A Music Cognition–Guided Framework for Multi-pitch Estimation

• DOI: 10.1007/s12559-022-10031-5
• Publication Date: 2022-06-14T01:02:27Z
• Authors: Xiaoquan Li; Yijun Yan; John Soraghan; Zheng Wang; Jinchang Ren
• Keywords: /dk/atira/pure/subjectarea/asjc/1700/1707; QA75; /dk/atira/pure/subjectarea/asjc/1700/1706; /dk/atira/pure/subjectarea/asjc/2800/2805; 02 engineering and technology; name=Computer Science Applications; name=Cognitive Neuroscience; 004; Multi-pitch estimation; Automatic music transcription; Music cognition; Electronic computers. Computer science; Harmonic structure detection (HSD); 0202 electrical engineering, electronic engineering, information engineering; Polyphonic music detection; name=Computer Vision and Pattern Recognition

AbstractAs one of the most important subtasks of automatic music transcription (AMT), multi-pitch estimation (MPE) has been studied extensively for predicting the fundamental frequencies in the frames of audio recordings during the past decade. However, how to use music perception and cognition for MPE has not yet been thoroughly investigated. Motivated by this, this demonstrates how to effectively detect the fundamental frequency and the harmonic structure of polyphonic music using a cognitive framework. Inspired by cognitive neuroscience, an integration of the constant Q transform and a state-of-the-art matrix factorization method called shift-invariant probabilistic latent component analysis (SI-PLCA) are proposed to resolve the polyphonic short-time magnitude log-spectra for multiple pitch estimation and source-specific feature extraction. The cognitions of rhythm, harmonic periodicity and instrument timbre are used to guide the analysis of characterizing contiguous notes and the relationship between fundamental frequency and harmonic frequencies for detecting the pitches from the outcomes of SI-PLCA. In the experiment, we compare the performance of proposed MPE system to a number of existing state-of-the-art approaches (seven weak learning methods and four deep learning methods) on three widely used datasets (i.e. MAPS, BACH10 and TRIOS) in terms of F-measure ($${F}_{1}$$F1) values. The experimental results show that the proposed MPE method provides the best overall performance against other existing methods.