CN108268530B

CN108268530B - Lyric score generation method and related device

Info

Publication number: CN108268530B
Application number: CN201611264888.5A
Authority: CN
Inventors: 叶舟; 王瑜; 张亚楠; 苏飞; 杨洋
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2016-12-30
Filing date: 2016-12-30
Publication date: 2022-04-29
Anticipated expiration: 2036-12-30
Also published as: CN108268530A; WO2018121368A1; TW201824249A

Abstract

The embodiment of the invention discloses a lyric score generation method and a related device, wherein a score server comprises a score library with a first corresponding relation between participles and score segments, after a plurality of participles obtained by participle processing on lyrics to be scored are obtained, a score segment set corresponding to the participles can be determined from the score library by using the participles according to the first corresponding relation, then the score segments corresponding to the participles in a character segment are used as a unit, a score segment set corresponding to the participles in the character segment is used for combining the sub-scores corresponding to one character segment, and the sub-scores of the character segments are spliced into the score of the lyrics to be scored, because the score segments corresponding to the participles are pre-stored, the score segments corresponding to the participles can be matched after the lyrics to be scored, so that the score segments to be matched and the positions of the participles in the lyrics to be scored can be used for automatically generating the score of the lyrics to be scored, the efficiency of the lyric music is effectively improved.

Description

Lyric score generation method and related device

Technical Field

The invention relates to the field of data processing, in particular to a lyric score generation method and a related device.

Background

In music creation, lyrics are created first and then music composition is performed according to the lyrics, that is, the lyrics are used to complete corresponding music matching, thereby forming a song.

How to rapidly complete music matching for lyrics is an urgent problem to be solved, and if the music matching can be automatically carried out according to the content of the lyrics, the efficiency of music creation is greatly improved.

However, there is currently no way in which lyrics can be automatically dubbed according to their content.

Disclosure of Invention

In order to solve the technical problem, the invention provides a lyric score generation method and a related device, so that the score of lyrics to be scored is automatically generated by utilizing the matched score segments and the positions of the participles in the lyrics to be scored, and the efficiency of scoring the lyrics is effectively improved.

The embodiment of the invention discloses the following technical scheme:

in a first aspect, the present invention provides a lyric score generating method, applied to a score server, where the score server includes a score library including a first correspondence between participles and score segments, where any participle has a score segment set including at least one score segment, the method including:

acquiring a plurality of participles acquired by participling the lyrics to be dubbed, wherein the lyrics to be dubbed comprise at least one text segment, and one text segment in the at least one text segment comprises at least one participle;

searching the score library according to the multiple participles, and determining a score fragment set corresponding to each participle by using the first corresponding relation;

determining sub-scores corresponding to the at least one text segment according to the score segment sets corresponding to the multiple scores and the scores included by the at least one text segment;

and generating the score of the lyrics to be scored by splicing the sub scores respectively corresponding to the at least one character fragment.

Optionally, the first text segment is one of the at least one text segment, and the determining, according to the score segment set corresponding to the multiple participles and the participles included in the at least one text segment, the sub-scores respectively corresponding to the at least one text segment includes:

acquiring a score segment set corresponding to the participles included in the first character segment;

determining the fluency between score segments selected from a score segment set corresponding to the participles included in the first text segment according to the adjacent relation of the participles included in the first text segment;

and taking a group of dubbing music segments with fluency meeting preset conditions as the sub-dubbing music of the first character segment.

Optionally, the determining, according to the adjacent relationship between the participles included in the first text segment, the fluency between the selected score segments in the score segment set corresponding to the participles included in the first text segment includes:

calculating the splicing degree between the score segment in the score segment set corresponding to the first participle and the score segment in the score segment set corresponding to the second participle in the first character segment to obtain at least one splicing score between the first participle and the second participle;

calculating at least one splicing total score of the first character segment according to the splicing score of adjacent participles in the first character segment;

the taking a group of score segments with fluency meeting a preset threshold as the sub-scores of the first text segment comprises:

and selecting a group of score segments corresponding to one splicing total score from the splicing total scores exceeding the preset threshold value as sub scores of the first character segment.

Optionally, the calculating a degree of concatenation between a score segment in a score segment set corresponding to a first participle in the first text segment and a score segment in a score segment set corresponding to a second participle to obtain at least one concatenation score between the first participle and the second participle includes:

acquiring one score segment in the score segment set corresponding to the first participle, and acquiring one score segment in the score segment set corresponding to the second participle;

and calculating the splicing degree between the two score segments to obtain a splicing score determined by the two score segments between the first participle and the second participle.

Optionally, the score server further includes a genre library, where the genre library includes possibility information that the participles belong to different song genres, and the score library further includes a second correspondence between score segments and song genres; before searching the score library according to the multiple participles and determining a score fragment set corresponding to the multiple participles, the method further comprises the following steps:

searching the type library according to the plurality of participles, and determining the song type of the song word to be matched;

the searching the score library according to the multiple participles and determining score fragment sets corresponding to the multiple participles by using the first corresponding relation comprise:

screening out score segments corresponding to the song types to which the song words to be scored belong from the score library according to the second corresponding relation;

and determining a score segment set corresponding to the plurality of participles from score segments corresponding to the song type to which the song word to be scored belongs according to the first corresponding relation.

Optionally, the method further includes:

acquiring historical songs, wherein the historical songs comprise historical lyrics, historical score and song types;

performing word segmentation on the historical lyrics;

counting the occurrence frequency of a third participle in the word segmentation process and the corresponding relation between the third participle and the song type of the historical song to which the third participle belongs, so as to determine the occurrence frequency of the third participle in the same song type, wherein the third participle is a participle obtained from the historical lyrics;

determining possibility information that the third participle belongs to different song types according to the occurrence frequency of the third participle in the participle process and the occurrence frequency of the third participle in the same song type;

and establishing the type library according to the possibility information that the participles obtained from the historical lyrics belong to different song types.

Optionally, the method further includes:

performing word segmentation on the historical lyrics;

determining a corresponding score segment in the historical score according to the obtained word segmentation;

and establishing the music matching library according to the first corresponding relation between the obtained word segmentation and the music matching segment in the historical music matching and the second corresponding relation between the music matching segment in the historical music matching and the song type.

Optionally, the obtaining a plurality of participles obtained by performing a participle processing on the lyrics of the to-be-dubbed music includes:

and performing word segmentation processing on the acquired lyrics to be matched to obtain a plurality of word segments.

In a second aspect, the present invention provides an apparatus for generating a score of a lyric, applied to a score server, the score server including a score library including a first correspondence relationship between participles and score pieces, wherein any one of the participles has a score piece set including at least one score piece, the apparatus including an obtaining unit, a searching unit, a determining unit, and a generating unit:

the acquiring unit is used for acquiring a plurality of participles acquired by performing participle processing on lyrics to be matched, wherein the lyrics to be matched comprise at least one text segment, and one text segment in the at least one text segment comprises at least one participle;

the searching unit is used for searching the score library according to the multiple participles and determining a score fragment set corresponding to each participle by using the first corresponding relation;

the determining unit is configured to determine sub-scores corresponding to the at least one text segment according to the score segment sets corresponding to the multiple scores and the scores included in the at least one text segment;

and the generating unit is used for generating the score of the lyrics to be scored by splicing the sub scores respectively corresponding to the at least one character fragment.

Optionally, the first text segment is one of the at least one text segment, and the determining unit is specifically configured to obtain a score segment set corresponding to a participle included in the first text segment; determining the fluency between score segments selected from a score segment set corresponding to the participles included in the first text segment according to the adjacent relation of the participles included in the first text segment; and taking a group of dubbing music segments with fluency meeting preset conditions as the sub-dubbing music of the first character segment.

Optionally, the first participle and the second participle are adjacent participles in the first text segment, and the determining unit includes a calculating subunit, an obtaining subunit, and as subunits:

the calculating subunit is configured to calculate a splicing degree between a score segment in a score segment set corresponding to a first participle in the first text segment and a score segment in a score segment set corresponding to a second participle, and obtain at least one splicing score between the first participle and the second participle;

the obtaining subunit is configured to calculate at least one total concatenation score of the first text segment according to the concatenation scores of adjacent participles in the first text segment;

and the sub-unit is used for selecting a group of score segments corresponding to one splicing total score from the splicing total scores exceeding the preset threshold value as sub-scores of the first character segment.

Optionally, the computing subunit is specifically configured to obtain one score segment in the score segment set corresponding to the first participle, and obtain one score segment in the score segment set corresponding to the second participle; and calculating the splicing degree between the two score segments to obtain a splicing score determined by the two score segments between the first participle and the second participle.

Optionally, the score server further includes a genre library, where the genre library includes possibility information that the participles belong to different song genres, and the score library further includes a second correspondence between score segments and song genres; the apparatus further comprises a determine song type unit:

the song type determining unit is used for searching the type library according to the plurality of participles and determining the song type of the song word to be matched;

the searching unit comprises a screening subunit and a determining subunit:

the screening subunit is configured to screen out, from the score library, a score segment corresponding to the song type to which the song word to be scored belongs according to the second correspondence;

and the determining subunit is configured to determine, according to the first correspondence, a score segment set corresponding to the multiple participles from score segments corresponding to the song type to which the to-be-scored song word belongs.

Optionally, the apparatus further includes a history song obtaining unit, a word segmentation unit, a statistics unit, an information determining unit, and an establishing unit:

the historical song obtaining unit is used for obtaining historical songs, and the historical songs comprise historical lyrics, historical score and song types;

the word segmentation unit is used for segmenting the historical lyrics;

the statistical unit is used for counting the occurrence frequency of a third participle in the word segmentation process and the corresponding relation between the third participle and the song type of the historical song to which the third participle belongs, so as to determine the occurrence frequency of the third participle in the same song type, wherein the third participle is a participle obtained from the historical lyrics;

the information determining unit is used for determining possibility information that the third participle belongs to different song types according to the occurrence frequency of the third participle in the participle process and the occurrence frequency of the third participle in the same song type;

the establishing unit is used for establishing the type library according to the possibility information of the participles obtained from the historical lyrics and belonging to different song types.

