CN108268530A - Dub in background music generation method and the relevant apparatus of a kind of lyrics - Google Patents

Dub in background music generation method and the relevant apparatus of a kind of lyrics Download PDF

Info

Publication number
CN108268530A
CN108268530A CN201611264888.5A CN201611264888A CN108268530A CN 108268530 A CN108268530 A CN 108268530A CN 201611264888 A CN201611264888 A CN 201611264888A CN 108268530 A CN108268530 A CN 108268530A
Authority
CN
China
Prior art keywords
background music
dubbing
segment
participle
music
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201611264888.5A
Other languages
Chinese (zh)
Other versions
CN108268530B (en
Inventor
叶舟
王瑜
张亚楠
苏飞
杨洋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba Group Holding Ltd
Original Assignee
Alibaba Group Holding Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alibaba Group Holding Ltd filed Critical Alibaba Group Holding Ltd
Priority to CN201611264888.5A priority Critical patent/CN108268530B/en
Priority to TW106126946A priority patent/TW201824249A/en
Priority to PCT/CN2017/117358 priority patent/WO2018121368A1/en
Publication of CN108268530A publication Critical patent/CN108268530A/en
Application granted granted Critical
Publication of CN108268530B publication Critical patent/CN108268530B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/63—Querying
    • G06F16/632—Query formulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/683—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/685—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using automatically derived transcript of audio data, e.g. lyrics

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management Or Editing Of Information On Record Carriers (AREA)
  • Machine Translation (AREA)

Abstract

The embodiment of the invention discloses dub in background music generation method and the relevant apparatus of a kind of lyrics,Server of dubbing in background music include the first correspondence between participle and segment of dubbing in background music with music storehouse,After obtaining by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing,Multiple participles can be utilized to determine the corresponding set of segments of dubbing in background music of multiple participles according to the first correspondence from music storehouse,Later using the word segment of the lyrics to be dubbed in background music as unit,Using segmenting corresponding set of segments of dubbing in background music in a word segment,The son corresponding to a word segment is combined into dub in background music,And the son of word segment is dubbed in background music and is spliced into dubbing in background music for the lyrics to be dubbed in background music,Due to the corresponding segment of dubbing in background music of the participle that prestored,Therefore the corresponding segment of dubbing in background music of participle can be matched after lyrics participle of dubbing in background music is treated,So as to automatically generate out dubbing in background music for the lyrics to be dubbed in background music using the position of segment and participle in the lyrics to be dubbed in background music of dubbing in background music matched,Effectively increase the efficiency dubbed in background music for the lyrics.

Description

Dub in background music generation method and the relevant apparatus of a kind of lyrics
Technical field
The present invention relates to data processing field, more particularly to dub in background music generation method and the relevant apparatus of a kind of lyrics.
Background technology
In musical composition, many situations are first to create the lyrics, are set a song to music further according to the lyrics, and as the lyrics complete phase That answers dubs in background music, so as to form a song.
How to quickly finish that dub in background music be a urgent problem needed to be solved for the lyrics, if it is possible to the basis of automation The content of the lyrics is dubbed in background music the efficiency that will greatly improve musical composition.
However, there is no a kind of modes that can be dubbed in background music automatically for the lyrics according to the content of the lyrics at present.
Invention content
In order to solve the above technical problem, the present invention provides dub in background music generation method and the relevant apparatus of a kind of lyrics, with Dubbing in background music for the lyrics to be dubbed in background music is automatically generated out using the position of segment and participle in the lyrics to be dubbed in background music of dubbing in background music matched, is had Effect improves the efficiency dubbed in background music for the lyrics.
The embodiment of the invention discloses following technical solutions：
In a first aspect, the present invention provides a kind of generation method of dubbing in background music of lyrics, it is described to dub in background music applied to server of dubbing in background music Server is included with music storehouse, and described to include participle and the first correspondence of segment of dubbing in background music with music storehouse, wherein any one is segmented Have there are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music, the method includes：
It obtains by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing, the lyrics to be dubbed in background music are included extremely Lack a word segment, a word segment includes at least one participle at least one word segment；
With music storehouse according to being searched the multiple participle, determine that each participle is distinguished using first correspondence Corresponding set of segments of dubbing in background music；
According to corresponding set of segments and point included by least one word segment of dubbing in background music of the multiple participle Word determines that the corresponding son of at least one word segment is dubbed in background music；
It is dubbed in background music by splicing at least one word segment corresponding son lyrics to be dubbed in background music described in generation of dubbing in background music.
Optionally, the first word segment is one at least one word segment, described according to the multiple point Word is corresponding to dub in background music set of segments and the participle included by least one word segment, determines at least one text The corresponding son of word slice section is dubbed in background music, including：
Obtain the corresponding set of segments of dubbing in background music of the included participle of the first word segment；
According to the neighbouring relations of included participle in the first word segment, determine to be wrapped from the first word segment Include the smooth degree segmented in corresponding set of segments of dubbing in background music between selected segment of dubbing in background music；
It smooth degree is met the one of preset condition assembles happy segment and dub in background music as the son of the first word segment.
Optionally, the first participle and the second participle are the adjacent participle in the first word segment, described in the basis The neighbouring relations of included participle in first word segment determine to dub in background music from participle is corresponding included by the first word segment Smooth degree in set of segments between selected segment of dubbing in background music, including：
Calculate dub in background music segment and second point in set of segments of dubbing in background music corresponding to the first participle in the first word segment The splicing degree between segment of dubbing in background music in set of segments of dubbing in background music corresponding to word is obtained between the first participle and the second participle At least one splicing score；
The first word segment is calculated extremely according to the splicing score of adjacent participle in the first word segment Few splicing total score；
It is described smooth degree met the one of predetermined threshold value assemble happy segment dub in background music as the son of the first word segment, Including：
One corresponding to a splicing total score is chosen from the splicing total score more than predetermined threshold value and assembles happy segment as institute The son for stating the first word segment is dubbed in background music.
Optionally, the piece of dubbing in background music calculated in set of segments of dubbing in background music corresponding to the first participle in the first word segment Splicing degree between the segment of dubbing in background music dubbed in background music in set of segments corresponding to section and the second participle obtains the first participle and the At least one splicing score between two participles, including：
A segment of dubbing in background music in set of segments of dubbing in background music corresponding to the first participle is obtained, obtains the second participle institute A segment of dubbing in background music in corresponding set of segments of dubbing in background music；
Calculate the splicing degree between the two segments of dubbing in background music, obtain the first participle and second participle between by this two The splicing score that a segment of dubbing in background music determines.
Optionally, the server of dubbing in background music further includes typelib, and the typelib includes participle and belongs to different song classes The possibility information of type, it is described to further include the second correspondence dubbed in background music between segment and types of songs with music storehouse；At described It is described with music storehouse according to the multiple participle lookup, before determining the corresponding set of segments of dubbing in background music of the multiple participle, further include：
The typelib is searched according to the multiple participle, determines the types of songs belonging to the lyrics to be dubbed in background music；
It is described to match music storehouse according to being searched the multiple participle, it is determined using first correspondence the multiple Corresponding set of segments of dubbing in background music is segmented, including：
According to second correspondence, from the song class with belonging to filtering out in music storehouse with the lyrics to be dubbed in background music The corresponding segment of dubbing in background music of type；
According to first correspondence, from segment of dubbing in background music corresponding with the types of songs belonging to the lyrics to be dubbed in background music Determine the corresponding set of segments of dubbing in background music of the multiple participle.
