Music clip selection method, device, equipment and medium

Abstract

The present invention discloses a method, device, equipment and medium for selecting music clips, the method comprising: obtaining comment information associated with each lyric block in a song, determining the popularity of each lyric block according to the comment information; determining a target lyric block according to the lyric block popularity, determining the correlation between the target lyric block and the comment information associated with it; selecting a hot lyric block in the target lyric block according to the correlation; and selecting a music clip corresponding to the hot lyric. The present invention improves the accuracy of the selected music clips.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, device, equipment and medium for selecting music clips. Background Art

[0002] As the music industry enters the era of mobile Internet, music platforms are paying more and more attention to music research and are committed to making personalized recommendations to users. There is a certain lyrics or melody fragment in a song that is the core fragment of the song, and the core fragment is often more likely to be liked by users. In existing solutions, popular music fragments are usually determined based on the popularity of song playback and the popularity of lyrics comments, and the popular music fragments are edited out. However, the popular music fragments selected by the existing solution may be due to high local popularity or attractive melody, which may deviate from the theme of the song to a certain extent, resulting in inaccurate selection of popular music fragments. Summary of the invention

[0003] The main purpose of the present invention is to provide a music clip selection method, device, equipment and medium, aiming to solve the problem of how to improve the accuracy of music clip selection.

[0004] To achieve the above object, the present invention provides a method for selecting a music clip, the method comprising the following steps:

[0005] Obtaining comment information associated with each lyric block in the song, and determining the lyric block heat of each lyric block according to the comment information;

[0006] Determine a target lyric block according to the popularity of the lyric block, and determine the relevance of the target lyric block and its associated comment information;

[0007] Selecting a popular lyrics block from the target lyrics block according to the correlation;

[0008] Select the music clip corresponding to the popular lyrics.

[0009] In one embodiment, the step of selecting a popular lyrics block from the target lyrics block according to the correlation comprises:

[0010] Obtaining the duration of the music segment corresponding to the target lyrics block;

[0011] A popular lyrics block is selected from each of the target lyrics blocks according to the relevance and the duration.

[0012] In one embodiment, the step of selecting a hot lyrics block from each of the target lyrics blocks according to the relevance and the duration includes:

[0013] Determine a difference between the duration and a preset duration threshold, and determine a deviation percentage according to a ratio of the difference to the preset duration threshold;

[0014] Sorting the association degrees from large to small, determining the ranking number of each association degree, and determining the deviation coefficient according to a preset value and the ranking number;

[0015] The hot lyrics block is selected from each of the target lyrics blocks according to the correlation degree, the deviation percentage and the deviation coefficient.

[0016] In one embodiment, the step of obtaining comment information associated with each lyric block in the song and determining the popularity of each lyric block according to the comment information includes:

[0017] Obtain comment information associated with each lyric block, and determine the keywords corresponding to each comment information;

[0018] Determine the frequency and probability of occurrence of the keyword;

[0019] Determining the comment popularity of the comment information according to the occurrence frequency and the occurrence probability;

[0020] The lyrics block heat of each lyrics block is determined according to the average heat of the comment heats of each of the comment information.

[0021] In one embodiment, the step of determining the occurrence frequency and occurrence probability of the keyword includes:

[0022] Determine each equivalent keyword corresponding to the keyword, and determine the total number of times each equivalent keyword appears in the comment information of the song;

[0023] Determine the occurrence frequency of the keyword according to the total number of times;

[0024] The occurrence probability of the keyword is determined according to the occurrence frequency and the total number of the comment information.

[0025] In one embodiment, the step of obtaining comment information associated with each lyric block and determining a keyword corresponding to each comment information includes:

[0026] Get the comment information associated with each lyric block;

[0027] If the comment information includes expression comments and / or picture comments, determining label information corresponding to the expression comments and / or picture comments, determining equivalent words corresponding to the expression comments and / or picture comments according to the label information; determining the keywords corresponding to the expression comments and / or picture comments according to the equivalent words;

[0028] If the comment information includes text comments, the word segments corresponding to the text comments are determined, and the keywords corresponding to the text comments are determined according to the word segments.

