Music analysis data set construction method and pitch and value extraction method based on data set

Abstract

The invention discloses a music analysis data set construction method and a pitch and value extraction method based on the data set, and mainly solves the problems that in the prior art, the performance of a traditional method is good in the single tone recognition subdivision field, but the performance in the complex tone recognition field is relatively weak. The music analysis data set construction method and the pitch and value extraction method based on the data set comprise the following steps of selecting and registering digital music scores into a digital music score library; registering a music score played by a player into the digital music score library; carrying out matching on a class label sequence and the sound characteristics to form a sound characteristic-music element dataset construction method; and based on the data set, executing the pitch and value extraction method and system from musical sound and musical instruments. According to the scheme, the value information and the pitch information of the digital sound signal generated by the musical instruments are extracted by using a supervised machine learning model, the purpose of extracting music elements in the complex tone can be well extracted, and the method has very high practical value and popularization value.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a method, system and equipment for extracting pitch and duration information from musical instrument sounds. Background technique [0002] Pitch and time value information are two basic and important elements in music. The performance analysis or automatic transcription of any musical instrument needs to be based on pitch and time value; traditional methods involving digital signal processing for music analysis mainly Use artificial rules to make logical judgments and extract relevant music elements. [0003] For example pitch analysis typically uses fundamental frequency estimation. The traditional method performs well in the field of monophonic recognition subdivision, but its performance in the field of polyphonic recognition is relatively weak, and it cannot extract the pitch and duration of the sound well. Contents of the invention [0004] The purpose of the present i...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method for constructing a music analysis data set and a method for extracting pitch and time value based on it, so as to solve the problem that the traditional method has good performance in the field of monophonic recognition subdivision, but has poor performance in the field of polyphonic recognition. Issues with relatively weak performance

Images