Optionally, the apparatus further includes a history song obtaining unit, a word segmentation unit, a score determining unit, and a creating unit:

the word segmentation unit is used for segmenting the historical lyrics;

the score determining unit is used for determining corresponding score segments in the historical scores according to the obtained participles;

the establishing unit is used for establishing the music matching library according to the first corresponding relation between the obtained segmented words and the music matching segments in the historical music matching and the second corresponding relation between the music matching segments in the historical music matching and the song types.

Optionally, the obtaining unit is specifically configured to perform word segmentation processing on the obtained song word to be matched to obtain a plurality of word segments.

In a third aspect, the present invention provides a lyric score generating method, applied to a score server, where the score server includes a score library including first corresponding relations of segments and score segments, where any one segment has a score segment set including at least one score segment, the method including:

acquiring at least one fragment obtained by performing word segmentation processing on lyrics to be dubbed music;

searching the music matching library according to the at least one segment, and determining a music matching segment set corresponding to the at least one segment respectively through the first corresponding relation;

according to the lyrics to be dubbed music, determining a dubbing music fragment corresponding to each fragment in the at least one fragment from a dubbing music fragment set corresponding to the at least one fragment respectively;

and generating the score of the lyrics to be scored by splicing the determined score segments.

Optionally, the obtaining at least one segment obtained by performing word segmentation processing on the lyrics to be dubbed music includes:

and acquiring a plurality of participles obtained by performing participle processing on the to-be-participled music song words.

Optionally, the segment is a text segment, and the song word to be matched includes at least one text segment.

Optionally, the to-be-dubbed lyrics include at least one text segment, and one text segment in the at least one text segment includes at least one participle, and then, according to the to-be-dubbed lyrics, determining a dubbing fragment corresponding to each participle in the at least one participle from a dubbing fragment set corresponding to each participle, respectively, includes:

determining sub-scores corresponding to the at least one text segment according to the score segment set corresponding to the multiple segments and the segments included in the at least one text segment, wherein the sub-scores corresponding to the text segment are obtained from the score segments corresponding to the segments included in the one text segment;

the generation of the score of the lyrics to be scored through the determined score segments by splicing comprises the following steps:

In a fourth aspect, the present invention provides an apparatus for generating a score of lyrics, applied to a score server, where the score server includes a score library including a first correspondence relationship between segments and score segments, where any one segment has a score segment set including at least one score segment, the apparatus includes an obtaining unit, a searching unit, a determining unit, and a splicing unit:

the acquiring unit is used for acquiring at least one fragment obtained by performing word segmentation processing on lyrics of the score to be dubbed;

the searching unit is configured to search the music matching library according to the at least one segment, and determine, according to the first corresponding relationship, a music matching segment set corresponding to the at least one segment respectively;

the determining unit is configured to determine, according to the lyrics to be dubbed music, a dubbing music fragment corresponding to each fragment in the at least one fragment from a dubbing music fragment set corresponding to the at least one fragment;

and the splicing unit is used for generating the score of the lyrics to be matched by splicing the determined score segments.

Optionally, the obtaining unit is specifically configured to obtain a plurality of segmented words obtained by performing word segmentation processing on the to-be-matched lyrics.

Optionally, the determining unit is specifically configured to determine, according to the score segment sets corresponding to the multiple score segments and the score included in the at least one text segment, the sub-scores corresponding to the at least one text segment, respectively, where the sub-scores corresponding to the text segment are obtained from the score segments corresponding to the score included in the text segment;

the splicing unit is specifically configured to generate the score of the lyric to be scored by splicing the sub-scores respectively corresponding to the at least one text segment.

In a fifth aspect, the present invention provides a method for acquiring an ancillary music of a lyric, which is applied to an interactive end, and the method includes:

sending the acquired song words to be matched to a music server, wherein the song words to be matched comprise at least one text segment, one text segment in the at least one text segment comprises at least one participle, the music server comprises a music library, the music library comprises a first corresponding relation of the participle and the music segment, and any participle has a music segment set comprising at least one music segment;

and acquiring the score corresponding to the lyrics to be scored from the score server, wherein the score is obtained by splicing score segments in a score segment set corresponding to the participles in the lyrics to be scored according to the score segment server, and the score segment set corresponding to the participles in the lyrics to be scored is obtained by searching the score library according to the first corresponding relation by the score server.

Optionally, the obtaining of the score corresponding to the lyric of the to-be-scored from the score server includes:

obtaining a plurality of pending scores corresponding to the lyrics of the to-be-dubbed music from the dubbing music server, wherein the pending scores carry the dubbing music information;

and selecting the undetermined score which meets the requirement as the score of the lyrics of the to-be-scored according to the score information of the to-be-scored scores.

Optionally, the score and/or score type of the pending score carrying the score information are included in the score information.

Optionally, after obtaining the score corresponding to the lyric of the to-be-scored from the score server, the method further includes:

and if the obtained score does not meet the requirement, sending feedback information to the score server so that the server regenerates the score of the lyrics to be scored according to the feedback information.

Optionally, the feedback information includes information describing the requirement.

In a sixth aspect, the present invention provides an apparatus for acquiring a score of a lyric, which is applied to an interactive end, and the apparatus includes a sending unit and an acquiring unit:

the sending unit is used for sending the acquired to-be-matched lyrics to a music server, wherein the to-be-matched lyrics comprise at least one text segment, one text segment in the at least one text segment comprises at least one participle, the music server comprises a music library, the music library comprises a first corresponding relation between the participle and the music segment, and any participle has a music segment set comprising at least one music segment;

the obtaining unit is used for obtaining the score corresponding to the lyrics to be matched from the score server, the score is obtained by splicing score segments in a score segment set corresponding to the participles in the lyrics to be matched by the score server according to the to-be-matched score, and the score segment set corresponding to the participles in the lyrics to be matched is obtained by searching the score library by the score server according to the first corresponding relation.

Optionally, the acquiring unit includes an acquiring subunit and a selecting subunit:

the acquiring subunit is configured to acquire, from the music server, a plurality of pending music matches corresponding to the lyrics of the music to be matched, where the pending music matches carry music matching information;

and the selection subunit is used for selecting the undetermined score according with the demand as the score of the lyrics of the to-be-assigned score according to the score information of the to-be-assigned scores.

Optionally, the apparatus further includes a feedback unit:

and if the obtained score does not meet the requirement, triggering the feedback unit, wherein the feedback unit is used for sending feedback information to the score server so that the score of the lyrics to be scored is regenerated by the server according to the feedback information.

In a seventh aspect, the present invention provides a lyric dubbing system, where the dubbing system includes a dubbing server and an interactive end:

the music score server comprises a music score library, wherein the music score library comprises a first corresponding relation of participles and music scores, any participle is provided with a music score set comprising at least one music score, the music score server is used for acquiring a plurality of participles acquired by performing word segmentation processing on lyrics to be matched, the lyrics to be matched comprise at least one text score, and one text score in the at least one text score comprises at least one participle; searching the score library according to the multiple participles, and determining a score fragment set corresponding to each participle by using the first corresponding relation; determining sub-scores corresponding to the at least one text segment according to the score segment sets corresponding to the multiple scores and the scores included by the at least one text segment; generating the score of the lyrics to be scored by splicing the sub scores respectively corresponding to the at least one text segment;

the interactive end is used for sending the acquired lyrics to be dubbed music to the dubbing music server; and acquiring the score corresponding to the lyrics of the to-be-assigned score from the score server.

In an eighth aspect, the present invention provides a music editor, the music editor having an input interface for obtaining lyrics of an input to-be-dubbed music, and an editing interface for displaying the dubbed music;

the input interface is used for acquiring lyrics to be matched with music, and the acquired lyrics to be matched with music are the lyrics input in the input interface or the lyrics obtained after the input voice is recognized;

when the output instruction is obtained, the music editor is used for sending the music words to be matched, which are obtained through the input interface, to a music server, the music server comprises a music library, the music library comprises participles and a first corresponding relation of music segments, and any participle is provided with a music segment set comprising at least one music segment;

when the score corresponding to the lyrics to be dubbed is obtained from the music server, the music editor is used for displaying the score corresponding to the lyrics to be dubbed in the editing interface, the score corresponding to the lyrics to be dubbed is obtained by splicing the music server according to the score segments in the music segment set corresponding to the participles in the lyrics to be dubbed, and the music segment set corresponding to the participles in the lyrics to be dubbed is obtained by searching the music library according to the first corresponding relation by the music server.

Optionally, the editing interface has an editing button, and when a trigger to the editing button is received, the editing interface is configured to set the displayed score as editable;

the score displayed by the editing interface further comprises the score obtained by identifying the recorded audio data.

In a ninth aspect, the present invention provides a music player having a mixing interface and a playing interface;

the mixing interface is used for mixing lyrics to be matched with music and the score corresponding to the lyrics to be matched with the music to obtain a song, the score corresponding to the lyrics to be matched with the music is obtained by splicing score segments in a score segment set corresponding to participles in the lyrics to be matched with the music according to a first corresponding relation, the score segment set corresponding to the participles in the lyrics to be matched is obtained by searching a score library by the score server according to the first corresponding relation, the score server comprises a score library, the score library comprises the first corresponding relation of the participles and the score segments, and any participle has a score segment set comprising at least one score segment;

the playing interface is used for playing the songs.