[0029] In one embodiment, the step of determining the relevance of the target lyrics block and the comment information associated therewith comprises:

[0030] Determine the vector angle between each target lyrics block and corresponding comment information;

[0031] Determine the average angle of the target lyrics block according to the vector angles of each of the comment information;

[0032] The degree of association is determined according to the average angle.

[0033] To achieve the above object, the present invention further provides a music clip selection device, the music clip selection device comprising:

[0034] An acquisition module, used to acquire comment information associated with each lyric block in the song, and determine the lyric block heat of each lyric block according to the comment information;

[0035] A calculation module, used to determine a target lyric block according to the popularity of the lyric block, and determine the correlation between the target lyric block and the comment information associated with it;

[0036] A determination module, configured to select a popular lyrics block from the target lyrics block according to the correlation;

[0037] The selection module is used to select the music clips corresponding to the popular lyrics.

[0038] To achieve the above-mentioned purpose, the present invention also provides a music clip selection device, which includes a memory, a processor, and a music clip selection program stored in the memory and executable on the processor, and when the music clip selection program is executed by the processor, it implements the various steps of the music clip selection method described above.

[0039] To achieve the above objectives, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a music clip selection program, and when the music clip selection program is executed by a processor, the various steps of the music clip selection method described above are implemented.

[0040] The present invention provides a method, device, equipment and medium for selecting music clips, which obtains comment information associated with each lyric block in a song, determines the lyric block heat of each lyric block according to the comment information; determines a target lyric block according to the lyric block heat, determines the correlation between the target lyric block and the comment information associated with it; selects a popular lyric block in the target lyric block according to the correlation; and selects the music clip corresponding to the popular lyrics. By determining the lyric block heat and correlation of the lyric block to select the popular lyric block, and determining the music clip corresponding to the popular lyric block, the accuracy of the selected music clip is improved, making the music clip more suitable for the theme of the song. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of the hardware structure of a music clip selection device involved in an embodiment of the present invention;

[0042] Figure 2 A schematic flow chart of a first embodiment of a method for selecting a music clip according to the present invention;

[0043] Figure 3 A schematic diagram of the vector angles between the target lyrics block and each comment information in the music clip selection method of the present invention;

[0044] Figure 4 A schematic flow chart of a second embodiment of the method for selecting a music clip according to the present invention;

[0045] Figure 5 This is a detailed flowchart of step S10 of the third embodiment of the method for selecting music clips of the present invention;

[0046] Figure 6 A schematic diagram of the comment popularity of comment information corresponding to each lyric block in the music clip selection method of the present invention;

[0047] Figure 7 A schematic diagram of the comment popularity of comment information corresponding to each lyric block in the music clip selection method of the present invention;

[0048] Figure 8 This is a detailed flowchart of step S12 of the fourth embodiment of the method for selecting music clips of the present invention;

[0049] Fig. 9 The figure is a schematic diagram of the logical structure of the music clip selection device of the present invention.

[0050] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0051] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0052] The main solution of the embodiment of the present invention is: obtaining comment information associated with each lyric block in a song, and determining the lyric block popularity of each lyric block based on the comment information; determining a target lyric block based on the lyric block popularity, and determining the correlation between the target lyric block and the comment information associated with it; selecting a popular lyric block in the target lyric block based on the correlation; and selecting a music clip corresponding to the popular lyrics.

[0053] By determining the heat and relevance of the lyrics blocks, a popular lyrics block is selected, and the music clip corresponding to the popular lyrics block is determined, thereby improving the accuracy of the selected music clip and making the music clip more suitable for the theme of the song.

[0054] As an implementation scheme, the music clip selection device can be as follows Figure 1 shown.

[0055] The embodiment of the present invention relates to a music clip selection device, which includes: a processor 101, such as a CPU, a memory 102, and a communication bus 103. The communication bus 103 is used to realize connection and communication between these components.

[0056] The memory 102 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. Figure 1 As shown, the memory 102 as a computer-readable storage medium may include a music clip selection program; and the processor 101 may be used to call the music clip selection program stored in the memory 102 and perform the following operations:

[0057] Obtaining comment information associated with each lyric block in the song, and determining the lyric block heat of each lyric block according to the comment information;

[0058] Determine a target lyric block according to the popularity of the lyric block, and determine the relevance of the target lyric block and its associated comment information;

[0059] Selecting a popular lyrics block from the target lyrics block according to the correlation;

[0060] Select the music clip corresponding to the popular lyrics.

