Song recommendation method based on singer voice characteristics

A recommendation method and technique for singers, applied in speech analysis, special data processing applications, instruments, etc., can solve the problems of inability to sing high-pitched parts, high-pitched songs, and not suitable for all users to sing, and achieve a wide range of applications.

Inactive Publication Date: 2017-07-28
FUZHOU UNIV
View PDF1 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, hit songs are not suitable for all users
It is possible that the pitch of the song is too high, and due to the limitation of the user's own singing range and singing a...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Song recommendation method based on singer voice characteristics
  • Song recommendation method based on singer voice characteristics
  • Song recommendation method based on singer voice characteristics

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] The present invention will be further described below in conjunction with accompanying drawings and embodiments.

[0040] This embodiment provides a song recommendation method based on the singer's voice characteristics, such as figure 1 Shown include the following steps:

[0041] Step S1: analyze the numbered notation information of the songs in the music library, obtain the MIDI pitch reference sequence of each song, analyze the histogram of the sound level distribution of the song, and obtain the singing range requirements of each song;

[0042]Step S2: Use the MELODIA algorithm to analyze the user's a cappella recording file, obtain the MIDI pitch value sequence of the song sung by the singer, obtain the MIDI pitch reference sequence of the same song obtained in step S1, calculate the singer's benchmark singing ability, and extract its singing range;

[0043] Step S3: Extract the time-frequency signal representation from the singer's a cappella file, input it into...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention relates to a song recommendation method based on singer voice characteristics. According to the method, a song characteristic file library is built according to song numbered musical notations, music-free singing tapes of singers and other information; singing ranges of songs are extracted, a voice timbre embedding space is built, and voice timbre characteristics of original singers are obtained; singing ranges and voice timbre representations are extracted from music-free record files of the singers, and the voice characteristics of the singers are depicted; the sound level distribution conditions of the songs and singing ability evaluation values of the singers at all sound levels are calculated, and the matching degree of the user singing range and song singing range requirements is calculated; voice fragments of the singers are embedded in the voice timbre embedding space, and the voice timbre similarity of the voice fragments and each singer in the embedding space is calculated. By means of the method, the singing range matching degree of the singers and the voice timbre similarity can be considered comprehensively, and the recommendation degree of each song to a user is calculated.

Description

technical field [0001] The invention relates to an audio signal processing method in the field of singing, in particular to a method for recommending songs based on vocal characteristics of a singer. Background technique [0002] The music recommendation system focuses on recommending songs that users may like to listen to. The recommendation techniques used can be mainly divided into content-based recommendation and collaborative filtering-based recommendation. The content-based recommendation algorithm mainly makes recommendations based on the audio characteristics of the music itself, including low-level features such as MFCC or features such as melody, rhythm, genre, and emotion. The recommendation algorithm based on collaborative filtering is mainly based on the on-demand behavior or playback records between users, and recommends based on the similarity between users. [0003] In recent years, under the dual stimulation of the rapid development of mobile Internet appli...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06F17/30G10L25/48
CPCG06F16/637G06F16/683G10L25/48
Inventor 余春艳苏金池刘煌郭文忠
Owner FUZHOU UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap