A method and device for extracting music clips

Through the sound acquisition and audio fingerprint comparison technology of natural environment, natural music clips are extracted and encoded, and the problem that the existing technology cannot provide personalized ringtones is solved, and ringtone clips that conform to natural characteristics and popular music are achieved, improving user experience .

CN114512137BActive Publication Date: 2024-12-03HUNAN UNIV OF SCI & TECH
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202210148938.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-02-18
Publication Date
2024-12-03
Estimated Expiration
2042-02-18

AI Technical Summary

Technical Problem

The prior art cannot effectively extract ringtone clips that conform to natural characteristics and are personalized, resulting in a decrease in user personalized experience.

Method used

By collecting and generating audio fingerprint data in natural environment sounds, comparing them with popular music clip databases, intercepting similar clips and encoding them, forming identification clips, storing them in the cloud and embedded in the ringtone library for users to choose.

Benefits of technology

It realizes providing unique ringtone clips that conform to the original natural characteristics, meeting users' personalized needs and improving user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114512137B_ABST
    Figure CN114512137B_ABST
Patent Text Reader

Abstract

The present invention discloses a music clip extraction method and a device thereof, belonging to the technical field of audio processing. The music clip extraction method comprises the following steps: S1, building an existing popular music clip into a popular audio fingerprint database; S2, setting a music clip extraction device in the corresponding natural environment sound, extracting the natural environment sound clip by the music clip extraction device, and generating a natural environment audio fingerprint data cloud; S3, comparing the natural environment audio fingerprint data cloud with the music clips in the popular audio fingerprint database to form a natural music clip; the scheme collects and extracts natural sound clips, and through the audio fingerprint data comparison, screens and intercepts audio clips similar to the popular music clips, thereby providing users with natural audio clips with melodies similar to popular music for users to choose as ringtones, thereby satisfying people's pursuit of personalization.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of audio processing, in particular to a method and device for extracting music segments. Background Art

[0002] At present, people's demand for smart terminals is increasingly personalized, for example, setting their favorite wallpapers, themes or ringtones, etc. When setting ringtones, many personal devices will set different ringtones. Users generally need to capture a certain music clip of their favorite music as a ringtone, or use song clips to combine them. However, some users prefer sounds close to nature, so different electronically synthesized natural sound clips have appeared, such as the gurgling sound of springs. Although such sounds are pleasant to the ears, they lack realism, and the rhythm is fixed and cannot be selected, which greatly reduces the user's freedom of use, thereby reducing the user's personalized experience. For this reason, we propose a music clip extraction method and device. Summary of the invention

[0003] The purpose of the present invention is to provide a music clip extraction method and device, which can extract sound clips by collecting different environmental sounds in nature, and match them with existing music clips, so as to extract unique ringtone clips that conform to the original natural characteristics and meet people's pursuit of personalization.

[0004] A second object of the present invention is a music piece extraction device used in the above method.

[0005] To solve the above problems, the present invention adopts the following technical solutions:

[0006] A music clip extraction method comprises the following steps:

[0007] S1. Build a popular audio fingerprint database from existing popular music clips;

[0008] S2, setting a music clip extraction device in the corresponding natural environment sound, extracting the natural environment sound clip through the music clip extraction device, and generating a natural environment audio fingerprint data cloud;

[0009] S3, comparing the natural environment audio fingerprint data cloud with the music clips in the popular audio fingerprint database, extracting the similar clips in the natural environment audio fingerprint data cloud and the music clips in the popular audio fingerprint database to form natural music clips;

[0010] S4, encoding the natural music clips according to the matched popular music clips to form recognition clips;

[0011] S5, storing the identified segments in a cloud server;

[0012] S6, the cloud server associates the identified segment with the matching popular music segment and sorts them according to the similarity to generate corresponding association links;

[0013] S7. Embed the associated link into the front-end ringtone library.

[0014] As a preferred solution of the present invention, the popular music clips in S1 are generated by sorting the number of user selections in the existing ringtone library, and are sorted according to the popularity of the music clips. The number of music clips intercepted in S3 is gradually reduced according to the popularity.

[0015] As a preferred solution of the present invention, the natural environment sound in S2 can be specifically set according to the music style of the target popular music clip.

[0016] As a preferred solution of the present invention, the natural environment sound in S2 can be collected by mixing multiple environmental sounds.

[0017] As a preferred solution of the present invention, if the natural environment audio fingerprint data in S3 has a chord similarity of more than 80% with the music clips in the popular audio fingerprint database, the natural music clips can be intercepted.

[0018] As a preferred solution of the present invention, the code for identifying the segment in S4 includes a popular music number, a natural sound type and a duration of the audio segment.