[0061] In one embodiment, the processor 101 may be used to call the music clip selection program stored in the memory 102 and perform the following operations:

[0062] Obtaining the duration of the music segment corresponding to the target lyrics block;

[0063] A popular lyrics block is selected from each of the target lyrics blocks according to the relevance and the duration.

[0064] In one embodiment, the processor 101 may be used to call the music clip selection program stored in the memory 102 and perform the following operations:

[0065] Determine a difference between the duration and a preset duration threshold, and determine a deviation percentage according to a ratio of the difference to the preset duration threshold;

[0066] Sorting the association degrees from large to small, determining the ranking number of each association degree, and determining the deviation coefficient according to a preset value and the ranking number;

[0067] The hot lyrics block is selected from each of the target lyrics blocks according to the correlation degree, the deviation percentage and the deviation coefficient.

[0068] In one embodiment, the processor 101 may be used to call the music clip selection program stored in the memory 102 and perform the following operations:

[0069] Obtain comment information associated with each lyric block, and determine the keywords corresponding to each comment information;

[0070] Determine the frequency and probability of occurrence of the keyword;

[0071] Determining the comment popularity of the comment information according to the occurrence frequency and the occurrence probability;

[0072] The lyrics block heat of each lyrics block is determined according to the average heat of the comment heats of each of the comment information.

[0073] In one embodiment, the processor 101 may be used to call the music clip selection program stored in the memory 102 and perform the following operations:

[0074] Determine each equivalent keyword corresponding to the keyword, and determine the total number of times each equivalent keyword appears in the comment information of the song;

[0075] Determine the occurrence frequency of the keyword according to the total number of times;

[0076] The occurrence probability of the keyword is determined according to the occurrence frequency and the total number of the comment information.

[0077] In one embodiment, the processor 101 may be used to call the music clip selection program stored in the memory 102 and perform the following operations:

[0078] Get the comment information associated with each lyric block;

[0079] If the comment information includes expression comments and / or picture comments, determining label information corresponding to the expression comments and / or picture comments, determining equivalent words corresponding to the expression comments and / or picture comments according to the label information; determining the keywords corresponding to the expression comments and / or picture comments according to the equivalent words;

[0080] If the comment information includes text comments, the word segments corresponding to the text comments are determined, and the keywords corresponding to the text comments are determined according to the word segments.

[0081] In one embodiment, the processor 101 may be used to call the music clip selection program stored in the memory 102 and perform the following operations:

[0082] Determine the vector angle between each target lyrics block and corresponding comment information;

[0083] Determine the average angle of the target lyrics block according to the vector angles of each of the comment information;

[0084] The degree of association is determined according to the average angle.

[0085] Based on the hardware architecture of the above-mentioned music clip selection device, an embodiment of the music clip selection method of the present invention is proposed.

[0086] Reference Figure 2 , Figure 2 This is the first embodiment of the music clip selection method of the present invention, and the music clip selection method includes the following steps:

[0087] Step S10, obtaining comment information associated with each lyric block in the song, and determining the lyric block heat of each lyric block according to the comment information.

[0088] Specifically, the comment information associated with each lyric block of the song is obtained, the comment time of the comment information in the song is determined, and the lyric block corresponding to the comment information is determined according to the comment time. For example, when the comment time of the comment information is 1 minute and 11 seconds in the song to be processed, the time of the lyric block corresponding to the comment information can be 1 minute and 10 seconds to 1 minute and 15 seconds.

[0089] Determining the lyrics block heat of each lyric block according to the comment information may be determining the lyrics block heat of each lyric block according to the comment heat of the comment information associated with each lyric block, wherein the comment heat of the comment information is determined by the number of occurrences and the frequency of occurrence of the keywords in the comment information. Optionally, determining the lyrics block heat of each lyric block according to the average heat of the comment heat of each comment information. Optionally, determining a preset number of comment information with a high comment heat ranking, determining the average heat according to the comment heat of the preset number of comment information, and determining the lyrics block heat of the lyric block according to the average heat.

