Singing evaluation method based on deep learning

Abstract

The invention provides a singing evaluation method based on deep learning in the technical field of singing evaluation, and the method comprises the following steps: S10, obtaining a large amount of singing data, and cleaning the singing data; step S20, extracting audio features of the cleaned singing data, and constructing a feature data set; step S30, creating an audio alignment model, and aligning each audio feature in the feature data set based on the audio alignment model; step S40, creating a multi-dimensional evaluation model based on deep learning, and training the multi-dimensional evaluation model by using the aligned feature data set; and step S50, acquiring an audio to be evaluated, inputting the audio to be evaluated into the multi-dimensional evaluation model to obtain an intonation evaluation value, a rhythm evaluation value and a pronunciation evaluation value, and displaying the intonation evaluation value, the rhythm evaluation value and the pronunciation evaluation value. The singing evaluation method has the advantages that the singing evaluation accuracy and interpretability are greatly improved.

Description

technical field [0001] The invention relates to the technical field of singing evaluation, in particular to a singing evaluation method based on deep learning. Background technique [0002] As music education gradually attracts the attention of the society, the number of people participating in music learning and engaging in music education continues to grow. In any subject, it is very important to receive timely evaluation and feedback at the initial stage of learning, and music study is no exception. Music learning is mainly a process of accumulating proficiency. Recognizing one's own deficiencies early can allow students to avoid early mistakes in the learning process and cultivate them into deep-rooted and difficult-to-correct bad habits after day-to-day practice. However, in the process of music learning, unless you are a student who grew up in a musical family, it is impossible to have a teacher by your side at all times, and students cannot receive immediate and prof...

AI Technical Summary

Problems solved by technology

However, in the process of music learning, unless you are a student who grew up in a musical family, you cannot always have a teacher by your side, and students cannot receive immediate and professional feedback, which greatly limits students' learning efficiency

[0003] Although there are some singing evaluation systems on the market, they are all designed and implemented around entertainment scenarios. The target users in these scenarios do not really care whether the final evaluation results are completely accurate, nor do they care about which part is not well done. , the evaluation results of singing evaluation often lack accuracy and interpretability

Images