Emergence of Musical Intelligence Based on Probabilistic Integration of Signal and Symbolic Processing

2017 Fiscal Year Final Research Report

• Project/Area Number: 26700020
• Research Category: Grant-in-Aid for Young Scientists (A)
• Allocation Type: Partial Multi-year Fund
• Research Field: Intelligent informatics
• Research Institution: Kyoto University
• Principal Investigator: YOSHII Kazuyoshi 京都大学, 情報学研究科, 講師 (20510001)
• Project Period (FY): 2014-04-01 – 2018-03-31
• Keywords: 音楽情報処理 / 機械学習 / 信号処理 / 記号処理 / ノンパラメトリックベイズ

Outline of Final Research Achievements

We have established a statistical framework that can jointly infer musical notes and syntactic structures from musical audio signals using a hierarchical Bayesian model based on integration of language and acoustic models. We tackled automatic music transcription for piano signals, musical note estimation for singing F0 trajectories, rhythm transcription for MIDI performances, automatic drum and guitar transcription for popular music audio signals. We applied these methods to assisting a user to sing a song and play a musical instrument. In addition, we developed an interactive music composition system based on a hierarchical Bayesian model of chords and melodies.

Free Research Field

音楽情報処理