[0090] Step S20, determining a target lyrics block according to the popularity of the lyrics block, and determining the relevance between the target lyrics block and the comment information associated with it.

[0091] Specifically, the target lyric block is determined according to the lyric block heat of each lyric block. The lyric block heat greater than a preset heat can be determined as the target lyric block, or the lyric block heat of a preset number of lyrics blocks ranked at the top can be determined as the target lyric block.

[0092] Determine the relevance of the target lyrics block and its associated comment information. Figure 3 As shown, the vector angles between each target lyric block and the corresponding comment information can be determined; the average angle of the target lyric block can be determined based on the vector angles of the comment information; and the correlation can be determined based on the average angle. The comment information corresponding to each target lyric block can be determined to have a comment heat greater than a preset heat, or it can be determined to have a preset number of comment information ranked high in comment heat.

[0093] The average angle of the target lyrics block is determined according to the vector angles of each comment information, for example, as shown in the following formula:

[0094]

[0095] Among them, θ k is the angle between the kth comment information and the lyrics block, and N is the number of comment information.

[0096] The correlation is determined based on the average angle. If the average angle is smaller, the correlation between the preset number of comment information and the corresponding lyrics block is higher, and the theme expressed by the lyrics block can be better reflected; if the average angle is larger, the correlation between the preset number of comment information and the corresponding lyrics block is lower, and the theme expressed by the lyrics block cannot be reflected.

[0097] Step S30, selecting a popular lyrics block from the target lyrics block according to the correlation.

[0098] Specifically, a hot lyric block is selected from each target lyric block according to the correlation. The target lyric block with a correlation greater than a preset threshold may be selected as the hot lyric block, or the target lyric block with the greatest correlation may be selected as the hot lyric block.

[0099] Step S40: Select the music clip corresponding to the popular lyrics.

[0100] Specifically, after selecting a popular lyrics block, selecting the recommended music clip corresponding to the popular lyrics block can be to display the music clip with a duration of t corresponding to the selected popular lyrics block in the recommendation area of ​​the music platform; or the music clip with a duration of t corresponding to the selected popular lyrics block can be edited, for example, by stretching, compressing, cropping, etc., to process the music clip with a duration of t into a music clip with a preset duration T, and display the music clip with a duration T in the recommendation area of ​​the music platform.

[0101] In the technical solution of this embodiment, the comment information associated with each lyric block in the song is obtained, and the lyric block heat of each lyric block is determined according to the comment information; the target lyric block is determined according to the lyric block heat, and the correlation between the target lyric block and the comment information associated with it is determined; the popular lyric block is selected in the target lyric block according to the correlation; and the music clip corresponding to the popular lyrics is selected. By determining the lyric block heat and correlation of the lyric block to select the popular lyric block, and determining the music clip corresponding to the popular lyric block, the accuracy of the selected music clip is improved, making the music clip more suitable for the theme of the song.

[0102] Reference Figure 4 , Figure 4 This is a second embodiment of the music clip selection method of the present invention, based on the first embodiment, step S20 includes:

[0103] Step S21, obtaining the duration of the music segment corresponding to the target lyrics block;

[0104] Step S22, selecting a popular lyrics block from each of the target lyrics blocks according to the relevance and the duration.

[0105] Specifically, the duration of the music clip corresponding to the target lyrics block is obtained, and a popular lyrics block is selected from each target lyrics block according to the correlation and duration. The target lyrics block can be determined, whose correlation is greater than a preset threshold and whose duration difference with a preset duration is less than a preset difference, and the target lyrics block is used as the popular lyrics block.

[0106] Select the popular lyrics block from each target lyrics block according to the relevance and duration, or determine the difference between the duration and a preset duration threshold, and determine the deviation percentage according to the ratio of the difference to the preset duration threshold. Exemplarily, determine X target lyrics blocks and sort them from small to large according to the average angle of the comment information as (S1, S2, S3, ..., S X ), calculate the deviation percentage between the duration of X target lyrics blocks and the preset duration threshold in turn, as shown in the following formula:

[0107] Y i =|t i -T| / T×100%;

[0108] Among them, Y i Indicates the percentage of deviation between the duration of the i-th target lyrics block and the preset duration threshold, t i represents the duration of the i-th target lyrics block, T represents the preset duration threshold. The duration of the target lyrics block may be longer or shorter than the preset duration threshold, and the deviation percentage represents the degree of deviation of the target lyrics block relative to the preset duration threshold.