[0019] As a preferred solution of the present invention, the similarity between the S6 identified segment and the matching popular music segment is determined based on the sum of the quantized values ​​of the difference between the fingerprint data of the identified segment and the fingerprint data of the popular segment. The smaller the difference, the higher the similarity.

[0020] As a preferred solution of the present invention, the natural environment sound segment in S2 includes the following steps:

[0021] A1. Extract the spectrum peak points of the natural environment sound clips;

[0022] A2. selecting at least one pair of adjacent spectrum peak points from the spectrum peak points of the natural environment sound segment;

[0023] A3. Determine the fingerprint feature value corresponding to the target adjacent spectrum peak point according to the frequency parameter and time parameter of the target adjacent spectrum peak point pair;

[0024] A4. Determine the audio fingerprint data of the natural environment sound segment according to the fingerprint feature values ​​corresponding to all adjacent spectrum peak points.

[0025] As a preferred solution of the present invention, the popular music clips in S1 include three clips with the most ringtones taken from multiple popular songs in the existing ringtone library.

[0026] A music clip extraction device comprises an audio acquisition module, an audio storage module and an audio processing module. The audio acquisition module uses a noise reduction microphone to collect pure natural environment sound; the audio storage module is used to store the collected natural environment sound; the audio processing module is connected to a cloud server through a wireless network, compares the natural environment sound stored in the audio storage module with a popular audio fingerprint database on the cloud server, and uploads the natural environment sound clip.

[0027] Compared with the prior art, the advantages of the present invention are:

[0028] This solution collects and extracts natural sound clips, and through audio fingerprint data comparison, screens and extracts audio clips similar to popular music clips, thereby providing users with natural audio clips with melodies similar to popular music for users to choose as ringtones, satisfying people's pursuit of personalization. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is the work flow chart of the present invention.

[0030] Figure 2 It is a structural block diagram of the extraction device of the present invention. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] In the description of the present invention, it should be noted that the terms "upper", "lower", "inner", "outer", "top / bottom" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific position, be constructed and operated in a specific position, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0033] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "provided with", "mounted / connected", "connected", etc. should be understood in a broad sense. For example, "connected" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0034] Example:

[0035] See also Figure 1 , a music clip extraction method, comprising the following steps:

[0036] S1. Build a popular audio fingerprint database from existing popular music clips;

[0037] S2, setting a music clip extraction device in the corresponding natural environment sound, extracting the natural environment sound clip through the music clip extraction device, and generating a natural environment audio fingerprint data cloud;

[0038] S3, comparing the natural environment audio fingerprint data cloud with the music clips in the popular audio fingerprint database, extracting the similar clips in the natural environment audio fingerprint data cloud and the music clips in the popular audio fingerprint database to form natural music clips;

[0039] S4, encoding the natural music clips according to the matched popular music clips to form recognition clips;

[0040] S5, storing the identified segments in a cloud server;

[0041] S6, the cloud server associates the identified segment with the matching popular music segment and sorts them according to the similarity to generate corresponding association links;

[0042] S7. Embed the associated link into the front-end ringtone library.

[0043] Specifically, the popular music clips in S1 are generated by sorting the number of user selections in the existing ringtone library, and are sorted according to the popularity of the music clips. The number of music clips intercepted in S3 is gradually reduced according to the popularity.

[0044] A popular music clip audio fingerprint data is extracted from the audio data of a certain popular music, so a popular music clip audio fingerprint data must have a corresponding popular music, and a popular music can have multiple corresponding popular music clip audio fingerprint data. In the preset popular audio fingerprint database, each popular music clip audio fingerprint data can have an identifier, which can adapt to the preferences of users who like different clips of popular music.

[0045] Specifically, the natural environment sound in S2 can be specifically set according to the music style of the target popular music clip.

[0046] In this embodiment, through targeted screening, natural environment sounds are purposefully collected, the amount of screening is reduced, and the screening speed is greatly improved.

[0047] Specifically, the natural environment sound in S2 can be collected by mixing multiple environmental sounds.

[0048] In this embodiment, a variety of ambient sounds are mixed to provide a rich composition of ambient sounds, thereby improving the effect of the finished music clip after screening.

[0049] Specifically, if the natural environment audio fingerprint data in S3 has a chord similarity of more than 80% with the music clips in the popular audio fingerprint database, the natural music clip can be intercepted.

[0050] In this embodiment, the similarity level needs to be set to maintain a high degree of overlap of the audio ripples.

[0051] Specifically, the code for identifying the segment in S4 includes a popular music number, a natural sound type, and a duration of the audio segment.

[0052] Specifically, the similarity between the identified segment in S6 and the matched popular music segment is determined according to the sum of the quantitative values ​​of the difference between the fingerprint data of the identified segment and the fingerprint data of the popular segment. The smaller the difference, the higher the similarity.