Optionally, it further includes：
Historical song is obtained, the historical song includes the history lyrics, history matches happy types of songs；
The history lyrics are segmented；
Count the occurrence number and the song of third participle and affiliated historical song that third segments during participle The correspondence of bent type, so that it is determined that going out the number that third participle occurs described in same types of songs, the third Segment a participle to be obtained from the history lyrics；
Occurrence number during participle and the third described in same types of songs point are segmented according to the third The number that word occurs determines that the third participle belongs to the possibility information of different types of songs；
The possibility information that participle according to being obtained in the history lyrics belongs to different types of songs establishes the type Library.
Optionally, it further includes：
Historical song is obtained, the historical song includes the history lyrics, history matches happy types of songs；
The history lyrics are segmented；
Participle according to obtaining determines the corresponding segment of dubbing in background music in the history is dubbed in background music；
Participle according to obtaining dubs in background music the first correspondence of segment in dubbing in background music with the history and the history is dubbed in background music In dub in background music segment and types of songs the second correspondence establish it is described with music storehouse.
Optionally, it is described to obtain by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing, including：
Word segmentation processing is carried out to the lyrics to be dubbed in background music of acquisition, obtains multiple participles.
Second aspect, it is described to dub in background music applied to server of dubbing in background music the present invention provides a kind of generating means of dubbing in background music of lyrics Server is included with music storehouse, and described to include participle and the first correspondence of segment of dubbing in background music with music storehouse, wherein any one is segmented There are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music, described device includes acquiring unit, searching unit, determines tool Unit and generation unit：
The acquiring unit, for obtaining by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing, institute It states the lyrics to be dubbed in background music and includes at least one word segment, a word segment includes at least one at least one word segment A participle；
The searching unit, for, with music storehouse, utilizing first correspondence according to the multiple participle lookup Determine the corresponding set of segments of dubbing in background music of each participle；
The determination unit, for according to corresponding set of segments and at least one text of dubbing in background music of the multiple participle Participle included by word slice section determines that the corresponding son of at least one word segment is dubbed in background music；
The generation unit splices the corresponding son of at least one word segment and dubs in background music described in generation for passing through The lyrics to be dubbed in background music are dubbed in background music.
Optionally, the first word segment is one at least one word segment, and the determination unit is specifically used Corresponding set of segments of dubbing in background music is segmented included by the first word segment in obtaining；It is wrapped according in the first word segment The neighbouring relations of participle are included, determine selected from the corresponding set of segments of dubbing in background music of participle included by the first word segment match Smooth degree between happy segment；Smooth degree is met the one of preset condition and assembles happy segment as the first word segment Son dub in background music.
Optionally, the first participle and second is segmented as the adjacent participle in the first word segment, the determination unit Including computation subunit, obtain subelement and as subelement：
The computation subunit, for calculating in set of segments of dubbing in background music corresponding to the first participle in the first word segment The segment of dubbing in background music dubbed in background music in set of segments corresponding to segment and the second participle of dubbing in background music between splicing degree, obtain described first At least one splicing score between participle and the second participle；
It is described to obtain subelement, for institute to be calculated according to the splicing score of adjacent participle in the first word segment State at least one splicing total score of the first word segment；
It is described to be used as subelement, for being chosen from the splicing total score more than predetermined threshold value corresponding to a splicing total score One assembles happy segment dubs in background music as the son of the first word segment.
Optionally, the computation subunit is specifically used for obtaining one in set of segments of dubbing in background music corresponding to the first participle A segment of dubbing in background music obtains a segment of dubbing in background music in set of segments of dubbing in background music corresponding to second participle；The two are calculated to dub in background music Splicing degree between segment obtains the spelling determined between the first participle and the second participle by the two segments of dubbing in background music Connect score.
Optionally, the server of dubbing in background music further includes typelib, and the typelib includes participle and belongs to different song classes The possibility information of type, it is described to further include the second correspondence dubbed in background music between segment and types of songs with music storehouse；Described device Further include determining types of songs unit：
The determining types of songs unit for searching the typelib according to the multiple participle, determines described to wait to match Types of songs belonging to music and songs word；
The searching unit includes screening subelement and determination subelement：
The screening subelement for according to second correspondence, is treated with described from described with being filtered out in music storehouse The corresponding segment of dubbing in background music of the types of songs belonging to the lyrics of dubbing in background music；
The determination subelement, for according to first correspondence, the song belonging to the lyrics to be dubbed in background music The corresponding set of segments of dubbing in background music of the multiple participle is determined in the corresponding segment of dubbing in background music of type.
Optionally, described device, which further includes, obtains historical song unit, participle unit, statistic unit, determines information unit With establish unit：
The acquisition historical song unit, for obtaining historical song, the historical song includes the history lyrics, history is matched Happy types of songs；
The participle unit, for being segmented to the history lyrics；
The statistic unit, for counting the occurrence number and third participle that third segments during participle With the correspondence of the types of songs of affiliated historical song, segmented out so that it is determined that going out the third described in same types of songs Existing number, the third participle are the participle obtained from the history lyrics；
The determining information unit, for according to third occurrence number of the participle during participle and same The number that the participle of third described in types of songs occurs determines that the third participle belongs to the possibility letter of different types of songs Breath；
It is described to establish unit, for belonging to the possibility of different types of songs according to the participle obtained in the history lyrics Information establishes the typelib.
Optionally, described device, which further includes, obtains historical song unit, participle unit, determines dub in background music slice unit and foundation Unit：
The acquisition historical song unit, for obtaining historical song, the historical song includes the history lyrics, history is matched Happy types of songs；
The participle unit, for being segmented to the history lyrics；
It is described to determine slice unit of dubbing in background music, for determining corresponding in the history is dubbed in background music dub in background music according to obtained participle Segment；
It is described to establish unit, for the first corresponding pass of segment of dubbing in background music in being dubbed in background music according to obtained participle with the history System and the history dub in background music in dubbing in background music segment and types of songs the second correspondence establish it is described with music storehouse.
Optionally, the acquiring unit is specifically used for carrying out word segmentation processing to the lyrics to be dubbed in background music of acquisition, obtains multiple points Word.
The third aspect, it is described to dub in background music applied to server of dubbing in background music the present invention provides a kind of generation method of dubbing in background music of lyrics Server is included with music storehouse, wherein first correspondence that segment and segment of dubbing in background music are included with music storehouse, any one segment Have there are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music, the method includes：
It obtains and carries out the obtained at least one segment of word segmentation processing by treating the lyrics of dubbing in background music；
With music storehouse according at least one segment search, by first correspondence determine it is described at least The corresponding set of segments of dubbing in background music of one segment；
According to the lyrics to be dubbed in background music, institute is determined from the corresponding set of segments of dubbing in background music of at least one segment State the corresponding segment of dubbing in background music of each segment at least one segment；
The lyrics to be dubbed in background music described in segment generation of dubbing in background music determined by splicing are dubbed in background music.
Optionally, the segment is participle, and the acquisition is obtained extremely by treating lyrics progress word segmentation processing of dubbing in background music A few segment, including：
It obtains by carrying out the obtained multiple participles of word segmentation processing to the lyrics to be dubbed in background music.
Optionally, the segment is word segment, and the lyrics to be dubbed in background music include at least one word segment.
Optionally, the lyrics to be dubbed in background music include at least one word segment, one at least one word segment Word segment includes at least one participle, then the lyrics to be dubbed in background music described in the basis, is corresponded to respectively from least one participle Set of segments of dubbing in background music in determine each to segment corresponding segment of dubbing in background music at least one participle, including：
Corresponding dub in background music included by set of segments and at least one word segment is segmented according to the multiple Participle, determine that the corresponding son of at least one word segment is dubbed in background music, wherein as included by a word segment point The son that the corresponding segment of dubbing in background music of word is obtained corresponding to this word segment is dubbed in background music；
The lyrics to be dubbed in background music described in segment generation of dubbing in background music determined by splicing are dubbed in background music, including：
It is dubbed in background music by splicing at least one word segment corresponding son lyrics to be dubbed in background music described in generation of dubbing in background music.
Fourth aspect, it is described to dub in background music applied to server of dubbing in background music the present invention provides a kind of generating means of dubbing in background music of lyrics Server is included with music storehouse, wherein first correspondence that segment and segment of dubbing in background music are included with music storehouse, any one segment There are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music, described device includes acquiring unit, searching unit, determines tool Unit and concatenation unit：
The acquiring unit carries out obtained at least one of word segmentation processing for obtaining by treating the lyrics of dubbing in background music Section；
The searching unit for matching music storehouse according at least one segment search, is corresponded to by described first Relationship determines the corresponding set of segments of dubbing in background music of at least one segment；
The determination unit for the lyrics to be dubbed in background music according to, is dubbed in background music from least one segment is corresponding The corresponding segment of dubbing in background music of each segment at least one segment is determined in set of segments；
The concatenation unit splices dubbing in background music for the lyrics to be dubbed in background music described in segment generation of dubbing in background music determined for passing through.
Optionally, the acquiring unit is specifically used for obtaining by obtained by carrying out the lyrics to be dubbed in background music word segmentation processing Multiple participles.
Optionally, the segment is word segment, and the lyrics to be dubbed in background music include at least one word segment.
Optionally, the determination unit be specifically used for according to it is the multiple participle it is corresponding dub in background music set of segments and Participle included by least one word segment determines that the corresponding son of at least one word segment is dubbed in background music, The son that the corresponding segment of dubbing in background music of participle is obtained corresponding to this word segment wherein as included by a word segment is dubbed in background music；
The concatenation unit is specifically used for dubbing in background music generation by splicing the corresponding son of at least one word segment The lyrics to be dubbed in background music are dubbed in background music.
5th aspect, the present invention provides a kind of acquisition methods of dubbing in background music of lyrics, applied to interaction end, the method packet It includes：
The acquired lyrics to be dubbed in background music are sent to server of dubbing in background music, the lyrics to be dubbed in background music include at least one letter plate Section, a word segment includes at least one participle at least one word segment, and the server of dubbing in background music includes dubbing in background music Library, described to include participle and the first correspondence of segment of dubbing in background music with music storehouse, wherein there are one include extremely for any one participle tool The set of segments of dubbing in background music of a few segment of dubbing in background music；
From the server of dubbing in background music obtain described in the lyrics to be dubbed in background music are corresponding dubs in background music, it is described to dub in background music as the server of dubbing in background music Fragment assembly of dubbing in background music in set of segments of being dubbed in background music according to corresponding to participle in the lyrics to be dubbed in background music obtains, the lyrics to be dubbed in background music Set of segments of dubbing in background music corresponding to middle participle is obtained for the server of dubbing in background music according to being searched first correspondence with music storehouse It arrives.
Optionally, it is described from the server of dubbing in background music obtain described in the lyrics to be dubbed in background music are corresponding dubs in background music, including：
From the server of dubbing in background music obtain described in the lyrics to be dubbed in background music it is corresponding it is multiple it is undetermined dub in background music, the carrying undetermined of dubbing in background music There is soundtrack information；
According to the multiple soundtrack information undetermined dubbed in background music, select to meet the undetermined of demand and dub in background music as the music and songs to be matched Word is dubbed in background music.
Optionally, the soundtrack information includes carrying the scoring undetermined dubbed in background music of the soundtrack information and/or type of dubbing in background music.
Optionally, it is described from it is described dub in background music server obtain described in the lyrics to be dubbed in background music are corresponding dub in background music after, further include：
If dubbing in background music for obtaining does not meet demand, feedback information is sent to the server of dubbing in background music, so that the server The lyrics to be dubbed in background music dubs in background music according to regenerating the feedback information.
Optionally, the feedback information includes describing the information of the demand.
6th aspect, the present invention provides a kind of acquisition device of dubbing in background music of lyrics, and applied to interaction end, described device includes Transmitting element and acquiring unit：
The transmitting element, for sending the acquired lyrics to be dubbed in background music, the lyrics packet to be dubbed in background music to server of dubbing in background music At least one word segment is included, a word segment includes at least one participle at least one word segment, described to match Happy server is included with music storehouse, first correspondence including participle and segment of dubbing in background music, wherein any one point with music storehouse There are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music for word tool；
The acquiring unit, it is described to match for the lyrics to be dubbed in background music to be corresponding dubs in background music described in being obtained from the server of dubbing in background music The happy fragment assembly of dubbing in background music in set of segments of dubbing in background music corresponding to being segmented in server lyrics to be dubbed in background music according to of dubbing in background music It obtains, set of segments of dubbing in background music corresponding to participle in the lyrics to be dubbed in background music is corresponded to for the server of dubbing in background music according to described first closes What system obtained described in searching with music storehouse.
Optionally, the acquiring unit includes obtaining subelement and selection subelement：
The acquisition subelement multiple undetermined is matched for the lyrics to be dubbed in background music described in being obtained from the server of dubbing in background music to be corresponding Happy, undetermined dub in background music carries soundtrack information；
The selection subelement, for according to the multiple soundtrack information undetermined dubbed in background music, selection to meet the undetermined of demand It dubs in background music dubbing in background music as the lyrics to be dubbed in background music.
Optionally, the soundtrack information includes carrying the scoring undetermined dubbed in background music of the soundtrack information and/or type of dubbing in background music.
Optionally, described device further includes feedback unit：
If what is obtained dubs in background music and do not meet demand, the feedback unit is triggered, the feedback unit, for the clothes of dubbing in background music Device of being engaged in sends feedback information, so that server lyrics to be dubbed in background music according to regenerating the feedback information are matched It is happy.
Optionally, the feedback information includes describing the information of the demand.
7th aspect, the present invention provides a kind of scoring system of the lyrics, the scoring system include dub in background music server and Interaction end：
The server of dubbing in background music is included with music storehouse, first correspondence including participle and segment of dubbing in background music with music storehouse, Wherein any one participle tool there are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music, use by the server of dubbing in background music In obtaining by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing, the lyrics to be dubbed in background music include at least one Word segment, a word segment includes at least one participle at least one word segment；According to the multiple participle With music storehouse described in searching, the corresponding set of segments of dubbing in background music of each participle is determined using first correspondence；According to The multiple participle corresponding dub in background music set of segments and the participle included by least one word segment, are determined described The corresponding son of at least one word segment is dubbed in background music；It is dubbed in background music by splicing the corresponding son of at least one word segment The lyrics to be dubbed in background music dubs in background music described in generation；
The interaction end, for the server of dubbing in background music send it is acquired described in the lyrics to be dubbed in background music；It dubs in background music from described The lyrics to be dubbed in background music are corresponding described in server acquisition dubs in background music.
Eighth aspect, the present invention provides a kind of music editor, the music editor has to be inputted for obtaining The input interface of the lyrics to be dubbed in background music and the editing interface dubbed in background music for displaying；
For the input interface for obtaining the lyrics to be dubbed in background music, the acquired lyrics to be dubbed in background music are defeated in the input interface The lyrics or the lyrics to be obtained after the voice of input is identified entered；
When obtaining output order, the music editor is used to send to server of dubbing in background music and be obtained by the input interface The lyrics to be dubbed in background music taken, the server of dubbing in background music are included with music storehouse, described the first couple for including participle and segment of dubbing in background music with music storehouse It should be related to, wherein there are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music for any one participle tool；
When from it is described dub in background music server obtain described in wait the lyrics of dubbing in background music are corresponding dub in background music when, the music editor be used for will Corresponding dub in background music of the lyrics to be dubbed in background music is illustrated in the editing interface, and corresponding dub in background music of the lyrics to be dubbed in background music is matched to be described Fragment assembly of dubbing in background music in set of segments of dubbing in background music corresponding to participle in happy server lyrics to be dubbed in background music according to obtains, described to treat The set of segments of dubbing in background music corresponding to participle in the lyrics of dubbing in background music is dubbed in background music for the server of dubbing in background music according to being searched the first correspondence What library obtained.