[0109] Sort the relevance from large to small, determine the ranking number of each relevance, and determine the deviation coefficient according to the preset value and the ranking number. For example, determine X target lyrics blocks, sort them from large to small according to the relevance, that is, sort them from small to large according to the average angle of the comment information (S1, S2, S3...S X ), for X lyrics blocks arranged in ascending order according to the total angle of comments. The deviation coefficient is determined according to the preset value and the sorting number, as shown in the following formula:

[0110] K i =(ai) / a;

[0111] Among them, K i represents the deviation coefficient of the i-th target lyrics block, i represents the ranking number of the i-th target lyrics block, a is a preset value, and a can be 100.

[0112] A hot lyric block is selected from each target lyric block according to the relevance, deviation percentage and deviation coefficient. A target lyric block having a relevance greater than a preset threshold, a deviation percentage less than a preset percentage and a deviation coefficient greater than a preset threshold can be determined as a hot lyric block.

[0113] In order to determine the target lyrics block whose duration has a smaller deviation from the preset duration threshold, and the target lyrics block whose comment information is more closely related to the target lyrics block, the comprehensive fit can also be determined based on the correlation, deviation percentage and deviation coefficient. When the comprehensive fit is greater than the preset threshold, the target lyrics block corresponding to the comprehensive fit is regarded as a popular lyrics block. The calculation of the comprehensive fit of the target lyrics block is shown in the following formula:

[0114] M=Avg θi ×K i / Y i ;

[0115] Among them, M represents the comprehensive fit of the target lyrics block, Avg θi represents the average angle of the i-th target lyrics block, K i represents the deviation coefficient of the i-th target lyrics block, Y i Indicates the percentage deviation between the duration of the i-th target lyrics block and the preset duration threshold.

[0116] In the technical solution of this embodiment, the duration of the music segment corresponding to the target lyric block is obtained; the popular lyric block is selected from each target lyric block according to the relevance and the duration. The target lyric block is determined by the lyric block popularity, the popular lyric block is selected from each target lyric block according to the relevance and the duration, and the music segment corresponding to the popular lyric block is determined. The duration of the selected music segment has a small difference from the preset duration threshold, avoiding the situation that the music segment is too long or too short, improving the accuracy of the selected music segment, and making the recommended music segment more in line with the song theme.

[0117] Referring to Figure 5 , Figure 5 This is the third embodiment of the music segment selection method of the present invention. Based on the first or second embodiment, the step S10 includes:

[0118] Step S11, obtaining the comment information associated with each lyric block, and determining the keyword corresponding to each comment information;

[0119] Step S12, determining the occurrence frequency and occurrence probability of the keyword;

[0120] Step S13, determining the comment popularity of the comment information according to the occurrence frequency and the occurrence probability;

[0121] Step S14, determining the lyric block popularity of each lyric block according to the average popularity of the comment popularity of each comment information.

[0122] Specifically, to obtain the comment information associated with each lyric block and determine the keyword corresponding to each comment information, the comment information can be segmented according to a preset algorithm to obtain multiple segmented words, and stop words are removed. After removing words such as "of", "this", "ah" that do not affect the semantics, the keyword corresponding to the comment information is obtained.

[0123] Obtain the comment information associated with each lyric block and determine the keyword corresponding to each comment information. The comment information associated with each lyric block can be obtained, where the comment information can include emoji comments and text comments, can include picture comments and text comments, and can also include emoji comments, picture comments and text comments. When the comment information includes emoji comments and / or picture comments, determine the tag information corresponding to the emoji comments and / or picture comments, and determine the equivalent words corresponding to the emoji comments and / or picture comments according to the tag information; determine the keyword corresponding to each comment information according to the equivalent words. When the comment information includes text comments, determine the segmented words corresponding to the text comments, and determine the keyword corresponding to the text comments according to the segmented words.