It can be seen from the above technical solutions that a score library including a first correspondence between the participles and score segments is preset in the score server, after a plurality of participles obtained by performing a segmentation process on the lyrics to be scored are obtained, the respective corresponding score segment sets of the participles can be determined from the score library according to the first correspondence by using the participles, then the score segments of the lyrics to be scored are taken as units, a score segment set corresponding to the participles in one text segment is used to combine the sub-scores corresponding to one text segment, and the sub-scores of the text segments are spliced into the score of the lyrics to be scored, and thus, since the score segments corresponding to the participles are prestored, the score segments corresponding to the participles can be matched after the lyrics to be scored, and thus the score of the lyrics to be scored can be automatically generated by using the matched score segments and the positions of the participles in the lyrics to be scored, the efficiency of the lyric music is effectively improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

FIG. 1 is a schematic diagram of a system for score generation of lyrics according to an embodiment of the present invention;

fig. 2 is a flowchart of a method for generating a score of lyrics according to an embodiment of the present invention;

fig. 2a is a flowchart of a method for determining a sub-score corresponding to a first text segment according to an embodiment of the present invention.

Fig. 3 is a schematic flow chart of a procedure for establishing a score library according to an embodiment of the present invention;

FIG. 4 is a flowchart illustrating a type library creating process according to an embodiment of the present invention;

fig. 5 is a flowchart of another method for generating lyrics score according to an embodiment of the present invention;

fig. 6 is a flowchart of a method for acquiring score of lyrics according to an embodiment of the present invention;

fig. 7 is a diagram illustrating an apparatus of an apparatus for generating a score of lyrics according to an embodiment of the present invention;

fig. 8 is a diagram showing an apparatus configuration of an apparatus for generating a score of lyrics according to another embodiment of the present invention;

fig. 9 is a device configuration diagram of a lyric score obtaining device according to an embodiment of the present invention;

fig. 10 is a system configuration diagram of a lyric dubbing system according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some embodiments, but not all embodiments, of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In the process of music creation, the completion of a song generally comprises the steps of firstly making words and then composing the song, namely firstly creating lyrics, and then composing the song (matching the music) according to the lyrics to finally form a song. And matching songs often takes a lot of time. With the development of deep learning in the field of voice, more and more tasks can be automatically completed through a machine, and the efficiency of music creation can be greatly improved if automatic music matching is performed according to the content of lyrics.

Therefore, the embodiment of the invention provides a lyric score generation method, wherein a score server acquires a plurality of participles acquired by performing word segmentation processing on lyrics to be scored, and matches score segments corresponding to the participles according to a pre-stored correspondence between the participles and the score segments, so that the score of the lyrics to be scored can be automatically generated by using the matched score segments and the positions of the participles in the lyrics to be scored.

Based on the above idea, the embodiment of the present invention can be implemented by a soundtrack server. As shown in fig. 1, the score server includes a score library 200, where the score library 200 stores the corresponding relationship between the participles and the score segments. Taking the first text segment 100 as an example, after the score server performs word segmentation processing on the first text segment 100, word segments included in the first text segment 100, such as a first word segment, a second word segment, and a third word segment, can be obtained. From the correspondence between the participles and the score segments stored in the score library 200, the score segment set 300 corresponding to the participles can be determined. As shown in fig. 1, the first participle corresponds to a score segment set a, the second participle corresponds to a score segment set B, and the third participle corresponds to a score segment set C. Each of the score segment sets includes at least one score segment, for example, score segment set a may include score segments a1, a2, and a3, score segment set B may include score segments B1 and B2, and score segment set C may include score segments C1, C2, and C3. The score server selects a suitable score from the score set corresponding to each participle as the sub-score 400 of the first text segment, for example, the score a1-b2-c3 can be selected as the sub-score 400 of the text segment.

In the embodiment of the present invention, a piece of lyric may be understood as lyric for forming a complete song, and a piece of lyric may include at least one text segment, where a text segment may be a paragraph in a category of a word structure, or may be a sentence or a piece of text in which punctuation is used as a division. A text segment may be composed of at least one word segment. The first text segment is one text segment of the at least one text segment. The invention does not limit the concrete implementation mode of dividing words of the lyrics, and can divide words with independent word senses or word structures from the lyrics through word senses or word structures, and the divided words can be a word or a word. For example, a word segment "i love beijing tiananmen" may be divided into segments including "i", "love", "beijing", and "tiananmen".

A score segment is understood to be a score segment determined for a participle, typically a participle is in a lyric, which lyric has a matched score, and the position of the participle in the lyric may be determined first, which position may be identified by the time range on the timeline of the lyric to which the participle corresponds. And then, a part of score corresponding to the position is intercepted from the matched score according to the position, for example, a part of score of the score in the time range is intercepted, and the part of score can be a score segment corresponding to the participle.

Correspondingly, a song may include lyrics and a score corresponding to the lyrics, where a text segment in the lyrics has a sub-score corresponding to the text segment, and the sub-score is a part of the score corresponding to the text segment in the score. The sub-score corresponding to the text segment can be composed of score segments corresponding to the participles in the text segment, for example, a text segment "i love beijing tiananmen" without score can be divided into participles including "i", "ai", "beijing" and "tiananmen". The participle "me" has a corresponding set a of score segments including score segments a1, a2 and a3, the participle "ai" has a corresponding set B of score segments including score segments B1 and B2, the participle "beijing" has a corresponding set C of score segments including score segments C1, C2 and C3, the participle "Tiananmen" has a corresponding set D of score segments including score segment D1, and the subpolor of this text segment "i beijing Tiananmen" may be composed of one score segment of score segments A, B, C and D, respectively, such as a1-B2-C3-D1, or a2-B1-C3-D1, etc.

The lyrics needing to be subjected to music matching can be called as lyrics to be subjected to music matching, the lyrics to be subjected to music matching can be known through the introduction, and the lyrics to be subjected to music matching can have two levels of word segmentation and character fragment in terms of composition form or division granularity.

Next, a manner of determining sub-score corresponding to the text segment from the score segment corresponding to the word segmentation so as to obtain score of lyrics to be scored will be taken as an example, and the score generation method of lyrics provided by the embodiment of the present invention will be described in detail. Fig. 2 is a flowchart of a method for generating a score of a lyric according to an embodiment of the present invention, where the method includes:

s201: and acquiring a plurality of participles acquired by performing participle processing on the lyrics of the to-be-dubbed music.

When the song words to be matched are required to be matched, the lyrics can be manually input or input into related equipment by voice.

In the embodiment of the invention, lyrics can be dubbed through the dubbing server. The score server comprises a score library, wherein a first corresponding relation between the word segmentation and the score fragment is stored in the score library in advance, and in order to facilitate the score server to automatically score the score for the lyrics to be scored, the word segmentation is firstly needed to be carried out on the lyrics to be scored.

The specific device for performing word segmentation processing on the lyrics to be dubbed music in the embodiment of the invention is not limited, and the word segmentation processing on the lyrics to be dubbed music can be performed by the dubbing music server to obtain a plurality of words. Or performing word segmentation processing on the lyrics to be segmented by other equipment to obtain a plurality of segmented words, and acquiring the plurality of segmented words obtained after word segmentation processing by the score server from the equipment.

When the word segmentation is performed, the word segmentation can be performed on the lyrics to be dubbed music in a progressive mode, that is, the lyrics to be dubbed music can be firstly divided into character segments, and then the word segmentation is performed on the character segments. The punctuation marks may be used as the basis for dividing the text segments, or the paragraphs may be used as the basis for dividing the text segments. The word segmentation is performed on the character segments, and specifically, the word segments with independent word senses or word structures can be divided from the character segments through word senses or word structures and the like. Or the words of the lyrics to be dubbed can be directly divided, and then the word segment to which the word is divided is determined.

For example, the lyric to be dubbed music, "i love beijing tiananmen and tiananmen rise to the sun", if the punctuation mark is used as the basis for dividing the text segments, the lyric can be divided into two text segments, i.e., "i love beijing tiananmen" and "tiananmen rise to the sun". The participles with independent word senses or word structures are divided from the lyrics through word senses or word structures, wherein the word segmentation processing is carried out on the word segment 'I love Beijing Tiananmen', four participles of 'I', 'love', 'Beijing' and 'Tiananmen' can be obtained, the word segmentation processing is carried out on the word segment 'Tiananmen Shantaiyang', and four participles of 'Tiananmen', 'Shangji', 'Taiyang' and 'Sheng' can be obtained.

For large-space music song words to be matched, the progressive form is adopted for word segmentation processing, so that the obtained word segmentation is more accurate, and the error probability is reduced.

S202: and searching the score library according to the plurality of participles, and determining a score fragment set corresponding to each participle by using the first corresponding relation.

The score server may previously establish a score library, and store a correspondence between the participles and the score pieces in the score library, and the correspondence may be referred to as a first correspondence. In the score library, a participle may have at least one score segment corresponding thereto.

One to-be-matched lyric can be divided into a plurality of participles, one participle in the music library can have at least one corresponding score segment, so that when the divided participles are matched with the music library, one participle can be matched with one or more score segments, the score segment matched with one divided participle can be used as a score segment set corresponding to the participle, and the participle has a corresponding score segment set. For example, the word "i" is matched to three score segments, a1, a2 and a3, in a song, which can be taken as a set of score segments of the word "i".

Taking a first participle, that is, one participle of the plurality of participles as an example, determining the score segment set corresponding to the first participle specifically includes: and searching the score library according to the first segmentation, and determining a score segment set which comprises at least one score segment and corresponds to the first segmentation.

The method for determining the corresponding score segment set by other participles is the same as the method for determining the corresponding score segment set by the first participle, and is not repeated herein.

S203: and determining the sub-score corresponding to the at least one text segment according to the score segment set corresponding to the plurality of the scores and the score included by the at least one text segment.

The song to be matched comprises at least one text segment, and the first text segment, namely one text segment in the at least one text segment, is taken as an example for introduction. If the first text segment only comprises one participle, a score segment can be directly determined from the participle as the sub-score of the first text segment. If the first text segment includes a plurality of word segments, the method for determining the sub-score corresponding to the first text segment may be as shown in fig. 2a, and specifically includes S2031 to S2033.

S2031: and acquiring a score segment set corresponding to the participles included in the first character segment.

At least one participle included in the first text segment can be acquired through the participle processing in S201, and a score segment set corresponding to each participle included in the first text segment can be determined through S202.

S2032: and determining the fluency between the score segments selected from the score segment set corresponding to the participles included in the first text segment according to the adjacent relation of the participles included in the first text segment.