Optionally, the editing interface has Edit button, when receiving the triggering to the Edit button, the volume Editing interface is used to dubbing in background music of being shown being set as editable；
What the editing interface was shown dub in background music to further include is dubbed in background music by what is identified to institute's recording audio data.
9th aspect, the present invention provides a kind of music player, the music player has mixed interface and broadcasting Interface；
The mixed interface is for the lyrics to be dubbed in background music and corresponding dub in background music of the lyrics to be dubbed in background music to be mixed to get Song, the lyrics to be dubbed in background music are corresponding to dub in background music as piece of dubbing in background music corresponding to participle in server lyrics to be dubbed in background music according to of dubbing in background music Fragment assembly of dubbing in background music in Duan Jihe obtains, and set of segments is dubbed in background music as the clothes of dubbing in background music corresponding to participle in the lyrics to be dubbed in background music Business device searches what is obtained with music storehouse according to the first correspondence, and the server of dubbing in background music is included with music storehouse, described to include with music storehouse Segment and dub in background music the first correspondence of segment, and wherein there are one include matching at least one segment of dubbing in background music for any one participle tool Happy set of segments；
The broadcast interface is used to play the song.
It dubs in background music it can be seen from above-mentioned technical proposal and is preset with including first pair between the segment that segments and dub in background music in server What should be related to matches music storehouse, after obtaining by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing, can utilize Multiple participle determines the corresponding set of segments of dubbing in background music of multiple participle from music storehouse according to the first correspondence, it Afterwards using the word segment of the lyrics to be dubbed in background music as unit, using corresponding set of segments of dubbing in background music is segmented in a word segment, combine Go out the son corresponding to a word segment to dub in background music, and the son of word segment is dubbed in background music and is spliced into dubbing in background music for the lyrics to be dubbed in background music, it can See, due to the corresponding segment of dubbing in background music of the participle that prestored, therefore can be matched after lyrics participle of dubbing in background music is treated and segment corresponding match Happy segment is waited to match so as to automatically generate out using the position of segment and participle in the lyrics to be dubbed in background music of dubbing in background music matched Music and songs word is dubbed in background music, and effectively increases the efficiency dubbed in background music for the lyrics.
Description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, to embodiment or will show below There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, may be used also To obtain other attached drawings according to these attached drawings.
Fig. 1 is a kind of system schematic of the generation of dubbing in background music of lyrics provided in an embodiment of the present invention；
Fig. 2 is a kind of method flow diagram of the generation method of dubbing in background music of lyrics provided in an embodiment of the present invention；
Fig. 2 a are a kind of flow of method for determining the corresponding son of the first word segment and dubbing in background music provided in an embodiment of the present invention Figure.
Fig. 3 is a kind of flow diagram that process is established with music storehouse provided in an embodiment of the present invention；
Fig. 4 is the flow diagram that a kind of typelib provided in an embodiment of the present invention establishes process；
Fig. 5 is the method flow diagram of the generation method of dubbing in background music of another lyrics provided in an embodiment of the present invention；
Fig. 6 is a kind of method flow diagram of the acquisition methods of dubbing in background music of lyrics provided in an embodiment of the present invention；
Fig. 7 is a kind of structure drawing of device of the generating means of dubbing in background music of lyrics provided in an embodiment of the present invention；
Fig. 8 is the structure drawing of device of the generating means of dubbing in background music of another lyrics provided in an embodiment of the present invention；
Fig. 9 is a kind of structure drawing of device of the acquisition device of dubbing in background music of lyrics provided in an embodiment of the present invention；
Figure 10 is a kind of system construction drawing of the scoring system of the lyrics provided in an embodiment of the present invention.
Specific embodiment
Purpose, technical scheme and advantage to make the embodiment of the present invention are clearer, below in conjunction with the embodiment of the present invention In attached drawing, the technical solution in the embodiment of the present invention is explicitly described, it is clear that described embodiment be the present invention Part of the embodiment, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not having All other embodiments obtained under the premise of creative work are made, shall fall within the protection scope of the present invention.
During musical composition, the completion of a song is usually song after first word, i.e., first creates the lyrics, further according to The lyrics are set a song to music and (are dubbed in background music), ultimately form a song.And the lyrics dub in background music and generally require to take a substantial amount of time.With Development of the deep learning in voice field, more and more tasks can by Machine automated be completed, if it is possible to from The content according to the lyrics of dynamicization is dubbed in background music the efficiency that musical composition will be greatly improved.
For this purpose, the embodiment of the present invention provides a kind of generation method of dubbing in background music of the lyrics, server of dubbing in background music is obtained is matched by treating Music and songs word carries out multiple participles acquired in word segmentation processing, the participle and the correspondence for segment of dubbing in background music that foundation pre-saves, The corresponding segment of dubbing in background music of participle is fitted on, so as to utilize the position of segment and participle in the lyrics to be dubbed in background music of dubbing in background music matched It puts and automatically generates out dubbing in background music for the lyrics to be dubbed in background music.
Based on above-mentioned thought, the embodiment of the present invention can be realized by dubbing in background music server.It may refer to shown in Fig. 1, match Happy server is included with music storehouse 200, wherein with the correspondence that participle and segment of dubbing in background music are preserved in music storehouse 200.With first For word segment 100, after server of dubbing in background music carries out word segmentation processing to the first word segment 100, the first letter plate can be obtained The participle that section 100 includes, such as the first participle, the second participle and third participle.By matching the participle preserved in music storehouse 200 and piece of dubbing in background music The correspondence of section, it may be determined that go out to segment corresponding set of segments 300 of dubbing in background music.As shown in Figure 1, the first participle and segment of dubbing in background music Set A is corresponding, and the second participle is corresponding with the set of segments B that dubs in background music, and third participle is corresponding with the set of segments C that dubs in background music.Each Set of segments of dubbing in background music includes at least one segment of dubbing in background music, and set of segments of such as dubbing in background music A can include dub in background music segment a1, a2 and a3, Set of segments of dubbing in background music B can include dub in background music segment b1 and b2, and set of segments of dubbing in background music C can include dub in background music segment c1, c2 and c3.Match Happy server chooses a suitable segment of dubbing in background music from the corresponding set of segments of dubbing in background music of each participle, as the first word segment Son dub in background music 400, can such as choose the segment a1-b2-c3 that dubs in background music and dub in background music 400 as the son of the letter plate section.
In embodiments of the present invention, a first lyrics can be understood as being used to form the lyrics of a first full songs, a first song It can include at least one word segment in word, a word segment can be a paragraph of word structural domain, can also It is the word or passage met using punctuate as division.It can be by least one participle group in one word segment Into.A word segment in first word segment i.e. at least one word segment.The present invention, which does not limit, carries out the lyrics The specific implementation of participle can be marked off from the lyrics by the meaning of a word or word structure etc. with the independent meaning of a word or word knot The participle of structure, the participle marked off can be a word or a word.Such as " I loves Beijing to a word segment The participle that Tian An-men " can mark off includes " I ", " love ", " Beijing " and " Tian An-men ".
Segment of dubbing in background music can be understood as to dub in background music segment determined by a participle, and a general participle is in one In the lyrics, this lyrics has what is prepared to dub in background music, and can first determine that this segments the position in this lyrics, the position Putting can be identified by the time range corresponding to the participle on the timeline of the lyrics.Match further according to the position from this A good part corresponding to middle interception position of dubbing in background music is dubbed in background music, such as the part dubbed in background music in the time range is dubbed in background music, It can be segment of dubbing in background music corresponding to the participle that this part, which is dubbed in background music,.
