進階搜尋


   電子論文尚未授權公開,紙本請查館藏目錄
(※如查詢不到或館藏狀況顯示「閉架不公開」,表示該本論文不在書庫,無法取用。)
系統識別號 U0026-2708202005272400
論文名稱(中文) 基於樂曲譜面資訊之和弦進行與旋律匹配
論文名稱(英文) Chord Progression and Melody Matching Based on Music Score
校院名稱 成功大學
系所名稱(中) 資訊工程學系
系所名稱(英) Institute of Computer Science and Information Engineering
學年度 108
學期 2
出版年 109
研究生(中文) 林煒婷
研究生(英文) Wei-Ting Lin
學號 p76074355
學位類別 碩士
語文別 英文
論文頁數 14頁
口試委員 指導教授-蘇文鈺
口試委員-楊奕軒
口試委員-蔡宗翰
口試委員-蘇黎
口試委員-劉奕汶
中文關鍵字 簡要總譜  音樂生成  旋律相似性 
英文關鍵字 lead sheet  music generation  melody similarity 
學科別分類
中文摘要 不論是古典、爵士、或是流行音樂,大部分樂曲都是由旋律和伴奏所組成。伴奏和旋律兩者的關係可以說是相輔相成,旋律是樂曲最明顯的表現形式,而和弦則是樂曲的色彩基調。
有趣的是,有些和弦進行非常受歡迎。以流行樂為例,無論是中文或西洋歌曲,很多歌曲都使用了相同的和弦進行。例如:張惠妹《記得》、王菲《紅豆》、周杰倫《千里之外》、Maroon 5《Memories》使用了IVviiiiIVIIVV 和弦進行(”Pachelbel’s Canon” Progression);五月天《轉眼》、Beyond《真的愛你》、Jason Mraz《I’m Yours》、OneRepublic《Apologize》、Alan Walker《Faded》使用了I–V–vi–IV 和弦進行(4chord song progression),等等。因此我們好奇是否可以將一首歌曲的和弦進行配上其他歌 曲的旋律,或是修改旋律的部分音符就能匹配。
本論文提出一個可以搭配不同歌曲的旋律與和弦進行的系統,以MusicXML 格式的lead sheet 作為資料集,首先由使用者指定資料集中的一段和弦進行片段,系統會根據和弦外音與休止符長度將資料集中的候選旋律片段排名,並提供可以調整旋律音高或音長的選擇來產生更多變化,最後再重新排名並根據與原旋律的相似度輸出旋律結果。我們使用一個序列對序列的自編碼器(客觀)來計算旋律間的相似度,並做了聽力測試(主觀)來驗證這個方法。
英文摘要 Whether it is classical, jazz, or pop music, most of the music is composed of melody and accompaniment. The relationship between accompaniment and melody can be said to complement each other, melody is the most obvious form of music, while chord is related to the color and harmony of the music.
Interestingly, some chord progressions are very popular. Taking pop music as an example, whether it is Chinese or Western songs, many songs use same chord progressions. For example, AMei ”Remember”, Faye Wong ”Red Bean”, Jay Chou ”Faraway”, and Maroon 5 “Memories” use IVviiiiIVIIVV
chord progression (”Pachelbel’s Canon” Progression);
While Mayday ”Final Chapter”, Beyond ”Really Love You", Jason Mraz “I'm Yours", OneRepublic ”Apologize”, and Alan Walker ”Faded” use I–V–vi–IV chord progression (4chord song progression), and so on. We are curious about whether we can combine the chord progressions of a song with the melody of other songs, or modify some notes of the melody to match it.
This paper proposes a system that can match the melody and chord progression of different song, using the lead sheet in the MusicXML format as the data set. Users first specify a chord progression segment of a song in the data set. Then the system rank the candidate melody segments in the data set according to the nonchord notes and rest duration, and
provide the option to adjust the pitch or duration of notes in the melody segments to produce more variations. Finally, the system rerank and output the melody result according to the similarity to the original melody. We use a sequencetosequence autoencoder (objective) to calculate the similarity between melodies, and apply a listening test (observation) to verify this method.
