透過您的圖書館登入
IP:3.142.53.216
  • 學位論文

改良式動態規畫和馬可夫模型與音樂旋律比對

Modified Dynamic Programming Algorithm and Markov Model for Melody Matching

指導教授 : 丁建均
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


常常我們腦袋中會浮起一段旋律,或是在路上聽到一段美妙的音樂卻忘記或是想了解更多此歌曲的歌名和歌的資訊確無法得知而感到可惜,因此,音樂檢所的應用越來越顯得其重要的地位,適當的音樂檢所方法會使得在找尋想了解的歌曲更加有效率且準確。 相對於以往用歌曲名稱以及作者當檢所,近年來有個系統,哼唱詢問(QBH)系統,使用者可以哼唱一段旋律,而此系統可將此旋律分析音樂上的特性,例如拍子資訊以及頻率資訊,並且將這些資訊進一步與資料庫裡面的歌曲作相似度比對,取得最大相似度的歌曲並且得到一系列可能的歌曲。對使用者來說,在不知道歌曲歌名或歌手以及歌詞的狀況下,想要搜尋想要的歌曲可說是相當的方便以及迷人。 但是在哼唱詢問系統中面臨著幾個問題,就是面對各式各樣演唱風格的使用者以及龐大的資料庫歌曲。不同使用者其演唱風格有所不同,其哼唱出來的旋律其音調高低以及音準的準確性皆有所相異,且在將使用者者哼唱出來的旋律與龐大資料歌曲旋律相似度比對時,當資料庫歌曲數目越多,則系統的處理時間也會跟著上升。因此在本篇論文中,將會針對這兩個問題,系統搜尋準確率以及系統搜尋處理的時間做研究,並且提出方法來改善。 實驗結果顯示在擴充的資料庫中,用提出的方法來做模擬將使得系統搜尋的準確率以及處理時間皆有改善,使得哼唱詢問系統在應用上,變得更為親近且方便。

關鍵字

哼唱詢問 旋律比對

並列摘要


We often memory a melody on our mind, or we want to know the more information about the music that we heard from the street. Therefore the application of music retrieval appears important these days. It is efficient that we use good music retrieval method to find the desire songs. Relative to the past that we use the song name or author name of the song, there is a system called query by humming (QBH) system in recent year. It uses the information of the melody that people hum to find the similar song in the database and generate a series of possible song. It is convenient for people that we do not know the song name or the author of the song. But there are some problem, including many kinds of singing style and large number of song of database. We focus on these problems in the thesis and propose method to deal with. Experiment results show that the hit rate will increase and the running time will decrease. It causes the convenience of QBH system.

並列關鍵字

Query by humming melody matching

參考文獻


B. Onset Detection
[4] A. Klapuri, “Sound Onset Detection by Applying Psychoacoustic Knowledge,” in Proc. of IEEE International Conference on Acoustics, Speech and Signal, 1999.
[7] J. Foote, “Automatic audio segmentation using a measure of audio novelty,” in Proc. of IEEE International Conference on Multimedia and Expo, issue 1, pp. 452–455, 1999
[8] P. Masri, and A. Bateman, ” Improved modelling of attack transients in music analysis-resynthesis,” in Proc. of International Computer Music Conference (ICMC 96), Hong-Kong, Aug 1996,
[10] A. de Cheveigne and H. Kawahara, “Yin, a fundamental frequency estimator for speech and music,” in Proc. of Acoust. Soc. Am., vol, 111, Issue, 4 pp. 1917-1930, April 2002.

延伸閱讀