Correspondingly, a song can include, the lyrics are corresponding with the lyrics to dub in background music, in the lyrics a word segment It dubs in background music with this corresponding son of word segment, a son part corresponding in this is dubbed in background music for this word segment of dubbing in background music is matched It is happy.This corresponding son of word segment, which is dubbed in background music, to be formed, such as one by segmenting corresponding segment of dubbing in background music in this word segment The participle that a word segment " I loves Beijing Tian An-men " do not dubbed in background music can mark off includes " I ", " love ", " Beijing " and " Tian An Door ".There is participle " I " the corresponding set of segments A that dubs in background music to include dub in background music segment a1, a2 and a3, and participle " love " has corresponding Set of segments of dubbing in background music B includes dub in background music segment b1 and b2, and there is participle " Beijing " the corresponding set of segments C that dubs in background music to include dubbing in background music There is the corresponding set of segments D that dubs in background music to include the segment d1 that dubs in background music for segment c1, c2 and c3, participle " Tian An-men ", then this text It can be respectively by a piece of dubbing in background music in dub in background music set of segments A, B, C and D that the son of word slice section " I loves Beijing Tian An-men ", which is dubbed in background music, Section is formed, such as a1-b2-c3-d1 or a2-b1-c3-d1 etc..
The lyrics dubbed in background music can be known as the lyrics to be dubbed in background music, the lyrics to be dubbed in background music are understood by above-mentioned introduction, from See can have and segment in composition form or granularity of division, the two levels of word segment, and what the embodiment of the present invention was provided Generation method of dubbing in background music can be applied to this kind of lyrics to be dubbed in background music or this granularity of division for treating the lyrics of dubbing in background music, and can also be applied to There are the lyrics to be dubbed in background music of other composition forms or applied to other granularity of division for treating the lyrics of dubbing in background music.
It next will be from corresponding segment of dubbing in background music is segmented, to determine that the corresponding son of word segment is dubbed in background music, be treated so as to obtain For the mode dubbed in background music of the lyrics of dubbing in background music, the generation method of dubbing in background music for the lyrics that the embodiment of the present invention is provided is discussed in detail.Fig. 2 is A kind of method flow diagram of the generation method of dubbing in background music of lyrics provided in an embodiment of the present invention, this method include：
S201：It obtains by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing.
When need be when the lyrics of dubbing in background music are dubbed in background music, the lyrics can be manually entered or phonetic entry to relevant equipment in.
It can be dubbed in background music in embodiments of the present invention by server of dubbing in background music to the lyrics.It dubs in background music to include in server and match Music storehouse is previously stored with participle and the first correspondence of segment of dubbing in background music, for the ease of server pair of dubbing in background music in this is with music storehouse The lyrics to be dubbed in background music are dubbed in background music automatically, it is necessary first to be treated the lyrics of dubbing in background music and be carried out word segmentation processing.
The specific equipment that the embodiment of the present invention treats lyrics progress word segmentation processing of dubbing in background music is not construed as limiting, and can be by clothes of dubbing in background music Business device treats the lyrics of dubbing in background music and carries out word segmentation processing, obtains multiple participles.Can also be that lyrics progress of dubbing in background music is treated by other equipment Word segmentation processing obtains multiple participles, and server of dubbing in background music can be got from the equipment carries out after word segmentation processing obtain multiple points Word.
When carrying out word segmentation processing, progressive form specifically may be used and treat lyrics progress word segmentation processing of dubbing in background music, you can The lyrics to be dubbed in background music first are divided into word segment, then word segmentation processing is carried out to word segment.Wherein it is possible to made with punctuation mark To divide the foundation of word segment or using paragraph as the foundation for dividing word segment.Word segmentation processing is carried out to word segment, The participle with the independent meaning of a word or word structure can be specifically marked off from word segment by the meaning of a word or word structure etc.. The lyrics of dubbing in background music can also directly be treated and carry out word segmentation processing, then determine the word segment belonging to participle.
For example, the lyrics to be dubbed in background music " I loves Beijing Tian An-men, sun liter on Tian An-men ", if using punctuation mark as dividing text The foundation of word slice section, the lyrics can be divided into " I loves Beijing Tian An-men " and " sun liter on Tian An-men " the two word segments. Participle with the independent meaning of a word or word structure is marked off from the lyrics by the meaning of a word or word structure etc., wherein to word segment " I loves Beijing Tian An-men " carries out word segmentation processing, " I ", " love ", " Beijing ", " Tian An-men " this four participles can be obtained, to text Word slice section " sun liter on Tian An-men " carries out word segmentation processing, can obtain " Tian An-men ", " on ", " sun ", this four points of " liter " Word.
For the lyrics to be dubbed in background music of big length, word segmentation processing, the participle that can cause are carried out using above-mentioned progressive form It is more accurate, reduce the probability of error.
S202：With music storehouse according to being searched the multiple participle, each point is determined using first correspondence The corresponding set of segments of dubbing in background music of word.
Server of dubbing in background music can be pre-established with music storehouse, and the correspondence of participle and segment of dubbing in background music is preserved in this is with music storehouse The correspondence can be known as the first correspondence by relationship.At this in music storehouse, a participle can have it is corresponding at least One segment of dubbing in background music.
The one head lyrics to be dubbed in background music can be divided into multiple participles, can have due to matching a participle in music storehouse with corresponding At least one segment of dubbing in background music, therefore when by the matching of the participle that marks off with music storehouse, a participle can match one or Multiple segments of dubbing in background music, the segment of dubbing in background music that can be matched the participle marked off for one segment corresponding match as this There are one corresponding set of segments of dubbing in background music for happy set of segments namely a participle tool.For example, segment " I " in a song Three segments of dubbing in background music, respectively a1, a2 and a3 are fitted on, segment that these three can be dubbed in background music is as the segment collection of dubbing in background music of participle " I " It closes.
With the first participle, i.e., for a participle in multiple participle, determine the corresponding piece of dubbing in background music of the first participle Duan Jihe is specifically included：With music storehouse according to being searched the first participle, the corresponding packet of the first participle is determined Include the set of segments of dubbing in background music of at least one segment of dubbing in background music.
The method for determining corresponding set of segments of dubbing in background music for other participles determines corresponding segment of dubbing in background music with the first participle The method of set is identical, and details are not described herein.
S203：Corresponding dub in background music included by set of segments and at least one word segment is segmented according to the multiple Participle, determine that the corresponding son of at least one word segment is dubbed in background music.
The lyrics to be dubbed in background music include at least one word segment, with the first word segment, i.e., at least one word segment A word segment for be unfolded introduce.If a participle is only included in the first word segment, then can be directly from this Determine that a segment of dubbing in background music is dubbed in background music as the son of the first word segment in participle.If the first word segment includes multiple Participle, determines that the method that the corresponding son of the first word segment is dubbed in background music can specifically include S2031- as shown in Figure 2 a S2033。
S2031：Obtain the corresponding set of segments of dubbing in background music of the included participle of the first word segment.
At least one participle that first word segment includes can be got by the word segmentation processing of S201, it can by S202 To determine the corresponding set of segments of dubbing in background music of at least one participle that the first word segment includes.
S2032：According to the neighbouring relations of included participle in the first word segment, determine from first letter plate Smooth degree in the corresponding set of segments of dubbing in background music of the included participle of section between selected segment of dubbing in background music.
When get at least one participle included by the first word segment it is corresponding dub in background music set of segments after, need A suitable segment of dubbing in background music is chosen from the corresponding set of segments of dubbing in background music of each participle suitably to dub in background music segment as the participle, So that the son of dub in background music fragment assembly into the first word segment selected is dubbed in background music.
