Vocal music teaching singing skill detection system

Abstract

The invention belongs to the technical field of singing skill detection, and discloses a vocal music teaching singing skill detecting system. The vocal music teaching singing skill detection system includes a sound acquisition module, a throat detecting module, a feature signal collecting module, a respiratory frequency acquisition module, a sound processing module, a database, a comparison evaluation module, and an expert evaluation module. The invention completes the acquisition and evaluation of the data in the whole vocal music skill detection process on the basis of the computer system, and has high automation degree and wide coverage; the system can perform the anti-noise pitch feature extraction, the note pitch model set and the note time value model set parameter training through the sound processing module, has high recognition rate and calculation speed, and can improve the accuracy of acquired sound audios; the invention can meet the use requirements of different vocal musictrainings, compares the acquired audio data with a standard audio so as to allow a student to intuitively understand his training defects.

Description

technical field [0001] The invention belongs to the technical field of singing skill detection, and in particular relates to a singing skill detection system for vocal music teaching. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: [0003] Vocal music refers to the form of music sung by human voice. Vocal music is mainly based on the human vocal cords, with the mouth, tongue, and nasal cavity acting on the breath to produce pleasant, continuous, and rhythmic sounds. According to the height of the range and the difference in timbre, it can be divided into soprano, mezzo-soprano, alto and tenor, baritone, and bass. The range of each human voice is about two octaves. Vocal music includes: bel canto, folk singing and popular singing. In 2006, original ecological singing appeared in China. Usually vocal refers to bel canto. However, most of the existing vocal music teaching and singing are manually assessed b...

AI Technical Summary

Problems solved by technology

[0007] Most of the existing vocal music teaching and singing are manually evaluated by teachers, and there will be deviations to a certain extent, and many singing skills cannot be tested, such as the position of the larynx and the sound penetration; the sound audio collected at the same time is easily affected by noise. lead to inaccurate test data

[0008] In recent years, some scholars have done some research on the carrier frequency estimation of the single carrier frequency signal under the Alpha stable distributed noise model, but the research results are few

In the prior art, when the mixed signal-to-noise ratio is -10dB and α is 1.5, the normalized mean square error of the carrier frequency estimation of the BPSK signal is 0.043, and the normalized mean square error of the carrier frequency estimation of the QPSK signal is 0.041, so the low The performance of carrier frequency estimation under signal-to-noise ratio still needs to be improved

Images