After acquiring the score segment sets respectively corresponding to at least one participle included in the first text segment, selecting a suitable score segment from the score segment set corresponding to each participle as a score segment suitable for the participle, so as to splice the selected score segments into sub scores of the first text segment.

Selecting one score segment from a score segment set corresponding to each participle, wherein the first character segment comprises a plurality of participles, the corresponding score segments can be selected, and the score segments can be used as a group of score segments, for example, the first character segment comprises three participles, namely a participle 1, a participle 2 and a participle 3, the participle 1 corresponds to a score segment set, and the score segment set comprises three score segments; the participle 2 corresponds to a score segment set which comprises a score segment; the segmentation 3 corresponds to a score segment set, the score segment set comprises two score segments, one score segment is selected from the score segment set corresponding to each segmentation and can be used as a group of score segments, and six groups of score segments can be selected for the first character segment.

Taking a group of score segments as an example, whether the group of score segments is suitable can be judged according to the fluency degree between the score segments. The smoothness degree can reflect the comfort degree between the dubbing music segments, the higher the smoothness degree is, the better the comfort degree between the dubbing music segments is, the more beautiful the generated tune is, and the better hearing enjoyment can be brought to the user.

In judging the fluency between the score segments, it refers to the fluency between a plurality of (at least two) score segments having adjacent relationships. The adjacent relationship between the score segments can be determined according to the adjacent relationship between the participles in the first text segment, for example, two adjacent participles in the first text segment, participle 1 and participle 2, so that the score segment corresponding to the participle 1 and the score segment corresponding to the participle 2 have an adjacent relationship.

The fluency can be specifically expressed by the splicing degree between the score segments.

In the embodiment of the invention, the music principle can be used as a basis for judging the splicing degree between two score segments. The concatenation degree can be understood as the concatenation degree of the score segments of two adjacent participles, and the higher the concatenation degree is, the more beautiful the melody between the two participles is. For example, a song and noise, which are the most essential differences, are that a song has an elegant melody, which may give a person an auditory enjoyment, and a noisy melody, which may be harsh to the person. Therefore, the splicing degree of the score segments in the song is one of the main factors influencing the melody of the song, the better the splicing degree of the two score segments is, the more the music spliced by the two score segments accords with the music principle, the more the melody is beautiful, and the better the auditory effect brought to people is.

In order to facilitate the intuitive understanding of the splicing degree between the score segments of the adjacent participles, the splicing score can be used for representing. The better the splicing degree of the two score segments is, the higher the corresponding splicing score is.

The degree of splicing between dubbing music pieces can be calculated according to the structure of the music, such as melody, rhythm, harmony, alignment, polyphony, melody, instrumental method, orchestration method and the like. The calculation of the splicing degree needs a plurality of factors to be considered, and the neural network has the capabilities of large-scale parallel, distributed storage and processing, self-organization, self-adaptation and self-learning, and is particularly suitable for processing the problems of inaccurate and fuzzy information processing which needs to consider a plurality of factors and conditions at the same time. In addition, an inverted index with lyrics as a main key can be established by using the score and the lyrics, and the cyclic neural network model is trained by using the inverted index so as to improve the performance of the model.

Taking any two adjacent participles in the first text segment as an example, the two participles can be respectively called as a first participle and a second participle, and the method for calculating the concatenation score between the two participles is as follows:

sa: and calculating the splicing degree between the score segment in the score segment set corresponding to the first participle in the first character segment and the score segment in the score segment set corresponding to the second participle to obtain at least one splicing score between the first participle and the second participle.

One participle can correspond to at least one score segment, the score segments of two adjacent participles can form various combinations, each combination corresponds to one splicing score, and the number of splicing scores can be obtained according to the number of combinations. For example, two adjacent participles "me" and "love" in the first text segment, i "has two pieces of dubbing music," love "has three pieces of dubbing music, and the dubbing music pieces of the two adjacent participles have six combination modes, and six splicing scores can be correspondingly obtained.

Taking a splicing score as an example, the specific calculation method of the splicing score is as follows:

According to the method, all the splicing scores corresponding to the two adjacent participles can be calculated. For example, the participles "me" and "love" are two adjacent participles, the participles "me" have corresponding score sets a including score pieces a1, a2 and a3, the participles "love" have corresponding score sets B including score pieces B1 and B2, and there are six combinations of score pieces of the two participles: a1b1, a1b2, a2b1, a2b2, a3b1 and a3b2, and the stitching scores corresponding to the six combinations are respectively as follows according to a recurrent neural network model: 3. 5, 8, 4, 7 and 2.

The first participle and the second participle are two adjacent participles randomly selected from the first character segment, the first character segment can comprise a plurality of participles, each participle has the participle adjacent to the participle, and the splicing score of other adjacent participles in the first character segment can be calculated by referring to the calculation method of the splicing score of the first participle and the second participle, so that the splicing score corresponding to all adjacent participles in the character segment is obtained.

For example, a text segment "i love beijing tiananmen" in the lyrics to be dubbed may be divided into four participles of "i", "love", "beijing", and "tiananmen", with three groups of adjacent participles: according to the method for calculating the splicing scores, at least one splicing score corresponding to the ' I ' and the ' love ', ' love ' and ' Beijing ', ' Beijing ' and ' Tiananmen ' can be calculated respectively, at least one splicing score corresponding to the ' I ' and the ' love ', ' at least one splicing score corresponding to the ' love ' and the ' Beijing ' and at least one splicing score corresponding to the ' Tiananmen '.

Sb: and calculating at least one splicing total score of the first character segment according to the splicing scores of the adjacent participles in the first character segment.

The total splicing score calculated by the splicing score can be used for representing the splicing degree of one character segment between the score segments corresponding to each participle, and the better the splicing degree, the higher the corresponding total splicing score.

The total splicing score can be calculated by adding the splicing scores, or by multiplying the splicing scores, or by a certain weight ratio. The calculation method can be selected according to the specific requirements of lyric score, and in the embodiment of the invention, the total splicing score is calculated by adopting a method of adding the splicing scores.

For example, a word segment "i love beijing tiananmen" is divided into four participles of "i", "ai", "beijing", "tiananmen", the participle "i" has a corresponding set a of score segments including score segments a1, a2 and a3, the participle "ai" has a corresponding set B of score segments including score segments B1 and B2, the participle "beijing" has a corresponding set C of score segments including score segments C1, C2 and C3, and the participle "tianann" has a corresponding set D of score segments including score segment D1. "i" and "ai" are two adjacent participles, and the dubbing music segments of the two participles are combined into six types: a1b1, a1b2, a2b1, a2b2, a3b1 and a3b2, and the corresponding splicing scores are 6, 4, 3, 7, 2 and 5 respectively; "love" and "Beijing" are two adjacent participles, and the dubbing music segments of the two participles are combined into six types: b1c1, b2c1, b1c2, b2c2, b1c3 and b2c3, and the corresponding splicing scores are 6, 2, 3, 1, 4 and 5 respectively; "Beijing" and "Tiananmen" are two adjacent participles, and the dubbing fragment combinations of the two participles are three types: c1d1, c2d1 and c3d1, and the corresponding splicing scores are 4, 2 and 5 respectively. In the text segment, the score segments corresponding to all the participles can form 18 combinations, and 18 total splicing scores can be correspondingly calculated. When a group of score segments is a1-b1-c3-d1, wherein the splicing score of a1b1 is 6 points, the splicing score of b1c3 is 4 points, and the splicing score of c3d1 is 5 points, if the total splicing score is determined in an additive mode, the total splicing score of the group of score segments is 15 points.

The method can calculate the total splicing score corresponding to all combinations of the score segments of the character segment. The higher the total score of the splicing is, the more the combination of the score segments conforms to the music principle, and the formed tune can bring better hearing effect to the audience.

S2033: and taking a group of dubbing music segments with fluency meeting preset conditions as the sub-dubbing music of the first character segment.

The preset condition may be a condition for judging the fluency level, and the preset condition may be preset, specifically, a relevant condition when the fluency level is measured according to a music principle. When the fluency between the score segments in a group of score segments meets a preset condition, the fluency is high enough to use the group of score segments as the sub-score of the first character segment.

For the case that the fluency is expressed by the splicing degree, at least one splicing total score of the first text segment may be calculated, and correspondingly, the preset condition may be a preset threshold. Specifically, a group of score segments corresponding to one total concatenation score can be selected from the total concatenation scores exceeding the preset threshold value to serve as sub-scores of the first text segment.

A text segment may include a plurality of segments, each segment may correspond to a plurality of score segments, a score segment selected from each segment may constitute a group of score segments,

one group of score segments corresponds to one total splicing score, and one character segment can have multiple groups of score segments, namely one character segment can comprise multiple total splicing scores. To complete the score of the text segment, a group of score segments needs to be selected from the group of score segments, and the selected group of score segments can be used as the sub-score corresponding to the text segment.

The sub-score of a text segment can be selected based on the total score of the splice. Since a text segment generally includes a plurality of total scores, the text segment can be selected by setting a preset threshold.

Specifically, the total splicing score exceeding a preset threshold is selected from all the total splicing scores corresponding to the text segment, then a proper total splicing score is selected from the total splicing scores exceeding the preset threshold, and a group of score segments corresponding to the total splicing score can be used as the sub-scores of the text segment. The preset threshold value is set, on one hand, the selected range can be narrowed, and on the other hand, the dubbing music pieces which do not meet the requirements can be combined and excluded.

The preset threshold value can be a fixed numerical value, and the total splicing score exceeding the preset threshold value is selected, namely the total splicing score with the score exceeding the fixed numerical value is selected. For example, the total concatenation scores of a text segment are 1, 3, 2, 9, 7, 5, 4, 8, and 6, respectively, and if the preset threshold is 5, the total concatenation scores exceeding the preset threshold are 6, 7, 8, and 9, respectively. And a group of score segments with the highest total splicing score can be selected from the four groups to splice the sub-scores of the character segment.