A segment of dubbing in background music is selected from the corresponding set of segments of dubbing in background music of each participle, the first word segment includes There are several participles, several segments of dubbing in background music can be selected accordingly, segment that this can be dubbed in background music assembles happy segment, example as one Such as, the first word segment includes three participles, participle 1, participle 2 and participle 3, and participle 1 is corresponded to there are one set of segments of dubbing in background music, Include that there are three dub in background music segment in the set of segments of dubbing in background music；Participle 2 is corresponded to there are one set of segments of dubbing in background music, in the segment of dubbing in background music Set includes that there are one dub in background music segment；Participle 3 is corresponded to there are one set of segments of dubbing in background music, and is included in this dubs in background music set of segments Two segments of dubbing in background music select a segment of dubbing in background music from the corresponding set of segments of dubbing in background music of each participle, can be assembled as one Happy segment can select six for the first word segment and assemble happy segment.
By taking one assembles happy segment as an example, this assembles the smoothness that whether happy segment is suitable, can be dependent between segment of dubbing in background music Degree differentiates.Smooth degree can reflect the comfort level between segment of dubbing in background music, and smooth degree is higher, between the segment that illustrates to dub in background music Comfort level it is better namely generation tune it is more graceful, better hearing enjoying can be brought to user.
Between segment is dubbed in background music in judgement during smooth degree, multiple (at least two) pieces of dubbing in background music with neighbouring relations are referred to Smooth degree between section.Wherein, the neighbouring relations between segment of dubbing in background music can be dependent in the first word segment between participle Neighbouring relations determine that for example, two adjacent participles in the first word segment, it is corresponding then to segment 1 for participle 1 and participle 2 It dubs in background music segment, segment of dubbing in background music corresponding with participle 2 is with neighbouring relations.
For the height of smooth degree, can specifically be represented by the splicing degree between segment of dubbing in background music.
In embodiments of the present invention, can be used as by music principle judge two segments of dubbing in background music between splicing degree according to According to.Splicing degree can be understood as the linking degree of the segment of dubbing in background music of two adjacent participles, and the linking degree the high then to represent two participles Between melody it is more graceful.For example, song and noise, the two is most essential, and difference lies in song has graceful melody, can To bring enjoyment acoustically to people, and the melody of noise is poor can make one to feel ear-piercing.Therefore the splicing for segment of dubbing in background music in song Degree is one of principal element for influencing song lyric, and the splicing degree of two segments of dubbing in background music is better, the two segment spellings of dubbing in background music The music being connected together more meets music principle, and melody is more graceful, brings auditory effect better to people.
For the ease of getting information about the splicing degree between the segment of dubbing in background music of adjacent participle, splicing Score Lists can be used Show.The splicing degree of two segments of dubbing in background music is better, then corresponding splicing score is higher.
The construction that splicing degree between segment of dubbing in background music can be dependent on music is calculated, for example, melody, rhythm and The factors such as sound, contraposition, polyphony, musical form, musical instrument method, orchestration.It is numerous to calculate splicing degree factor needed to be considered, and nerve net Network has large-scale parallel, distributed storage and processing, self-organizing, adaptive and self-learning ability, and it is same to be particularly suitable for processing needs When consider many factors and condition, inaccurate and fuzzy information-processing problem, therefore can lead in embodiments of the present invention It crosses Recognition with Recurrent Neural Network model and dub in background music and splice the calculating of degree between segment, splicing journey is obtained so as to efficiently and accurately Spend corresponding splicing score.Further, it is also possible to be established with the happy lyrics using the lyrics as the inverted index of major key, the row's of falling rope is utilized Trained Recognition with Recurrent Neural Network model is attracted, with the performance of lift scheme.
By taking the adjacent participle of any two in the first word segment as an example, the two can be referred to as first point Word and second participle, calculate the two participle between splicing score method it is as follows：
Sa：Calculate the segment and second of dubbing in background music in set of segments of dubbing in background music corresponding to the first participle in the first word segment The splicing degree between segment of dubbing in background music in set of segments of dubbing in background music corresponding to participle obtains the first participle and the second participle Between at least one splicing score.
One participle can correspond at least one segment of dubbing in background music, and the segment of dubbing in background music of two neighboring participle may be constructed a variety of groups It closes, each combination is corresponding, and there are one splicing scores, how many kind combination is corresponding can to obtain how many splicing scores.Example Such as, two adjacent participles " I " and " love " in the first word segment, there are two segments of dubbing in background music for " I " tool, and there are three dub in background music " love " tool Segment, the segment of dubbing in background music of the two adjacent participles has six kinds of combinations, corresponding to obtain six splicing scores.
By taking one splices score as an example, the circular of the splicing score is as follows：
A segment of dubbing in background music in set of segments of dubbing in background music corresponding to the first participle is obtained, obtains the second participle institute A segment of dubbing in background music in corresponding set of segments of dubbing in background music；
Calculate the splicing degree between the two segments of dubbing in background music, obtain the first participle and second participle between by this two The splicing score that a segment of dubbing in background music determines.
The corresponding all splicing scores of two neighboring participle can be calculated according to the above method.For example, participle " I " and It is two adjacent participles to segment " love ", participle " I " with the corresponding set of segments A that dubs in background music include dubbing in background music segment a1, a2 and There is the corresponding set of segments B that dubs in background music to include dub in background music segment b1 and b2 for a3, participle " love ", and the segment of dubbing in background music of the two participles has Six kinds of combinations：A1b1, a1b2, a2b1, a2b2, a3b1 and a3b2 can obtain this six kinds of groups by Recognition with Recurrent Neural Network model Closing corresponding splicing score is respectively：3rd, 5,8,4,7 and 2.
The above-mentioned first participle and the second participle are adjacent two participle arbitrarily chosen in the first word segment, and first is literary Word slice section can include multiple participles, and each participle has participle adjacent thereto, segmented with reference to the above-mentioned first participle and second Splicing score computational methods, the splicing scores of other adjacent participles in the first word segment can be calculated, so as to obtain The corresponding splicing score of all adjacent participles in the letter plate section.
For example, the word segment " I loves Beijing Tian An-men " in the lyrics to be dubbed in background music can be divided into " I ", " love ", " Beijing " and " Tian An-men " this four participles, have three groups of adjacent participles：" I " and " love ", " love " and " Beijing ", " Beijing " " Tian An-men " according to the method for above-mentioned calculating splicing score, it is corresponding at least one can to calculate " I " and " love " respectively Splice score, " love " and " Beijing " corresponding at least one splicing score, " Beijing " and " Tian An-men " corresponding at least one spelling Connect score.
Sb：The first word segment is calculated according to the splicing score of adjacent participle in the first word segment At least one splicing total score.
The splicing total score calculated by splicing score can be used to indicate that a word segment as corresponding to each participle Segment of dubbing in background music between splicing degree, splicing degree more it is good then it is corresponding splicing total score it is higher.
Wherein, splice total score computational methods can there are many, can go out splice total score by splicing score addition calculation, Splicing total score can also be calculated by splicing score multiplication, can also will splice score be added according to certain weight proportion or Multiplication calculates splicing total score.The specific requirement that computational methods can dub in background music according to the lyrics is selected, in the embodiment of the present invention In, splicing total score is calculated using the method that splicing score is added.