[0053] Specifically, the natural environment sound clip in S2 includes the following steps:

[0054] A1. Extract the spectrum peak points of the natural environment sound clips;

[0055] A2. selecting at least one pair of adjacent spectrum peak points from the spectrum peak points of the natural environment sound segment;

[0056] A3. Determine the fingerprint feature value corresponding to the target adjacent spectrum peak point according to the frequency parameter and time parameter of the target adjacent spectrum peak point pair;

[0057] A4. Determine the audio fingerprint data of the natural environment sound segment according to the fingerprint feature values ​​corresponding to all adjacent spectrum peak points.

[0058] Specifically, the popular music clips in S1 include three clips with the most ringtones taken from multiple popular songs in the existing ringtone library.

[0059] See also Figure 2 A music clip extraction device includes an audio acquisition module, an audio storage module and an audio processing module. The audio acquisition module uses a noise reduction microphone to collect pure natural environment sound; the audio storage module is used to store the collected natural environment sound; the audio processing module is connected to a cloud server through a wireless network, compares the natural environment sound stored in the audio storage module with a popular audio fingerprint database on the cloud server, and uploads the natural environment sound clip.

[0060] Working principle: This solution collects and extracts natural sound clips, and through audio fingerprint data comparison, screens and extracts audio clips similar to popular music clips, thereby providing users with natural audio clips with melodies similar to popular music for users to choose as ringtones, satisfying people's pursuit of personalization.

[0061] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and improved concepts of the present invention within the technical scope disclosed by the present invention, and they should be covered by the protection scope of the present invention.

Claims

1. A method for extracting music clips, characterized in that: It includes the following steps: S1. Build a popular audio fingerprint database from existing popular music clips; S2, setting a music clip extraction device in the corresponding natural environment sound, extracting the natural environment sound clip through the music clip extraction device, and generating a natural environment audio fingerprint data cloud; S3, comparing the natural environment audio fingerprint data cloud with the music clips in the popular audio fingerprint database, extracting the similar clips in the natural environment audio fingerprint data cloud and the music clips in the popular audio fingerprint database to form natural music clips; S4, encoding the natural music clips according to the matched popular music clips to form recognition clips; S5, storing the identified segments in a cloud server; S6, the cloud server associates the identified segment with the matching popular music segment and sorts them according to the similarity to generate corresponding association links; S7. Embed the associated link into the front-end ringtone library.

2. A method for extracting music clips according to claim 1, characterized in that: The popular music clips in S1 are generated by sorting the number of user selections in the existing ringtone library, and are sorted according to the popularity of the music clips. The number of music clips intercepted in S3 is gradually reduced according to the popularity.

3. A method for extracting music clips according to claim 1, characterized in that: The natural environment sound in S2 can be specifically set according to the music style of the target popular music clip.

4. A method for extracting music clips according to claim 1, characterized in that: The natural environment sound in S2 can be collected by mixing multiple environment sounds.

5. A method for extracting music clips according to claim 1, characterized in that: If the natural environment audio fingerprint data in S3 has a chord similarity of more than 80% with the music clips in the popular audio fingerprint database, the natural music clips can be intercepted.

6. A method for extracting music clips according to claim 1, characterized in that: The code for identifying the segment in S4 includes the popular music number, the natural sound type and the duration of the audio segment.

7. A method for extracting music clips according to claim 1, characterized in that: The similarity between the identified segment and the matched popular music segment in S6 is determined according to the sum of the quantitative values ​​of the difference between the fingerprint data of the identified segment and the fingerprint data of the popular segment. The smaller the difference, the higher the similarity.

8. A method for extracting music clips according to claim 1, characterized in that: Generating the natural environment audio fingerprint data cloud in S2 includes the following steps: A1. Extract the spectrum peak points of the natural environment sound clips; A2. selecting at least one pair of adjacent spectrum peak points from the spectrum peak points of the natural environment sound segment; A3. Determine the fingerprint feature value corresponding to the target adjacent spectrum peak point according to the frequency parameter and time parameter of the target adjacent spectrum peak point pair; A4. Determine the audio fingerprint data of the natural environment sound segment according to the fingerprint feature values ​​corresponding to all adjacent spectrum peak points.

9. A method for extracting music clips according to claim 1, characterized in that: The popular music clips in S1 include three clips with the most ringtones taken from multiple popular songs in the existing ringtone library.

Citation Information

Patent Citations

  • Music clip extraction method and device

    CN106708990A

  • Music listening and recognizing method and system based on audio fingerprint and audio equipment

    CN111161758A

  • System and method for generating ring tone / ring backtone based on user preference melody part by real-timemusic identification

    KR100774708B1