The timing for setting the preset threshold value may be selected in various ways, for example, the preset threshold value may be set before the lyric score is assigned, or the preset threshold value may be set before the total score is selected after the total score is calculated.

The first text segment is a text segment randomly selected from the plurality of text segments, and therefore, the generation method of the sub-score of the other text segments is the same as that of the first text segment, and is not described herein again.

S204: and generating the score of the lyrics to be scored by splicing the sub scores respectively corresponding to the at least one character fragment.

The sub-score corresponding to all the text segments included in the lyric to be dubbed can be calculated by S203, and each text segment has a fixed position in the lyric, so that the sub-scores corresponding to all the text segments in the lyric can be spliced according to the positions of the text segments, and a complete dubbing, namely the dubbing of the lyric to be dubbed, can be generated.

The method for generating the score of the lyric introduces the score of the lyric in detail, and can match the score of the lyric to be matched with the score corresponding to the participle after the lyric to be matched is participled because the score corresponding to the participle is prestored in the score library, so that the score of the lyric to be matched can be automatically generated by utilizing the matched score and the position of the participle in the lyric to be matched with the lyric, and the efficiency of matching the lyric to the lyric is effectively improved.

Next, a detailed description will be given of the process of creating the music library, where the basis for creating the music library may be history music, or may be music possibility that a word segmentation may have been set in advance through a music principle. Alternatively, the present invention will explain a manner of creating a music library by history music.

As shown in fig. 3, a schematic flow chart of a building process of a score library provided in an embodiment of the present invention is shown, where the building process includes:

s301: and acquiring historical songs, wherein the historical songs comprise historical lyrics, historical score and song types.

The historical songs may be songs that already exist today that include complete lyrics and soundtracks. The historical songs can be stored in the cloud or the server, and can be obtained from the cloud or the server when the historical songs need to be used. To facilitate subsequent processing of the historical songs, the historical songs may include historical lyrics, historical score, and song type.

The song type may be divided according to the melody of the song, the content of the lyrics, the rhythm of the score, etc. For example, the types of songs can be roughly divided into ten categories according to the development history of the songs: tempo complaints are music (R & B), RAP (RAP), rock, jazz, country, new century, classical pop, folk song, disco, english rock (Trip-Hop). Or the song types which can be divided according to the melody of the song comprise lyrics, feelings of injury and the like.

S302: and performing word segmentation on the historical lyrics.

The method for segmenting the lyrics may be the same as the method for obtaining a plurality of segmentations in S201, and is not described herein again.

S303: and determining the corresponding score segment in the historical score according to the obtained word segmentation.

Specifically, the position of the word segmentation in the lyric may be determined, and the position may be identified by a time range corresponding to the word segmentation on a time line of the lyric, and then a part of score corresponding to the position is intercepted from the score of the lyric according to the position, and is used as a score segment corresponding to the word segmentation.

For example, in a song, the word "Tiananmen" is positioned at the 50 th second of the song playing according to the time sequence, the time range corresponding to the word can be 45 seconds to 55 seconds, and the score corresponding to 45 seconds to 55 seconds can be intercepted in the score as the score segment of the word.

A participle may appear multiple times in a song, and each occurrence may intercept a corresponding score segment, so that a participle may correspond to multiple score segments in a song. In addition, the same segmentation may occur in different songs.

It can be seen that the score segment corresponding to each word segmentation may be from not only one song but also other songs. Therefore, the more history songs are acquired, the more segmented words are obtained, and the wider the score segments corresponding to the segmented words are, so that the segmented words and the corresponding score segments are more perfect.

S304: and establishing the music matching library according to the first corresponding relation between the obtained word segmentation and the music matching segment in the historical music matching and the second corresponding relation between the music matching segment in the historical music matching and the song type.

In S303, the score segment corresponding to the segmented word may be obtained, and the correspondence between the obtained segmented word and the score segment in the historical score may be referred to as a first correspondence.

The score segments are obtained by dividing from the historical songs, so each score segment has the historical song to which the score segment belongs, the corresponding relation between the score segments obtained by dividing from the historical songs and the song types of the historical songs can be determined according to the song types included in one historical song, the corresponding relation is different from the first corresponding relation, and the corresponding relation between the score segments and the song types in the historical scores can be called as a second corresponding relation.

And establishing a music matching library according to the first corresponding relation between the obtained word segmentation and the music score in the historical music matching and the second corresponding relation between the music score in the historical music matching and the song type. After the establishment of the score library is completed, the score library can be stored in a score server.

In the process of building the score library, in order to make the score library more complete, a large number of historical songs are often analyzed, so that score segments corresponding to each participle are as many as possible, but when the lyrics are scored, the more score segments are, the greater the calculation difficulty of selecting the score segment suitable for each participle is. Therefore, when the proper score segments are selected for word segmentation, the score segments can be primarily screened, and then the proper score segments are selected from the screened score segments.

Considering that the lyrics are often closely related to the types of songs, the preliminary screening of the score segments can be performed according to the degree of association between the participles included in the lyrics and the types of songs.

Specifically, the type of the song to which the lyric to be matched belongs may be determined, and then, according to the second correspondence stored in the score library, a score segment corresponding to the type of the song to which the lyric to be matched belongs may be screened out from the score library.

After the preliminary screening, determining a score segment set corresponding to the multiple participles from score segments corresponding to the song types to which the to-be-scored song words belong according to the first corresponding relation stored in the score library.

The song type of the lyrics to be dubbed may be preset, may be determined according to the content of the lyrics to be dubbed, or may be determined according to a type library established in the dubbing server, and the type library may include possibility information that the participles belong to different song types. Specifically, the type library can be searched according to a plurality of participles included in the lyric to be matched, the song type to which the lyric to be matched belongs is determined, that is, the song type to which the lyric to be matched belongs is determined according to the possibility information in the type library.

Whether the determined song type is accurate or not directly influences the quality of subsequently generated music score is determined, and the more accurate the song type is, the better the subsequently generated music score is, so that the finally generated song can bring better auditory effect to the user. The naive Bayes model is one of the most widely used classification models at present, originates from classical mathematical theory, has solid mathematical foundation and stable classification efficiency. Therefore, in the embodiment of the invention, the type library can specifically determine the song type to which the lyric to be matched belongs through a naive Bayes model. In addition, a naive Bayes model can be trained through the word segmentation and the song type corresponding to the word segmentation, so that the accuracy of judging the song type to which the word segmentation belongs is improved.

If the situation that the song word to be dubbed belongs to one song type is determined, the dubbing can be generated for the dubbing word to be dubbed through the dubbing fragment of the song type. The score fragment belonging to the song type can better embody the characteristics of the song type, so the generated score can better accord with the style which the lyrics of the score to be distributed want to embody. Furthermore, the score segments for scoring the song words to be scored can be effectively reduced by screening the score segments according to the types of the songs, so that the calculation difficulty of automatic scoring is reduced.

The song type may be determined based on a type library prior to the preliminary screening.

Next, a detailed description will be given of a type library establishing process, as shown in fig. 4, which is a schematic flow chart of the type library establishing process provided in the embodiment of the present invention, and the establishing process includes:

s401: and acquiring historical songs, wherein the historical songs comprise historical lyrics, historical score and song types.

S402: and performing word segmentation on the historical lyrics.

S401-S402 are the same as S301-S302 in FIG. 3, and are not described herein again.

S403: counting the occurrence frequency of a third participle in the word segmentation process and the corresponding relation between the third participle and the song type of the historical song to which the third participle belongs, so as to determine the occurrence frequency of the third participle in the same song type, wherein the third participle is a participle obtained from the historical lyrics.

The word segmentation is carried out on the historical songs, and the word segmentation processing is carried out on the historical lyrics by taking each song as a unit, so that the times of the occurrence of the word segmentation in each song can be obtained. Taking the third participle as an example, after the historical lyrics are participled, the frequency of the third participle appearing in the historical song can be counted, and the corresponding relation between the third participle and the song type can be counted according to the song type included in the historical song, so that the frequency of the third participle appearing in the same song type can be determined. For example, the third participle appears 10 times in the first history song, 20 times in the second history song, 5 times in the third history song, and 15 times in the fourth history song, each song having its corresponding song type, the first history song and the second history song both belonging to the RAP type, the third history song belonging to the rock type, and the fourth song belonging to the jazz type, so that it can be found that the third participle appears 30 times in the RAP type, 5 times in the rock type, and 15 times in the jazz type.

It should be noted that the third participle may be a participle arbitrarily selected from a plurality of participles included in the historical lyrics, where the third participle is distinguished from the first participle and the second participle mentioned above for name, and is not limited in other meanings such as sequence.

S404: and determining the possibility information that the third participle belongs to different song types according to the occurrence frequency of the third participle in the participle process and the occurrence frequency of the third participle in the same song type.

Each historical song may include historical lyrics and a song type, each song may include a participle corresponding to the song type of the song, and the same participle may be from different historical songs, i.e., the same participle may correspond to different song types.

The likelihood information may be information indicating the degree of association of the participle with the song type, and may be stored in the form of a numerical value, where the numerical value may be a percentage.

For example, a genre library is created based on 15 history songs, 10 history songs in the 15 songs belong to the RAP type, the remaining 5 songs belong to the rock type, and the participle "i" appears 100 times in the 15 songs, wherein 70 times appears in the RAP type history songs and 30 times appears in the rock type history songs, so that it can be found that the possibility information that the participle "i" belongs to the RAP type is 70%, and the possibility information that the participle "i" belongs to the rock type is 30%.

S405: and establishing the type library according to the possibility information that the participles obtained from the historical lyrics belong to different song types.

The third participle is an optional participle in all the participles included in the historical song, the processing process of the other participles is the same as that of the third participle, and the possibility information that all the participles in the historical song belong to different song types can be determined through S403 and S404, so that a type library for storing the possibility information that the participles obtained from the historical lyrics belong to different song types can be established. After the type library is established, the type library can be stored in the music server.