For example, a word segment " I loves Beijing Tian An-men " be divided into " I ", " love ", " Beijing ", " Tian An-men " this four There is the corresponding set of segments A that dubs in background music to include dub in background music segment a1, a2 and a3 for a participle, participle " I ", participle " love " with pair The set of segments B that dubs in background music answered includes dub in background music segment b1 and b2, and there is participle " Beijing " the corresponding set of segments C that dubs in background music to include There is the corresponding set of segments D that dubs in background music to include the segment d1 that dubs in background music for segment of dubbing in background music c1, c2 and c3, participle " Tian An-men "." I " and " love " is two adjacent participles, and the fragment combination of dubbing in background music of the two participles has six kinds：a1b1、a1b2、a2b1、a2b2、a3b1 And a3b2, corresponding splicing score is respectively 6,4,3,7,2,5；" love " and " Beijing " is two adjacent participles, the two divide The fragment combination of dubbing in background music of word has six kinds：B1c1, b2c1, b1c2, b2c2, b1c3 and b2c3, corresponding splicing score is respectively 6, 2、3、1、4、5；" Beijing " and " Tian An-men " is two adjacent participles, and there are three types of the fragment combinations of dubbing in background music of the two participles： C1d1, c2d1, c3d1, corresponding splicing score is respectively 4,2,5.It is all to segment corresponding segment of dubbing in background music in the letter plate section 18 kinds of combinations are may be constructed, it is corresponding to calculate 18 splicing total scores.Happy segment is assembled as a1-b1-c3-d1 when one, The splicing of middle a1b1 is scored at 6 points, and the splicing of b1c3 is scored at 4 points, and the splicing of c3d1 is scored at 5 points, if in a manner of addition If determining splicing total score, then the splicing total score for assembling happy segment is 15 points.
The corresponding splicing total score of all fragment combinations of dubbing in background music of the letter plate section can be calculated by the above method.Splicing is total It is point higher, then illustrate that the combination of the segment of dubbing in background music more meets music principle, the tune formed can be brought preferably to audience Auditory effect.
S2033：It smooth degree is met the one of preset condition assembles happy segment and match as the son of the first word segment It is happy.
Preset condition can be the condition for judging smooth degree height, and preset condition can pre-set, specifically may be used To be to weigh correlated condition during smooth degree according to music principle.The smoothness between segment of dubbing in background music in happy segment is assembled when one When degree meets preset condition, illustrate that smooth degree is sufficiently high, it is high to this can be assembled to happy segment as the first word The son of segment is dubbed in background music.
For the above-mentioned situation that smooth degree is represented with splicing degree, the first word segment can be calculated at least One splicing total score, corresponding, which can be a predetermined threshold value.It specifically, can be from more than default threshold One corresponding to a splicing total score is chosen in the splicing total score of value to assemble happy segment and as the son of the first word segment match It is happy.
One word segment can include multiple participles, and each participle can correspond to multiple segments of dubbing in background music, from each participle One segment of dubbing in background music of middle selection may be constructed one and assemble happy segment,
One, which assembles happy segment, corresponds to a splicing total score, can have multigroup segment of dubbing in background music, i.e., one in a word segment It can include multiple splicing total scores in word segment.It completes to dub in background music to the letter plate section, need from multigroup segment of dubbing in background music It selects one and assembles happy segment, happy segment can be assembled using select one and dubbed in background music as the corresponding son of letter plate section.
The son of one word segment dub in background music can be dependent on splicing total score chosen.Since a word segment is generally wrapped Multiple splicing total scores are included, therefore the mode that setting predetermined threshold value may be used is chosen.
Specifically, from the corresponding all splicing total scores of the letter plate section, the splicing total score more than predetermined threshold value is selected, Suitable splicing total score is chosen from the splicing total score more than predetermined threshold value again, which assembles happy piece The son that section can be used as the letter plate section is dubbed in background music.Predetermined threshold value is set, on the one hand can reduce the range of selection, on the other hand Undesirable fragment combination of dubbing in background music can be excluded.
Wherein, predetermined threshold value can be a fixed numbers, and the splicing total score selected more than predetermined threshold value chooses spelling The score value for connecing total score is more than the splicing total score of the fixed numbers.For example, the splicing total score of a word segment is respectively 1,3,2, 9th, 7,5,4,8,6, predetermined threshold value 5 is respectively then 6,7,8,9 more than the splicing total score of predetermined threshold value.It can be from this four groups Arbitrary selection one assembles happy segment and dubs in background music for being spliced into the son of this word segment, can also select splicing total from this four groups Divide highest one to assemble happy segment and dub in background music for being spliced into the son of this word segment.
It sets the opportunity of predetermined threshold value can be there are many selection, such as can just be pre-set before dubbing in background music for the lyrics, It can also be set before splicing total score is chosen after splicing total score is calculated.
Above-mentioned first word segment is the word segment arbitrarily selected in the multiple word segment, thus may be used Know, the generation method that the son of remaining word segment is dubbed in background music is identical with the generation method of above-mentioned first word segment, no longer superfluous herein It states.
S204：It is dubbed in background music the lyrics to be dubbed in background music described in generation by splicing at least one word segment corresponding son It dubs in background music.
The corresponding son of all word segments that the lyrics to be dubbed in background music include can be calculated by S203 to dub in background music, each word Segment has its fixed position in the lyrics, and therefore, the corresponding son of word segment all in the lyrics can be dubbed in background music Spliced according to the position where word segment, completely dubbed in background music so as to generate a head, that is, the lyrics to be dubbed in background music are dubbed in background music.
The above describes the generation method of dubbing in background music of the lyrics in detail, and due to matching, the participle that prestored in music storehouse is corresponding to dub in background music Segment, therefore the corresponding segment of dubbing in background music of participle can be matched after lyrics participle of dubbing in background music is treated, so as to utilize what is matched Segment of dubbing in background music and the position segmented in the lyrics to be dubbed in background music automatically generate out dubbing in background music for the lyrics to be dubbed in background music, and effectively increase as song The efficiency that word is dubbed in background music.
Next, by describing in detail to the process of establishing for matching music storehouse, it can be history sound to establish the foundation with music storehouse Find pleasure in or what a is pre-set by music principle and segment the possibility of dubbing in background music that can have.Optionally, the present invention will be right It is illustrated in a manner that history music is established and matches music storehouse.
As shown in figure 3, for a kind of flow diagram that process is established with music storehouse provided in an embodiment of the present invention, this was established Journey includes：
S301：Historical song is obtained, the historical song includes the history lyrics, history matches happy types of songs.
Historical song can be currently existed including the complete lyrics and the song dubbed in background music.The historical song can be deposited Storage is beyond the clouds or in server, when needing using these historical songs, can be obtained from high in the clouds or server.For the ease of right Historical song carries out subsequent processing, can include the history lyrics in the historical song, history matches happy types of songs.
Types of songs can be divided according to the melody of song, the content of the lyrics, rhythm for dubbing in background music etc..Such as according to song Bent development course, types of songs can substantially be divided into ten classes：Rhythm resentment bent (R＆B), rap music (RAP), rock music, jazz Pleasure, country music, new century music, Classical Pop pleasure, folk song, disco, Britpop (Trip-Hop).Alternatively, according to song The types of songs that bent melody can mark off includes lyric, sentiment etc..
S302：The history lyrics are segmented.
The lyrics are segmented can be identical with obtaining the method for multiple participles in S201, and details are not described herein.
S303：Participle according to obtaining determines the corresponding segment of dubbing in background music in the history is dubbed in background music.
Specifically, can first determine that this segments the position in the lyrics, which can pass through the timeline of the lyrics The time range corresponding to the participle is gone up to identify, further according to the position dubbing in background music one corresponding to middle interception position from the lyrics Part is dubbed in background music, as the corresponding segment of dubbing in background music of participle.
For example, in a song, according to time sequencing, the 50th second of " Tian An-men " in playback of songs, this point are segmented Time range corresponding to word can be -55 seconds 45 seconds, can be intercepted in dubbing in background music -55 seconds 45 seconds it is corresponding dub in background music, as this The segment of dubbing in background music of participle.
