classifying audio in the CueVideo system,” ACM Multimedia (1) 1999: 393-400
[Erli96] Erling Wold, Thom Blum, Douglas Keislar, James Wheaton, “Content-Based Classification, Search, and Retrieval of Audio,” IEEE MultiMedia 3: 27-36 (1996)
[Lu01] Lie Lu, Hao Jiang, HongJiang Zhang, “A robust audio classification and segmentation method,” ACM Multimedia 2001: 203-211
[Lu02] Lie Lu, Hao Jiang, Senior Member, IEEE, HongJiang Zhang, “Content Analysis for Audio Classification and Segmentation,” IEEE Transactions on Speech and Audio Processing, VOL. 10, NO. 7, October 2002
[Alex00] Alex Weiss, “Music Selection for Internet Radio,” November 3, 2000
[Zhan01] Tong Zhang, and C.-C. Jay Kuo, “Content-Based Classification and Retrieval of Audio”
[Jian04] M.H. Jian, C.H. Lin, and A.L.P. Chen, “Perceptual Analysis for Music Segmentation,” SPIE Storage and Retrieval Methods and Applications for Multimedia 2003
[Sche97] Scheirer, E. and M. Slaney, “Construction and evaluation of a robust multi-feature speech/music discriminator,” in IEEE Transactions on Acoustics, Speech and Signal Processing, PP. 1331-1334, 1997
[Mart03] Martin F. McKinney, and Jeroen Breebaart, “Features for Audio and Music Classification,” ISMIR (International Symposium on Music Information Retrieval) 2003
[Liu03] Dan Liu, Lie Lu, Hong-Jiang Zhang, “Automatic Mood Detection from Acoustic Music Data,” ISMIR (International Symposium on Music Information Retrieval) 2003
[Saun96] J.Saunder, “Real-Time Discrimination of Broadcast Speech/Music,” Proc. ICASSP (International Conference on Acoustic, Speech and Signal Processing) ’96, VOL II PP.93-996, Atlanta, May, 1996.
[Sche97] E.Scheirer, M. Slaney, “Construction and Evaluation of a Robust Multifeature Speech/Music Discriminator,” Proc. ICASSP (International Conference on Acoustic, Speech and Signal Processing) ’97, Munich, Germany, April, 1997
[Wyse95] L.Wyse, S. Smoliar, “Toward Content-based Audio Indexing and Retrieval and a New Speaker Discrimination Technique,” download from http://www.zwhome.org/~lonce, Institute of Systems Science, National Univ. of Singapore, Dec., 1995