[0124] Determine the occurrence frequency and occurrence probability of the keyword. Determine the comment popularity of the keyword according to the occurrence frequency and occurrence probability of the keyword. Exemplarily, as shown in the following formula:

[0125] r i =WFS i ×FFI i ;

[0126] Among them, r i Indicates the popularity of comments on keyword i, WFS i Indicates the frequency of occurrence of keyword i, FFI i represents the occurrence probability of keyword i.

[0127] The comment popularity of the comment information is determined according to the sum of the comment popularity of each keyword corresponding to the comment information. For example, the following formula is shown:

[0128]

[0129] Among them, R represents the comment popularity of the comment information, r i represents the popularity of comments for keyword i, and n represents the total number of keywords.

[0130] like Figure 6 As shown in FIG. 1 , the comment heat of multiple comment information corresponding to each lyric block. The lyric block heat of each lyric block is determined according to the average heat of the comment heat of each comment information. Figure 7 As shown, the lyrics block heat of each lyrics block can be determined according to the m comments with the highest comment heat ranking, where m is 4, as shown in the following formula for example:

[0131]

[0132] Among them, P represents the lyrics block heat of the lyrics block, R j represents the comment popularity of comment information j of the lyrics block, and m represents the total number of comment information.

[0133] In the technical solution of this embodiment, the comment information associated with each lyric block is obtained to determine the keywords corresponding to each comment information; the occurrence frequency and occurrence probability of the keywords are determined; the comment heat of the comment information is determined based on the occurrence frequency and occurrence probability; the lyric block heat of each lyric block is determined based on the average heat of the comment heat of each comment information. The comment heat of the comment information is determined by the occurrence frequency and occurrence probability of the keywords, and the lyric block heat of the lyric block is determined based on the comment heat of each comment information, so as to determine the target lyric block based on the lyric block heat, so that the determined target lyric block is more accurate, and the accuracy of subsequent hot lyric block selection is improved.

[0134] Reference Figure 8 , Figure 8 This is a fourth embodiment of the music clip selection method of the present invention, based on the third embodiment, and step S12 includes:

[0135] Step S121, determining each equivalent keyword corresponding to the keyword, and determining the total number of times each equivalent keyword appears in the comment information of the song;

[0136] Step S122, determining the occurrence frequency of the keyword according to the total number of times;

[0137] Step S123: determining the occurrence probability of the keyword according to the occurrence frequency and the total number of the comment information.

[0138] Specifically, determine the equivalent keywords corresponding to the keyword. The keyword may correspond to multiple equivalent keywords. For example, if the keyword is admire, the equivalent keywords include admire, kneel down, and prostrate oneself. It is also possible to determine the word frequency S of each equivalent keyword in the comment information of the song to be processed. i , take the equivalent keyword with the maximum word frequency as the equivalent representative word, and use the equivalent representative word to replace the keyword. For example, in the lyrics block, the equivalent keyword "admire" has a word frequency of 10, "bow" has a word frequency of 2, and "prostrate oneself" has a word frequency of 11, then "prostrate oneself" is taken as the equivalent representative word.

[0139] Determine the total number of times each equivalent keyword appears in the comment information of the song, and determine the frequency of occurrence of the keyword based on the total number of times. For example, the equivalent keywords corresponding to keyword W100 include keyword W100, keyword W1023, and keyword W1125, and the frequency of occurrence of keyword W100 is shown in the following formula:

[0140] WFS_100=WF_100+WF_1023+WF_1125;

[0141] Among them, WFS_100 represents the occurrence frequency of keyword W100, WF_100 represents the word frequency of keyword W100, WF_1023 represents the word frequency of keyword W1023, and WF_1125 represents the word frequency of keyword W1125.

[0142] The occurrence probability of a keyword is determined based on the occurrence frequency and the total number of comment information. Exemplarily, specifically, the occurrence probability of a keyword is shown in the following formula:

[0143]

[0144] Among them, FFI represents the probability of occurrence of keywords, WFS represents the frequency of occurrence of keywords, and C represents the total number of comment information.

