Research and Implementation of Automatic Score Recording Algorithm for Piano Music Based on Feature Extraction

Authors

  • Xue Biyun

Abstract

Music is closely related to human life, and it is an important way of expressing people's feelings and singing life. With the rapid progress of artificial intelligence in recent years and its application in various fields, it has also brought great development to computer music. In this paper, an automatic notation algorithm for piano music based on feature extraction is studied and implemented. Firstly, the input piano music sound signal is analyzed in time frequency, and the original time domain waveform of the piano music sound signal is transformed into the representation of frequency distribution changing with time; Then detect that starting point of note; Finally, multi-pitch detection is performed to detect the pitch of newly played notes at the starting point of each note. The performance evaluation experiment is based on UCHO set and RAND set of MAPS data set. Compared with the best piano music automatic notation system in MIREX, the performance of piano multi-note estimation algorithm proposed in this paper has been greatly improved.

Published

2021-07-01

Issue

Section

Articles