Musical sound recognition method
A recognition method and noise technology, applied in the field of noise recognition, can solve problems such as large fluctuations, single recognition ability, and general accuracy, and achieve the effect of improving recognition accuracy, high recognition accuracy, and high stability
Active Publication Date: 2019-05-24
深圳蜜蜂云科技有限公司
View PDF7 Cites 7 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
NMF does not have a good way to obtain the contextual association of notes and apply this relationship to musical tone (sound) recognition
[0007] 3) The algorithm does not have the ability to deal with different instrumental music and identify different instrumental music data, and the recognition ability is single
[0008] To sum up the above reasons, the accuracy of the NMF algorithm in the recognition of musical tones (sounds) to pitch and sound length is average. In known polyphonic music recognition applications, the recognition accuracy is maintained between 50% and 70%, and fluctuates. Larger, there is instability, which greatly limits the application space of tone (sound) recognition in the field of music
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0040] Now in conjunction with the accompanying drawings, the preferred embodiments of the present invention will be described in detail.
[0041] Such as Figure 1 to Figure 4 As shown, the present invention provides a preferred embodiment of a tone recognition method.
[0042] Described tone recognition method comprises the steps:
[0043] S10, collecting a large number of music samples;
[0044] S20. Using the collected tone samples, train an automatic tone recognition model for identifying pitches and durations of notes;
[0045] S30. Input the musical tone data to be recognized, call the automatic musical tone recognition model, and generate a musical note result set.
[0046] By collecting a large number of tone samples and using the collected tone samples to train the tone automatic recognition model, the tone automatic recognition model can identify the pitch and duration of the note in the tone data to be recognized and generate a note result set to realize the rec...
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
Login to View More Abstract
The invention relates to the field of musical sound (instrumental musical sound) recognition, in particular to a musical sound recognition method, comprising the steps of collecting massive musical sound samples; with the collected musical sound samples, training a musical sound auto-recognition module which recognizes pitches of notes and their durations; inputting musical sound data to be recognized, calling the musical sound auto-recognition module to generate a set of note results. The massive musical sound samples are collected and utilized to train the musical sound auto-recognition model; the musical sound auto-recognition model can recognize pitches of notes in the musical sound data to be recognized and their durations to generate a set of note results, so that musical sound recognition is achieved, recognition accuracy is high, recognition stability is high, and the influence from sound muffling is avoided.
Description
technical field [0001] The invention relates to the field of musical tone (musical instrument sound) recognition, in particular to a musical tone recognition method. Background technique [0002] Tone (sound) recognition was first proposed in 1977. With the understanding of digital audio engineering by audio researchers, these researchers believe that computers can analyze digital music data through certain algorithms to detect melody sounds. High and chord patterns, as well as instrumental rhythms. [0003] In the field of traditional musical instrument recognition, the most widely used in the early stage is to use a nonnegative matrix factorization (Nonnegative matrix factorization, NMF) algorithm for recognition. Non-negative matrix factorization was proposed by Lee and Seung in Nature in 1999, which makes all decomposed components non-negative (requiring a purely additive description), and at the same time realizes nonlinear dimension reduction. The psychological and p...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G10L25/51G10L25/30
Inventor 钟毅陆建刘强李湘
Owner 深圳蜜蜂云科技有限公司