論文目次 摘要 i
Abstract ii
誌謝 iii
Table of Contents iv
List of Tables v
List of Figures vi
Chapter 1. Introduction 1
1.1. Background and Motivation 1
1.2. This Work 1
Chapter 2. Related Work 2
2.1. Music Mashup 2
2.2. Lead Sheet Generation 2
Chapter 3. Method 3
3.1. Data 3
3.2. Rank Method 4
3.3. Melody Similarity 4
3.4. Note Pitches Adjustment 6
3.5. Note Durations Adjustment 6
Chapter 4. Experiments 9
4.1. Note Pitches Adjustment 9
4.2. Note Durations Adjustment 10
Chapter 5. Conclusion and Future Works 12
References 13
參考文獻 [1] Matthew E. P. Davies, Philippe Hamel, Kazuyoshi Yoshii, and Masataka Goto. Automashupper: automatic creation of multisong music mashups. IEEE ACM Trans. Audio Speech Lang. Process., 22(12):1726–1737, 2014.
[2] Garth Griffin, Youngmoo E. Kim, and Douglas Turnbull. Beat-sync-mash-coder: A web application for realtime creation of beatsynchronous music mashups. In Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010, 14-19 March 2010, Sheraton Dallas Hotel, Dallas, Texas, USA, pages 437–440. IEEE, 2010.
[3] Gaëtan Hadjeres and Frank Nielsen. Deep rank-based transposition-invariant distances on musical sequences. CoRR, abs/1709.00740, 2017.
[4] Daniel D. Johnson, Robert M. Keller, and Nicholas Weintraut. Learning to create jazz melodies using a product of experts. In Ashok K. Goel, Anna Jordanous, and Alison Pease, editors, Proceedings of the Eighth International Conference on Computational Creativity, Atlanta, Georgia, USA, June 19-23, 2017, pages 151–158. Association for Computational Creativity (ACC), 2017.
[5] Robert M. Keller and David R. Morrison. A grammatical approach to automatic improvisation, 2007.
[6] Chuan-Lung Lee, Yin-Tzu Lin, Zun-Ren Yao, Feng-Yi Lee, and Ja-Ling Wu. Automatic mashup creation by considering both vertical and horizontal mashabilities. In Meinard Müller and Frans Wiering, editors, Proceedings of the 16th International Society for Music Information Retrieval Conference, ISMIR 2015, Málaga, Spain, October 26-30, 2015, pages 399–405, 2015.
[7] Hyungui Lim, Seungyeon Rhyu, and Kyogu Lee. Chord generation from symbolic melody using BLSTM networks. In Sally Jo Cunningham, Zhiyao Duan, Xiao Hu, and Douglas Turnbull, editors, Proceedings of the 18th International Society for Music Information Retrieval Conference, ISMIR 2017, Suzhou, China, October 23-27, 2017, pages 621–627, 2017.
[8] Hao-Min Liu and Yi-Hsuan Yang. Lead sheet generation and arrangement by conditional generative adversarial network. In M. Arif Wani, Mehmed M. Kantardzic, Moamar Sayed Mouchaweh, João Gama, and Edwin Lughofer, editors, 17th IEEE International Conference on Machine Learning and Applications, ICMLA 2018, Orlando, FL, USA, December 17-20, 2018, pages 722–727. IEEE, 2018.
[9] Alexandre Papadopoulos, Pierre Roy, and François Pachet. Assisted lead sheet composition using flowcomposer. In Michel Rueher, editor, Principles and Practice of Constraint Programming - 22nd International Conference, CP 2016, Toulouse, France, September 5-9, 2016, Proceedings, volume 9892 of Lecture Notes in Computer Science, pages 769–785. Springer, 2016.
[10] Pierre Roy, Alexandre Papadopoulos, and François Pachet. Sampling variations of lead sheets. CoRR, abs/1703.00760, 2017.
[11] Hiroaki Tsushima, Eita Nakamura, Katsutoshi Itoyama, and Kazuyoshi Yoshii. Function- and rhythm-aware melody harmonization based on tree-structured parsing and split-merge sampling of chord sequences. In Sally Jo Cunningham, Zhiyao Duan, Xiao Hu, and Douglas Turnbull, editors, Proceedings of the 18th International Society
for Music Information Retrieval Conference, ISMIR 2017, Suzhou, China, October 23-27, 2017, pages 502–508, 2017.
[12] Li-Chia Yang, Szu-Yu Chou, and Yi-Hsuan Yang. Midinet: A convolutional generative adversarial network for symbolic-domain music generation. In Sally Jo Cunningham, Zhiyao Duan, Xiao Hu, and Douglas Turnbull, editors, Proceedings of the 18th International Society for Music Information Retrieval Conference, ISMIR 2017, Suzhou, China, October 23-27, 2017, pages 324–331, 2017.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2020-09-03起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw