Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Intonation evaluation method based on Deep Conventional Neural Network DCNN and CTC algorithm

A neural network, deep convolution technology, applied in the field of pitch assessment, which can solve problems such as errors

Active Publication Date: 2019-10-22
XIAN CONSERVATORY OF MUSIC +1
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Then, when the melody is matched, the forward algorithm of HMM is used to calculate the matching similarity as the matching probability. The limitation of this method is that it will cause errors when the query melody length is greater than the longest path in HMM.

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
  • Intonation evaluation method based on Deep Conventional Neural Network DCNN and CTC algorithm
  • Intonation evaluation method based on Deep Conventional Neural Network DCNN and CTC algorithm
  • Intonation evaluation method based on Deep Conventional Neural Network DCNN and CTC algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] Such as figure 1 and figure 2 Shown, the pitch evaluation method based on deep convolutional neural network DCNN and CTC algorithm of the present invention comprises the following steps: 1. The pitch evaluation method based on deep convolutional neural network DCNN and CTC algorithm is characterized in that, the method comprises the following step:

[0053] Step 1. Train the deep learning network model. The process is as follows:

[0054] Step 101, using audio recording equipment to collect multiple sets of vocal audio data of professional singers within a specified time, and transmitting and storing them to a computer, naming each set of vocal audio data in the computer according to the name of the person and the number of the music segment, and A MIDI file is provided for each group of human voice audio data, and multiple groups of human voice audio data constitute a standard audio data set;

[0055]Step 102, the computer performs feature extraction on each group ...

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

PropertyMeasurementUnit
Frame lengthaaaaaaaaaa
Login to View More

Abstract

The invention discloses an intonation evaluation method based on a Deep Conventional Neural Network DCNN and a CTC algorithm. The method comprises the following steps: firstly, a deep learning networkmodel is trained; secondly, recognition of notes in audio data is tested; thirdly, recognition of pitch in the audio data is tested; and fourthly, recognition of rhythm in the audio data is tested. According to the method in the invention, the notes in a test melody are recognized and segmented through the deep learning method, then, the feature information of the pitch and duration of each noteis extracted for each segmented note, and the feature information is compared and analyzed with a standard audio band so as to give an evaluation result and achieve the purpose of music teaching; andmeanwhile, a visual mark of the image is given, so that a relatively reliable and comprehensive intonation evaluation solution can be provided for effective development of music basic teaching, and the audio quality is evaluated.

Description

technical field [0001] The invention belongs to the technical field of pitch evaluation, and in particular relates to a pitch evaluation method in music teaching based on a deep convolutional neural network (DCNN) and a CTC algorithm. Background technique [0002] Pitch, intensity, duration, and timbre are the four properties of sound. Among them, the pitch and length are decisive for the quality of intonation. In the field of music education, solfeggio and ear training is the most important music basic course. Through solfeggio and listening, this course cultivates learners' correct intonation, rhythm and score-sight-singing ability, so that learners have professional music skills. literacy, and promote the improvement of their musical aesthetic ability. However, the current solfeggio teaching is generally based on the "factory model" of the 19th century. All students learn at the same time and place at the same speed and in the same way, using the "assembly line" model t...

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): G10L25/60G10L25/30G10L25/03G10L25/24G06N3/04G06N3/08
CPCG10L25/60G10L25/30G10L25/03G10L25/24G06N3/08G06N3/045
Inventor 冯勇王薇许鹏飞康金龙
Owner XIAN CONSERVATORY OF MUSIC
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com