Beat analysis method, audio processing method and device, equipment, medium
An analysis method and audio processing technology, applied in the field of equipment, media, audio processing method and device, and beat analysis method, can solve the problems of poor beat analysis effect, no obvious rhythm, unpredictable a cappella content, etc., and achieve strong anti-interference , high accuracy effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0047] figure 1 It is a flow chart of a beat analysis method provided by Embodiment 1 of the present invention. This embodiment is applicable to the case of performing beat analysis on a cappella audio. Specifically, the tempo analysis method can be executed by a tempo analysis device, which can be implemented in the form of software and / or hardware, and integrated into the device. Further, the device includes, but is not limited to: electronic devices such as computers, notebook computers, and MIDI (Musical Instrument Digital Interface, Musical Instrument Digital Interface) devices.
[0048] It should be noted that the beat analysis refers to the analysis of the combination rules of strong beats and weak beats in a piece of music. For a piece of a cappella data, various strong and weak beats will appear repeatedly in a certain order within the same length of time, forming regular changes in strength and weakness, from one strong beat to the next strong beat. That is called ...
Embodiment 2
[0061] figure 2 It is a flowchart of a tempo analysis method provided by Embodiment 2 of the present invention. On the basis of the above-mentioned embodiments, this embodiment specifically describes the steps of identifying the voice features of a cappella data, by pre-training the voice recognition model, using the voice recognition model to identify the phoneme features of the a cappella data, and improving the performance of voice feature recognition Accuracy.
[0062] refer to figure 2 , the beat analysis method includes:
[0063] S101. Obtain a cappella data sample, a dialogue data sample, a cappella voice feature, and a dialogue voice feature.
[0064] Specifically, this embodiment trains the speech recognition model for the difference between a cappella data and normal dialogue data. First, a large number of a cappella data samples and dialogue data samples are obtained, and a cappella voice feature and a dialogue voice feature are extracted from the samples. The...
Embodiment 3
[0081] Figure 4 It is a schematic diagram of the principle of a tempo analysis method provided by Embodiment 3 of the present invention. In this embodiment, specific optimization is performed on the basis of the above-mentioned embodiments, and the determination of the takt period value and the takt analysis result is described in detail. It should be noted that for technical details not exhaustively described in this embodiment, reference may be made to any of the foregoing embodiments.
[0082] Specifically, refer to Figure 4 , the method specifically includes the following steps:
[0083] S210. Obtain a cappella data of the user.
[0084] S220. Input the cappella data into a speech recognition model to obtain corresponding speech features.
[0085] Specifically, input the a cappella data into the speech recognition model, and the speech recognition model can recognize the occurrence time of each phoneme in the a cappella data according to the learned recognition rules, ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


