手法を比較

	音楽セグメンテーション	コード認識	メロディ抽出
分野	音楽情報検索	音楽情報検索	音楽情報検索
系統	Machine learning	Machine learning	Machine learning
提唱年	2001	2005	2008
提唱者	Masataka Goto	Christopher Harte	Anssi Klapuri
種類	Audio structural analysis	Harmonic audio analysis	Polyphonic audio analysis
原典	Goto, M., & Hasegawa, Y. (2001). Automatic transcription of popular music audio. In Proceedings of the Fourth International Conference on Music Information Retrieval.	Harte, C., Sandler, M. B., Abdallah, S. A., & Gómez, E. (2005). Symbolic representation of musical chords: Proposed extensions to the HarmO ontology. In Proceedings of the International Society for Music Information Retrieval Conference.	Salamon, J., & Gómez, E. (2014). Melody extraction from polyphonic music signals using pitch contour characteristics. IEEE Transactions on Audio, Speech, and Language Processing, 20(6), 1759-1770.
別名	structural segmentation, music structure analysis, section boundary detection	chord estimation, harmonic analysis, chord detection	pitch contour extraction, melodic line extraction, f0 tracking
関連	5	5	5
概要	Music segmentation is the task of dividing a musical recording into distinct structural sections (e.g., verse, chorus, bridge, pre-chorus, outro). Introduced by Goto (2001), it identifies major structural boundaries and labels sections according to musical form. Segmentation is essential for music understanding, audio editing, and composition analysis. It enables higher-level tasks like cover song identification and song structure-aware music generation.	Chord recognition is the task of automatically identifying the harmonic chords present in a musical recording and estimating when chord changes occur. Introduced formally by Harte et al. (2005), it is a cornerstone of music analysis and widely used in music education, cover song analysis, and musical structure understanding. Modern systems use deep learning to classify and sequence chords in real time.	Melody extraction is the task of automatically isolating the main melodic contour from polyphonic music recordings. It originated from music transcription research in the 2000s and addresses the core challenge of human pitch perception: identifying the perceptually dominant pitch when many instruments play simultaneously. Modern approaches use deep learning and are essential for music analysis, cover song detection, and music-to-lyrics alignment.