The establishment of the type library enables the score server to directly determine the song type of the song word to be scored according to the possibility information that the participles stored in the type library belong to different song types before the preliminary screening, simplifies the step of determining the song type, and further improves the efficiency of automatic score scoring.

The lyrics to be dubbed described in the above embodiments may have two levels of word segmentation and text segment in terms of composition form or granularity of division, but the method for generating dubbing music provided in the embodiments of the present invention may be applied to the lyrics to be dubbed or the granularity of division of the lyrics to be dubbed, and may also be applied to lyrics to be dubbed having other composition forms or to the situation of other granularity of division of the lyrics to be dubbed. Next, how the embodiments of the present invention are applied to the candidate songs in possible composition forms or possible granularity of division will be described.

Fig. 5 is a flowchart of a method for generating a lyric score of a musical instrument according to an embodiment of the present invention, which is applied to a musical instrument server, where the musical instrument server includes a musical instrument library, and the musical instrument library includes a first correspondence between segments and musical instrument segments, where any one segment has a musical instrument segment set including at least one musical instrument segment, and here, description about relevant features of the musical instrument server may refer to fig. 1, and description about relevant features in an embodiment corresponding to fig. 2 is omitted here for brevity.

The method comprises the following steps:

s501: and acquiring at least one fragment obtained by performing word segmentation processing on the lyrics of the score.

According to the division granularity of word segmentation processing or the composition form of lyrics to be dubbed, at least one fragment can be obtained through the lyrics to be dubbed, wherein the fragment is related to the division granularity and can also be related to the composition form. For example, when the division granularity is fine, the segment may be in the form of a word segmentation, and when the division granularity is coarse, the segment may be in the form of a text segment. For example, when there are few punctuations in the lyrics to be matched and there is no clear paragraph relationship, the segment may be in the form of text segment, and when there are many punctuations in the lyrics to be matched and the paragraph division is clear, the segment may be in the form of word segmentation. Specifically, the form of the segment into which the lyrics to be matched are divided can be adjusted according to the scene requirements, the calculation accuracy and the like, which is not limited in the present invention.

It should be noted that what form of the section into which the lyrics to be dubbed are divided is also the same as the form of the section stored in the dubbing library in the dubbing server, and thus the set of dubbing sections can be matched by the divided section and the first correspondence. For example, the segment stored in the music library is in the form of a participle, and the song word to be matched needs to be subjected to participle processing to obtain at least one participle. If the fragments stored in the score library include both the form of word segmentation and the form of text fragment, the fragments obtained by segmenting the lyrics of the score to be assigned may be related to the above-mentioned granularity or lyric composition.

S502: and searching the music matching library according to the at least one segment, and determining a music matching segment set corresponding to the at least one segment respectively according to the first corresponding relation.

Since the divided segments are the same as the segments stored in the score library, the determination of the score segment set corresponding to each divided segment can be realized through the first corresponding relationship.

S503: and according to the lyrics to be dubbed music, determining a dubbing music fragment corresponding to each fragment in the at least one fragment from the dubbing music fragment set corresponding to the at least one fragment respectively.

S504: and generating the score of the lyrics to be scored by splicing the determined score segments.

Because the set of the score segments comprises at least one score segment, the score segments corresponding to each segment can be determined from the set of the score segments according to the composition relationship of each segment in the lyrics to be scored, and the specific determination mode and the splicing mode can be different according to the different specific forms of the segments.

It can be seen that, by presetting a score library including a first corresponding relationship between the participles and score segments in a score server, after the obtained song words to be scored are participled to obtain at least one segment, the at least one segment can be used to determine a score segment set corresponding to the at least one segment according to the first corresponding relationship from the score library, and according to the lyrics to be scored, the score corresponding to the song words to be scored is spliced by determining the score segments from the score segment set.

Embodiments of the invention will be further illustrated by the possible forms of the segments.

In a first possible form, a segment may be a word segmentation:

in this case, since the word segment itself is approximately the length of one word or one phrase, the possibility that a single word segment is used as the lyric to be dubbed is not high, and generally at least a plurality of word segments are required to form the lyric to be dubbed. Therefore, under the condition that the fragments are word segmentation, a plurality of word segmentation obtained by performing word segmentation processing on the lyrics to be dubbed music can be obtained.

During the lyric matching process of the lyrics to be matched, determined score segments corresponding to the participles can be directly used for matching according to the positions of the participles in the lyrics to be matched so as to match the score corresponding to the lyrics to be matched; and under the condition that the song words to be matched also have the hierarchy of the character segments, obtaining the sub-scores of the character segments, and splicing the scores corresponding to the song words to be matched by the sub-scores according to the positions of the character segments in the song words to be matched.

For the case that the lyric to be matched also has the hierarchy of the text segments, the lyric to be matched comprises at least one text segment, and one text segment in the at least one text segment comprises at least one participle.

When determining the score segments corresponding to the participles, determining the sub-scores corresponding to the at least one text segment according to the score segment set corresponding to the participles and the participles included in the at least one text segment, wherein the sub-scores corresponding to the text segment are obtained from the score segments corresponding to the participles included in the text segment. The specific determination manner may refer to the related description in the embodiment corresponding to fig. 2, and is not described herein again.

After obtaining the sub-score corresponding to the text segment, the score of the lyric to be scored can be generated by splicing the sub-scores respectively corresponding to the at least one text segment.

The specific splicing manner may refer to the related description in the embodiment corresponding to fig. 2, and is not described herein again.

In a second possible form, the segment may be a text segment:

because the text segment may include contents of a certain length, it is possible that one text segment may be used as lyrics of the to-be-dubbed music, and in this case, when the set of the to-be-dubbed music segments corresponding to the text segment is found, the to-be-dubbed music segment determined in the set of the to-be-dubbed music segments may be directly used as the to-be-determined dubbed music.

When the lyrics to be dubbed are divided into a plurality of character segments, the corresponding dubbing music segments can be determined from the dubbing music segment set corresponding to each character segment, and the dubbing music corresponding to the lyrics to be dubbed can be determined by splicing according to the positions of the character segments in the lyrics to be dubbed.

The method for obtaining lyrics music according to the embodiment of the present invention is described in detail below from an interactive end, where the interactive end may be an intelligent device related to lyrics music, and may implement interaction with a music server and interaction with a user. Fig. 6 is a flowchart of a method for acquiring score of lyrics according to an embodiment of the present invention, where the method includes:

s601: the method comprises the steps of sending the acquired song words to be matched to a music server, wherein the song words to be matched comprise at least one text segment, one text segment in the at least one text segment comprises at least one participle, the music server comprises a music library, the music library comprises a first corresponding relation of the participle and the music segment, and any participle has a music segment set comprising at least one music segment.

The interactive terminal can acquire the lyrics to be dubbed music, the form of acquiring the lyrics to be dubbed music is not limited in the embodiment of the invention, and the lyrics to be dubbed music can be manually input or input into the interactive terminal by voice.

After the interactive terminal obtains the lyrics of the music to be matched, the lyrics of the music to be matched can be directly sent to the music matching server. Or sending the lyrics to be matched to the music server when receiving a request sent by the music server for obtaining the lyrics to be matched.

S602: and acquiring the score corresponding to the lyrics to be scored from the score server, wherein the score is obtained by splicing score segments in a score segment set corresponding to the participles in the lyrics to be scored according to the score segment server, and the score segment set corresponding to the participles in the lyrics to be scored is obtained by searching the score library according to the first corresponding relation by the score server.

For the description of the features in the embodiment corresponding to fig. 6, reference may be made to fig. 1 and the related description of the embodiment corresponding to fig. 2, which are not repeated here.

In the embodiment of the present invention, the interactive end may implement the related processing of the score, for example, presenting the score to the user. Therefore, after the music server determines the music corresponding to the song word to be matched, the music can be directly sent to the interactive terminal, or the music corresponding to the song word to be matched is sent to the interactive terminal when a request for obtaining the music sent by the interactive terminal is received.

The interactive end can take the acquired score as pending score and select the pending score meeting the requirement from the multiple pending scores as the score of the song word to be matched. Optionally, a plurality of pending scores corresponding to the lyrics of the to-be-dubbed music may be obtained from the dubbing server, where the pending scores carry the dubbing information; and selecting the undetermined score which meets the requirement as the score of the lyrics of the to-be-scored according to the score information of the to-be-scored scores.

The score information can be information used for identifying pending scores, each pending score has corresponding score information, and the score information can embody the relevant characteristics of the pending score to a certain extent. Specifically, the score and/or score type of the pending score carrying the score information may be included in the score information.

The score may be used to indicate the degree to which the pending score conforms to the musical principle, and may be derived based on the degree of fluency between score segments in the pending score. The higher the score is, the more the pending score accords with the music principle, namely the melody of the pending score is more pleasant, so that better hearing enjoyment can be brought to the user.

The score type may be used to indicate the type of song to which the pending score belongs, e.g. the pending score belongs to rock or jazz, etc.

When the pending score meeting the requirement is selected as the score of the lyrics of the to-be-dubbed music, a plurality of modes can be provided, the first mode can be selected according to the score included in the music information, the highest score of the to-be-dubbed music can be used as the score of the lyrics of the to-be-dubbed music, or the higher score of the to-be-dubbed music can be used as the score of the lyrics of the to-be-dubbed music, or a preset numerical value can be set, and the pending score higher than the preset numerical value can be used as the score of the lyrics of the to-be-dubbed music.

The second way can be selected according to the type of the score included in the score information, for example, the user can select a favorite type of the pending score according to the types of the scores to be determined, and use the pending score as the score of the lyrics of the score to be determined.

The third mode may be to select the score based on the score and the type of the score included in the score information, for example, when there are a plurality of pending scores of the same score, the score with the highest score in the pending scores may be used as the score of the lyrics of the to-be-dubbed music, or when there are a plurality of pending scores with the same score, the user may select a favorite type of the to-be-dubbed music as the score of the lyrics of the to-be-dubbed music according to the to-be-dubbed music types.