[0145] In the technical solution of this embodiment, each equivalent keyword corresponding to the keyword is determined, and the total number of times each equivalent keyword appears in the comment information of the song is determined; the frequency of occurrence of the keyword is determined based on the total number; and the probability of occurrence of the keyword is determined based on the frequency of occurrence and the total number of comment information. By determining the frequency of occurrence and probability of occurrence of the keyword, it is convenient to subsequently determine the comment heat of the comment information, thereby improving the accuracy of the comment heat calculation.

[0146] Reference Fig. 9 The present invention also provides a music clip selection device, the music clip selection device comprising:

[0147] The acquisition module 100 is used to acquire comment information associated with each lyric block in the song, and determine the lyric block heat of each lyric block according to the comment information;

[0148] The calculation module 200 is used to determine the target lyrics block according to the popularity of the lyrics block, and determine the correlation between the target lyrics block and the comment information associated with it;

[0149] A determination module 300, configured to select popular lyrics in the target lyrics block according to the relevance;

[0150] The selection module 400 is used to select the music clip corresponding to the popular lyrics.

[0151] In one embodiment, in selecting a popular lyric block from the target lyric block according to the correlation, the determination module 300 is specifically used to:

[0152] Obtaining the duration of the music segment corresponding to the target lyrics block;

[0153] A popular lyrics block is selected from each of the target lyrics blocks according to the relevance and the duration.

[0154] In one embodiment, in selecting a hot lyric block from each of the target lyric blocks according to the relevance and the duration, the determination module 300 is specifically configured to:

[0155] Determine a difference between the duration and a preset duration threshold, and determine a deviation percentage according to a ratio of the difference to the preset duration threshold;

[0156] Sorting the association degrees from large to small, determining the ranking number of each association degree, and determining the deviation coefficient according to a preset value and the ranking number;

[0157] The hot lyrics block is selected from each of the target lyrics blocks according to the correlation degree, the deviation percentage and the deviation coefficient.

[0158] In one embodiment, in obtaining comment information associated with each lyric block in a song and determining the lyric block popularity of each lyric block according to the comment information, the obtaining module 100 is specifically used to:

[0159] Obtain comment information associated with each lyric block, and determine the keywords corresponding to each comment information;

[0160] Determine the frequency and probability of occurrence of the keyword;

[0161] Determining the comment popularity of the comment information according to the occurrence frequency and the occurrence probability;

[0162] The lyrics block heat of each lyrics block is determined according to the average heat of the comment heats of each of the comment information.

[0163] In one embodiment, in determining the occurrence frequency and occurrence probability of the keyword, the acquisition module 100 is specifically used to:

[0164] Determine each equivalent keyword corresponding to the keyword, and determine the total number of times each equivalent keyword appears in the comment information of the song;

[0165] Determine the occurrence frequency of the keyword according to the total number of times;

[0166] The occurrence probability of the keyword is determined according to the occurrence frequency and the total number of the comment information.

[0167] In one embodiment, in terms of obtaining the comment information associated with each lyric block and determining the keywords corresponding to each comment information, the obtaining module 100 is specifically used to:

[0168] Get the comment information associated with each lyric block;

[0169] If the comment information includes expression comments and / or picture comments, determining label information corresponding to the expression comments and / or picture comments, determining equivalent words corresponding to the expression comments and / or picture comments according to the label information; determining the keywords corresponding to the expression comments and / or picture comments according to the equivalent words;

[0170] If the comment information includes text comments, the word segments corresponding to the text comments are determined, and the keywords corresponding to the text comments are determined according to the word segments.

[0171] In one embodiment, in determining the relevance between the target lyrics block and the comment information associated therewith, the calculation module 200 is specifically used to:

[0172] Determine the vector angle between each target lyrics block and corresponding comment information;

[0173] Determine the average angle of the target lyrics block according to the vector angles of each of the comment information;

[0174] The degree of association is determined according to the average angle.

[0175] The present invention also provides a music clip selection device, which includes a memory, a processor, and a music clip selection program stored in the memory and executable on the processor. When the music clip selection program is executed by the processor, the various steps of the music clip selection method described in the above embodiment are implemented.

[0176] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a music clip selection program, and when the music clip selection program is executed by a processor, the various steps of the music clip selection method described in the above embodiment are implemented.

[0177] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0178] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, system, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, system, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, system, article or device including the element.

[0179] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment system can be implemented by means of software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a computer-readable storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a parking management device, an air conditioner, or a network device, etc.) to execute the system described in each embodiment of the present invention.

[0180] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for selecting a music clip, characterized in that: The music clip selection method comprises: Obtaining comment information associated with each lyric block in the song, and determining the lyric block heat of each lyric block according to the comment information; Determine a target lyric block according to the popularity of the lyric block, and determine the relevance of the target lyric block and its associated comment information; Obtaining the duration of the music segment corresponding to the target lyrics block; Selecting a popular lyrics block from each of the target lyrics blocks according to the relevance and the duration; Select the music segment corresponding to the popular lyrics block.

2. The method for selecting a music clip according to claim 1, wherein: The step of selecting a hot lyrics block from each of the target lyrics blocks according to the relevance and the duration comprises: Determine a difference between the duration and a preset duration threshold, and determine a deviation percentage according to a ratio of the difference to the preset duration threshold; Sorting the association degrees from large to small, determining the ranking number of each association degree, and determining the deviation coefficient according to a preset value and the ranking number; The hot lyrics block is selected from each of the target lyrics blocks according to the correlation degree, the deviation percentage and the deviation coefficient.

3. The method for selecting a music clip according to claim 1, wherein: The step of obtaining comment information associated with each lyric block in the song and determining the popularity of each lyric block according to the comment information comprises: Obtain comment information associated with each lyric block, and determine the keywords corresponding to each comment information; Determine the frequency and probability of occurrence of the keyword; Determining the comment popularity of the comment information according to the occurrence frequency and the occurrence probability; The lyrics block heat of each lyrics block is determined according to the average heat of the comment heats of each of the comment information.

4. The method for selecting a music clip according to claim 3, wherein: The step of determining the occurrence frequency and occurrence probability of the keyword includes: Determine each equivalent keyword corresponding to the keyword, and determine the total number of times each equivalent keyword appears in the comment information of the song; Determine the occurrence frequency of the keyword according to the total number of times; The occurrence probability of the keyword is determined according to the occurrence frequency and the total number of the comment information.

5. The method for selecting a music clip according to claim 3, wherein: The step of obtaining the comment information associated with each lyric block and determining the keyword corresponding to each comment information comprises: Get the comment information associated with each lyric block; If the comment information includes expression comments and / or picture comments, determining label information corresponding to the expression comments and / or picture comments, determining equivalent words corresponding to the expression comments and / or picture comments according to the label information; determining the keywords corresponding to the expression comments and / or picture comments according to the equivalent words; If the comment information includes text comments, the word segments corresponding to the text comments are determined, and the keywords corresponding to the text comments are determined according to the word segments.

6. The method for selecting a music clip according to claim 1, wherein: The step of determining the relevance of the target lyrics block and the comment information associated therewith comprises: Determine the vector angle between each target lyrics block and corresponding comment information; Determine the average angle of the target lyrics block according to the vector angles of each of the comment information; The degree of association is determined according to the average angle.

7. A music clip selection device, characterized in that: The music clip selection device comprises: An acquisition module, used to acquire comment information associated with each lyric block in the song, and determine the lyric block heat of each lyric block according to the comment information; A calculation module, used to determine a target lyric block according to the popularity of the lyric block, and determine the correlation between the target lyric block and the comment information associated with it; A determination module, configured to obtain the duration of the music segment corresponding to the target lyrics block; and select a popular lyrics block from each of the target lyrics blocks according to the correlation and the duration; The selection module is used to select the music segment corresponding to the popular lyrics block.

8. A music clip selection device, characterized in that: The music clip selection device includes a memory, a processor, and a music clip selection program stored in the memory and executable on the processor. When the music clip selection program is executed by the processor, it implements the various steps of the music clip selection method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a music clip selection program, and when the music clip selection program is executed by a processor, each step of the music clip selection method according to any one of claims 1 to 6 is implemented.

Images