In consideration of the fact that the score corresponding to the to-be-matched song word determined by the score server does not meet the requirement, for example, the user wants to obtain the score of the rock type, but the score of the rock type is not included in the score type of the score determined by the score server. In this case, it is described that the score determined by the score server has a problem, and the score server needs to determine the score again.

Therefore, for the situation that the obtained score does not meet the requirement, the interactive end can send feedback information to the score server, so that the server regenerates the score of the lyrics of the to-be-scored according to the feedback information.

The feedback information may be information indicating that the score is not in demand. In order to make the regenerated score better meet the requirement, the feedback information may include information describing the requirement, for example, if the user wants to obtain a rock-type score, the requirement is a rock-type score, and the feedback information may carry the information.

The music score corresponding to the lyrics to be matched determined by the music score server can better meet the requirements of the user through the feedback operation, and the accuracy of music score matching for the lyrics to be matched is further improved.

Fig. 7 is a device structure diagram of an apparatus for generating a score of lyrics according to an embodiment of the present invention, which is applied to a score server including a score library, where the score library includes a first correspondence between participles and score segments, where any participle has a score segment set including at least one score segment, and the apparatus includes an obtaining unit 701, a searching unit 702, a determining unit 703, and a generating unit 704:

the acquiring unit 701 is configured to acquire a plurality of participles acquired by performing a participle processing on a lyric to be dubbed, where the lyric to be dubbed includes at least one text segment, and one text segment in the at least one text segment includes at least one participle;

the searching unit 702 is configured to search the score library according to the multiple participles, and determine a score segment set corresponding to each participle by using the first corresponding relationship;

the determining unit 703 is configured to determine, according to the score segment sets corresponding to the multiple word segments and the word segments included in the at least one text segment, sub-scores corresponding to the at least one text segment respectively;

the generating unit 704 is configured to generate the score of the lyric to be scored by splicing the sub-scores respectively corresponding to the at least one text segment.

the searching unit comprises a screening subunit and a determining subunit:

the word segmentation unit is used for segmenting the historical lyrics;

For the above description of the relevant features of the score server, reference may be made to fig. 1, and the relevant description in the embodiment corresponding to fig. 2 is not repeated here.

It can be seen that, a score library including a first corresponding relationship between the participles and score segments is preset in the score server, after a plurality of participles obtained by performing the participle processing on the lyrics to be scored are obtained, the score segment sets corresponding to the participles are determined from the score library according to the first corresponding relationship by using the participles, then the score segments of the lyrics to be scored are taken as units, the score segment set corresponding to the participles in one text segment is used for combining the sub-scores corresponding to one text segment, and the sub-scores of the text segments are spliced into the score of the lyrics to be scored, therefore, the score segments corresponding to the participles are pre-stored, the score segments corresponding to the participles can be matched after the lyrics to be scored, and the score of the lyrics to be scored can be automatically generated by using the matched score segments and the positions of the participles in the lyrics to be scored, the efficiency of the lyric music is effectively improved.

Fig. 8 is a device structure diagram of an apparatus for generating a score of lyrics according to an embodiment of the present invention, which is applied to a score server, where the score server includes a score library, and the score library includes a first corresponding relationship between segments and score segments, where any one segment has a score segment set including at least one score segment, and the apparatus includes an obtaining unit 801, a searching unit 802, a determining unit 803, and a splicing unit 804:

the acquiring unit 801 is configured to acquire at least one segment obtained by performing word segmentation processing on lyrics of a to-be-dubbed music;

the searching unit 802 is configured to search the music library according to the at least one segment, and determine, according to the first corresponding relationship, a music segment set corresponding to the at least one segment respectively;

the determining unit 803 is configured to determine, according to the lyrics to be dubbed, a dubbing fragment corresponding to each fragment in the at least one fragment from a set of dubbing fragments corresponding to the at least one fragment;

the splicing unit 804 is configured to generate the score of the lyric to be scored by splicing the determined score segments.

Fig. 9 is a device structure diagram of an apparatus for acquiring a score of a lyric according to an embodiment of the present invention, which is applied to an interactive end, and the apparatus includes a sending unit 901 and an acquiring unit 902:

the sending unit 901 is configured to send the obtained to-be-matched lyrics to a music server, where the to-be-matched lyrics include at least one text fragment, one text fragment of the at least one text fragment includes at least one participle, the music server includes a music library, and the music library includes a first correspondence between the participle and the music fragment, where any participle has a music fragment set including at least one music fragment;

the obtaining unit 902 is configured to obtain, from the music server, the music corresponding to the lyrics to be matched, where the music is obtained by splicing the music pieces in the music piece set corresponding to the participles in the lyrics to be matched by the music server, and the music piece set corresponding to the participles in the lyrics to be matched is obtained by searching the music library by the music server according to the first corresponding relationship.

Optionally, the apparatus further includes a feedback unit:

Therefore, the interaction with the music matching server and the interaction with the user can be realized by the interaction end, the music matching corresponding to the lyrics to be matched, which are determined by the music matching server, can better meet the requirements of the user through the feedback of the interaction end, and the accuracy of matching the music for the lyrics to be matched is further improved.

Fig. 10 is a system configuration diagram of a lyric dubbing system according to an embodiment of the present invention, where the dubbing system includes a dubbing server 1001 and an interactive terminal 1002:

the score server 1001 includes a score library including a first correspondence relationship between a participle and score segments, where any participle has a score segment set including at least one score segment, the score server 1001 is configured to obtain a plurality of participles obtained by performing a participle process on a to-be-scored lyric, where the to-be-scored lyric includes at least one text segment, and one text segment of the at least one text segment includes at least one participle; searching the score library according to the multiple participles, and determining a score fragment set corresponding to each participle by using the first corresponding relation; determining sub-scores corresponding to the at least one text segment according to the score segment sets corresponding to the multiple scores and the scores included by the at least one text segment; generating the score of the lyrics to be scored by splicing the sub scores respectively corresponding to the at least one text segment;

the interactive terminal 1002 is configured to send the acquired lyrics to be dubbed music to the dubbing server; and acquiring the score corresponding to the lyrics of the to-be-assigned score from the score server.

Next, a music editor related to a lyric score generation method provided by the embodiment of the present invention is introduced, where the music editor may obtain lyrics of a to-be-assigned music and may also edit a score corresponding to the lyrics of the to-be-assigned music provided by a score server.

The music editor may include a display having an input interface for obtaining lyrics of an input to be dubbed music, and an editing interface for presenting the dubbing music.

The input interface is used for acquiring lyrics to be matched, and the acquired lyrics to be matched can be the lyrics input in the input interface or the lyrics acquired after the input voice is recognized.

That is, the user may manually input lyrics to be assigned to music in the input interface or import lyrics to be assigned to music from other text editing tools or texts through the input interface while using the music editor. When the intelligent equipment provided with the music editor has a voice input recognition function, a user can play an audio file with the lyrics to be matched or directly speak the lyrics to be matched, the intelligent equipment can recognize the input audio content and recognize the lyrics to be matched, and the input interface can acquire the recognized lyrics to be matched. The abundant mode of obtaining the lyrics of the music to be matched improves the application range of the music editor.

After the input interface acquires the lyrics to be dubbed music, the music editor can be connected with the dubbing server, so that the lyrics to be dubbed music can be sent to the dubbing server when needed. The user can generate an output instruction through a function key arranged on the music editor to instruct the music editor to send lyrics of the to-be-dubbed music to the dubbing server.

When the output instruction is obtained, the music editor is used for sending the music words to be matched, which are obtained through the input interface, to a music server, and the music server comprises a music library, wherein the music library comprises first corresponding relations of participles and music sections, and any participle is provided with a music section set comprising at least one music section.

For the description of the lyrics to be dubbed music, the dubbing music and the related features of the dubbing music server, reference may be made to fig. 1 and the related description in the embodiment corresponding to fig. 2, which are not described herein again.

The editing interface can enable a user to observe specific information of the score, such as notes, beats, a time axis and the like, by displaying the score, and can be used as a basis for editing the score. It is noted that the editing interface may present other soundtracks, such as soundtracks identified from recorded audio data, in addition to the soundtrack sent by the soundtrack server.

In order to enable editing of the score, optionally, the editing interface has an editing button, and when a trigger to the editing button is received, the editing interface is configured to set the displayed score as editable. In the editable state, the score can be deleted, added, modified and the like to finally meet the requirements of the user.

Therefore, lyrics of the music to be matched can be obtained through the input interface through the music editor, the lyrics of the music to be matched can be output to the music server, and when the music to be matched which is returned by the music server and aims at the lyrics of the music to be matched is received, the music to be matched can be edited by displaying the music in the editing interface, so that the user experience is improved.

Next, a music player related to the lyric score generation method provided by the embodiment of the present invention is introduced, and the music player can obtain the lyrics of the lyric to be scored and the corresponding score, and mix the lyrics and the corresponding score into a song for playing.

The music player has a mixing interface and a playing interface.

The mixing interface is used for mixing the lyrics to be matched with the score corresponding to the lyrics to be matched with the score to obtain the song, the score corresponding to the lyrics to be matched with the score is obtained by splicing score segments in a score segment set corresponding to the participles in the lyrics to be matched with the score server according to a first corresponding relation, the score segment set corresponding to the participles in the lyrics to be matched with the score library is obtained by searching the score library according to the first corresponding relation by the score server, the score library comprises a score library, the score library comprises the first corresponding relation of the participles and the score segments, and any participle has a score segment set comprising at least one score segment.

The playing interface is used for playing the songs.

It can be seen that the mixing interface can mix the pre-obtained lyrics to be dubbed with the corresponding dubbing music, the mixing mode is not limited in the invention, the lyrics obtained after mixing are the lyrics to be dubbed, the music is the song of the dubbing music, when the song is played, the dubbing music can be played only, and the lyrics corresponding to the time axis of the played dubbing music (the lyrics at the position corresponding to the time axis in the lyrics to be dubbed music) are displayed at the relevant position, and the simulated sound of the lyrics to be dubbed music can also be output simultaneously when the dubbing music is played through the simulated sound. Therefore, the user can see or hear whether the corresponding lyrics are matched with the score while playing the score, and the user experience is improved.

Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium may be at least one of the following media: various media that can store program codes, such as read-only memory (ROM), RAM, magnetic disk, or optical disk.

It should be noted that, in the present specification, all the embodiments are described in a progressive manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus and system embodiments, since they are substantially similar to the method embodiments, they are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described embodiments of the apparatus and system are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A lyric score generation method applied to a score server including a score library including first correspondences of participles and score pieces, wherein any participle has a score piece set including at least one score piece, the method comprising:

and generating the score of the lyrics to be matched with the music by splicing the sub-scores respectively corresponding to the at least one character fragment according to the position of each character fragment in the lyrics to be matched with the music.

2. The method of claim 1, wherein the first text segment is one of the at least one text segment, and the determining the respective sub-score of the at least one text segment according to the score segment set corresponding to the plurality of scores and the score included in the at least one text segment comprises:

3. The method of claim 2, wherein the first word segmentation and the second word segmentation are adjacent word segmentation in the first text segment, and the determining the fluency between score segments selected from a score segment set corresponding to word segmentation included in the first text segment according to the adjacent relation of the word segmentation included in the first text segment comprises:

4. The method of claim 3, wherein the calculating the degree of concatenation between the score segment in the score segment set corresponding to the first participle in the first text segment and the score segment in the score segment set corresponding to the second participle to obtain at least one concatenation score between the first participle and the second participle comprises:

5. The method according to any one of claims 1 to 4, wherein the score server further comprises a genre library, the genre library comprises possibility information that the participles belong to different song types, and the score library further comprises a second correspondence between score pieces and song types; before searching the score library according to the multiple participles and determining a score fragment set corresponding to the multiple participles, the method further comprises the following steps:

6. The method of claim 5, further comprising:

performing word segmentation on the historical lyrics;

7. The method of claim 5, further comprising:

performing word segmentation on the historical lyrics;

8. The method of claim 1, wherein the obtaining a plurality of participles obtained by participling lyrics of a to-be-dubbed music comprises:

9. An apparatus for generating a score of a lyric, applied to a score server including a score library including a first correspondence relationship of participles and score pieces, wherein any one of the participles has a score piece set including at least one score piece, the apparatus comprising an acquiring unit, a searching unit, a determining unit, and a generating unit:

the generating unit is used for generating the score of the lyric to be dubbed music by splicing the sub-scores respectively corresponding to the at least one character segment according to the position of each character segment in the lyric to be dubbed music.

10. The apparatus according to claim 9, wherein a first text segment is one of the at least one text segment, and the determining unit is specifically configured to obtain a score segment set corresponding to a participle included in the first text segment; determining the fluency between score segments selected from a score segment set corresponding to the participles included in the first text segment according to the adjacent relation of the participles included in the first text segment; and taking a group of dubbing music segments with fluency meeting preset conditions as the sub-dubbing music of the first character segment.

11. The apparatus of claim 10, wherein the first and second segments are adjacent segments in the first text segment, and wherein the determining unit comprises a calculating subunit, a obtaining subunit, and as subunits:

12. The apparatus according to claim 11, wherein the computing subunit is specifically configured to obtain one score segment in the score segment set corresponding to the first participle, and obtain one score segment in the score segment set corresponding to the second participle; and calculating the splicing degree between the two score segments to obtain a splicing score determined by the two score segments between the first participle and the second participle.

13. The apparatus according to any one of claims 9 to 12, wherein the score server further comprises a genre library, wherein the genre library comprises possibility information that the participles belong to different song genres, and the score library further comprises a second correspondence between score pieces and the song genres; the apparatus further comprises a determine song type unit:

the searching unit comprises a screening subunit and a determining subunit:

14. The apparatus according to claim 13, further comprising a history song acquisition unit, a word segmentation unit, a score piece determination unit, and a creation unit:

the word segmentation unit is used for segmenting the historical lyrics;

15. A lyric score generation method applied to a score server including a score library including a first correspondence of segments and score segments, any one of which has a score segment set including at least one score segment, the method comprising:

and generating the score of the lyrics to be scored by splicing the determined score segments according to the position of each segment in the lyrics to be scored in the at least one segment.

16. The method of claim 15, wherein the segment is a segmentation word, and the obtaining at least one segment obtained by performing a segmentation word processing on the lyrics of the score music comprises:

17. The method of claim 15, wherein the segment is a text segment, and the song to be dubbed comprises at least one text segment.

18. The method of claim 16, wherein the lyrics to be dubbed comprises at least one text fragment, and one text fragment of the at least one text fragment comprises at least one participle, and the determining, according to the lyrics to be dubbed, a respective dubbed fragment of the at least one participle from a set of respective dubbed fragments corresponding to the at least one participle comprises:

the generating of the score of the lyrics to be score by splicing the determined score segments according to the position of each segment in the lyrics to be score comprises the following steps:

19. An apparatus for generating a score of a lyric, applied to a score server including a score library including a first correspondence relationship between segments and score segments, any one of the segments having a score segment set including at least one score segment, the apparatus comprising an obtaining unit, a searching unit, a determining unit, and a splicing unit:

and the splicing unit is used for splicing the determined score fragment according to the position of each of the at least one fragment in the lyrics to be scored to generate the score of the lyrics to be scored.

20. The apparatus according to claim 19, wherein the segment is a participle, the lyric to be dubbed music comprises at least one text segment, one of the at least one text segment comprises at least one participle, and the determining unit is specifically configured to determine the sub-dubbing music corresponding to the at least one text segment according to a set of dubbing music segments corresponding to a plurality of participles and the participle included in the at least one text segment, wherein the sub-dubbing music corresponding to the text segment is obtained from the dubbing music segment corresponding to the participle included in the one text segment;

the splicing unit is specifically configured to splice sub-scores respectively corresponding to the at least one text segment according to the position of each text segment in the lyrics to be dubbed music to generate the dubbing music of the lyrics to be dubbed music.

21. A lyric score obtaining method is applied to an interactive end, and the method comprises the following steps:

and acquiring the score corresponding to the lyrics to be scored from the score server, wherein the score is obtained by splicing score segments in a score segment set corresponding to the participles in the lyrics to be scored according to the position of each participle in the at least one participle in the lyrics to be scored in the score server, and the score segment set corresponding to the participles in the lyrics to be scored is obtained by searching the score library according to the first corresponding relation by the score server.

22. The method of claim 21, wherein the obtaining the score corresponding to the lyric of the to-be-scored from the score server comprises:

23. The method of claim 22, wherein the score information includes a score and/or score type of the pending score that carries the score information.

24. The method of claim 21, wherein after the obtaining the score corresponding to the lyric of the to-be-scored from the score server, further comprising:

25. The method of claim 24, wherein the feedback information includes information describing the need.

26. The device for acquiring the score of the lyrics is applied to an interactive end, and comprises a sending unit and an acquisition unit:

the obtaining unit is configured to obtain, from the music server, the music score corresponding to the lyrics to be matched, where the music score is obtained by splicing, by the music server, music score segments in a music score set corresponding to the participles in the lyrics to be matched according to the position of each participle in the at least one participle in the lyrics to be matched, and the music score segment set corresponding to the participle in the lyrics to be matched is obtained by the music server searching the music library according to the first correspondence.

27. The apparatus of claim 26, wherein the obtaining unit comprises an obtaining subunit and a selecting subunit:

28. The lyric dubbing system is characterized by comprising a dubbing server and an interactive terminal:

the music score server comprises a music score library, wherein the music score library comprises a first corresponding relation of participles and music scores, any participle is provided with a music score set comprising at least one music score, the music score server is used for acquiring a plurality of participles acquired by performing word segmentation processing on lyrics to be matched, the lyrics to be matched comprise at least one text score, and one text score in the at least one text score comprises at least one participle; searching the score library according to the multiple participles, and determining a score fragment set corresponding to each participle by using the first corresponding relation; determining sub-scores corresponding to the at least one text segment according to the score segment sets corresponding to the multiple scores and the scores included by the at least one text segment; generating the score of the lyrics to be score by splicing the sub scores respectively corresponding to at least one character fragment according to the position of each character fragment in the lyrics to be score;

29. The music editor is characterized in that the music editor is provided with an input interface used for acquiring lyrics of input music to be distributed and an editing interface used for displaying the music;

when the score corresponding to the lyrics to be dubbed is obtained from the dubbing server, the music editor is used for displaying the score corresponding to the lyrics to be dubbed in the editing interface, the score corresponding to the lyrics to be dubbed is obtained by splicing the dubbing fragments in the set of the dubbing fragments corresponding to the participles in the lyrics to be dubbed by the dubbing server according to the position of each participle in the at least one participle in the lyrics to be dubbed, and the set of the dubbing fragments corresponding to the participles in the lyrics to be dubbed is obtained by searching the dubbing library by the dubbing server according to the first corresponding relation.

30. The music editor of claim 29, wherein the editing interface has an edit button, the editing interface operable to set the displayed soundtrack to be editable upon receipt of a trigger to the edit button;

31. A music player, wherein the music player has a mixing interface and a playing interface;

the mixing interface is used for mixing lyrics to be dubbed and the dubbing music corresponding to the lyrics to be dubbed to obtain a song, the dubbing music corresponding to the lyrics to be dubbed is obtained by splicing the dubbing music in the dubbing music server according to the position of each participle in at least one participle in the lyrics to be dubbed in a dubbing music fragment set corresponding to the participle in the lyrics to be dubbed, the dubbing music fragment set corresponding to the participle in the lyrics to be dubbed is obtained by searching a dubbing music library according to a first corresponding relation by the dubbing music server, the dubbing music server comprises a dubbing music library, the dubbing music library comprises the first corresponding relation of the participle and the dubbing music fragment, and any participle has a dubbing music fragment set comprising at least one dubbing music fragment;

the playing interface is used for playing the songs.