One participle may repeatedly occur in a song, and can be truncated to every time by occurring by corresponding one dubs in background music Segment, therefore, in a song, a participle can be corresponding with multiple segments of dubbing in background music.In addition, in different songs, also may be used Identical participle can be will appear.
It follows that a song may be come not only from by each segmenting corresponding segment of dubbing in background music, it is also possible to come from it Its song.Therefore, the historical song of acquisition is more, and obtained participle is more, and it is also wider to segment corresponding segment of dubbing in background music It is general, so as to make participle and corresponding segment of dubbing in background music more perfect.
S304：Participle according to obtaining dub in background music in dubbing in background music with the history segment the first correspondence and described go through History dub in background music in dubbing in background music segment and types of songs the second correspondence establish it is described with music storehouse.
It can be obtained segmenting corresponding segment of dubbing in background music by S303, dubbed in background music during can obtained participle be dubbed in background music with the history The correspondence of segment is known as the first correspondence.
Since segment of dubbing in background music all is to divide to obtain from historical song, therefore each segment of dubbing in background music has the history song belonging to it Song, the types of songs included according to a historical song, it may be determined that go out and dubbing in background music of obtaining is divided from this historical song Correspondence between segment and the types of songs of this historical song is different from above-mentioned first correspondence, can be by described in Dub in background music in the dubbing in background music correspondence of segment and types of songs of history is known as the second correspondence.
Participle according to obtaining dubs in background music the first correspondence of segment in dubbing in background music with the history and the history is dubbed in background music In dub in background music the second correspondence of segment and types of songs can establish out with music storehouse.After the completion of being established with music storehouse, it can preserve In server of dubbing in background music.
It is more perfect with music storehouse in order to make during establishing with music storehouse, often a large amount of historical song is carried out Analysis, so as to make each corresponding segment of dubbing in background music of participle more as far as possible, but when dubbing in background music for the lyrics, segment of dubbing in background music is got over More, the difficulty in computation for choosing the suitable segment of dubbing in background music of each participle is bigger.Therefore, segment of suitably dubbing in background music is being chosen for participle When, first segment of dubbing in background music tentatively can be screened, then segment of suitably dubbing in background music is chosen from the segment of dubbing in background music after screening.
Often exist with types of songs in view of the lyrics and closely contact, therefore can be according to the participle and song that the lyrics include The correlation degree of bent type dub in background music the preliminary screening of segment.
Specifically, the types of songs belonging to the lyrics to be dubbed in background music can be determined first, further according to the preserved in music storehouse Two correspondences, from described with filtering out segment of dubbing in background music corresponding with the types of songs belonging to the lyrics to be dubbed in background music in music storehouse.
After tentatively screening, further according to the first correspondence preserved in music storehouse, from the lyrics institute to be dubbed in background music The corresponding set of segments of dubbing in background music of the multiple participle is determined in the corresponding segment of dubbing in background music of the types of songs of category.
The types of songs of the lyrics to be dubbed in background music can be pre-set or be determined according to the content of the lyrics to be dubbed in background music , the typelib established in server of dubbing in background music can also be dependent on to determine, participle can be included in the type library and belong to different The possibility information of types of songs.Specifically, the typelib can be searched according to multiple participles that the lyrics to be dubbed in background music include, really Types of songs belonging to the fixed lyrics to be dubbed in background music, i.e., according to the possibility information in typelib, determine lyrics institute to be dubbed in background music The types of songs of category.
Whether the types of songs determined accurately directly affects the quality for quality of dubbing in background music being subsequently generated, and types of songs is got over Accurately, then the quality of dubbing in background music being subsequently generated is better, so that the song ultimately generated can bring the better sense of hearing to user Effect.Model-naive Bayesian is one of current most popular disaggregated model, rises in classical mathematics theory, there is heavily fortified point Real Fundamentals of Mathematics and the classification effectiveness stablized.Therefore, the type library can specifically pass through simplicity in embodiments of the present invention Bayesian model determines the types of songs belonging to the lyrics to be dubbed in background music.Further, it is also possible to pass through participle and the corresponding song of participle Type trains model-naive Bayesian, promotes its accuracy judged types of songs belonging to participle.
If it is determined that the lyrics to be dubbed in background music belong to a types of songs, then can pass through the segment of dubbing in background music of this types of songs It waits to dub in background music for this to segment to generate and dub in background music.Belong to the spy that the segment of dubbing in background music of this types of songs can more embody this types of songs Point, therefore the lyrics to be dubbed in background music also can more be met with this dubbing in background music of generating and want the style embodied.Further, pass through song class Screening of the type to segment of dubbing in background music can effectively reduce the segment of dubbing in background music for dubbing in background music for the lyrics to be dubbed in background music, so as to reduce Automatically the difficulty in computation dubbed in background music.
Typelib can be dependent on before above-mentioned preliminary screening by, which carrying out, first determines types of songs.
Next, by describing in detail to the process of establishing of typelib, as shown in figure 4, being provided for the embodiment of the present invention A kind of typelib establish the flow diagram of process, this is established process and includes：
S401：Historical song is obtained, the historical song includes the history lyrics, history matches happy types of songs.
S402：The history lyrics are segmented.
S401-S402 is identical with S301-S302 implementation procedures in Fig. 3, and details are not described herein.
S403：Count the occurrence number and third participle and affiliated history song that third segments during participle The correspondence of bent types of songs, so that it is determined that going out the number that third participle occurs described in same types of songs, institute It is the participle obtained from the history lyrics to state third participle.
Historical song is segmented, is word segmentation processing to be carried out to the history lyrics, therefore can be with as unit of every song Obtain the number that appearance is segmented in every song.By taking third segments as an example, after being segmented to the history lyrics, third point can be counted The number that word occurs in historical song, and the types of songs included by historical song, can count third participle with The correspondence of types of songs may thereby determine that out the number that third participle occurs described in same types of songs.Example Such as, third participle occurs 10 times in First historical song, occurs 20 times in the second first historical song, third head history song Occur in song 5 times, occur 15 times in the 4th first historical song, have its corresponding types of songs, First history per song Song and the second historical song belong to RAP types, and third head historical songs belong to rock and roll music type, and the 4th song belongs to the rank of nobility The happy type of scholar so as to show that third participle occurs 30 times in RAP types, occurs 5 times, in jazz in rock and roll music type Occur 15 times in type.
It should be noted that above-mentioned third participle can be one arbitrarily chosen from multiple participles that the history lyrics include A participle, wherein, third participle is in order to which title and the first participle mentioned above and the second participle are mutually distinguished, and is had no successively suitable Restriction in other meanings such as sequence.
S404：The according to third occurrence number of the participle during participle and described in same types of songs The number that three participles occur determines that the third participle belongs to the possibility information of different types of songs.
It can include the history lyrics and types of songs, the participle included per song and the song in per head historical songs Types of songs it is corresponding, an identical participle may come from different historical songs, i.e., identical participle can be with Corresponding different types of songs.
Possibility information can represent the information of participle and the correlation degree of types of songs, and the form of numerical value may be used It is stored, wherein percentage may be used in numerical value.
For example, establishing typelib according to 15 first historical songs, there are 10 historical songs to belong to RAP classes in this 15 song Type, 5 head of residue belong to rock and roll music type, and participle " I " occurs 100 times altogether in this 15 song, wherein, in RAP types Occur 70 times in historical song, occur 30 times in the historical song of rock and roll music type, then can show that participle " I " belongs to RAP classes The possibility information of type is 70%, and the possibility information that participle " I " belongs to rock and roll music type is 30%.
S405：Participle according to being obtained in the history lyrics belongs to described in the possibility information foundation of different types of songs Typelib.
Third participle is that one optionally gone out in all participles that historical song includes segments, the processing procedure of remaining participle It is identical with the processing procedure of third participle, it can determine that all participles belong to different songs in historical song by S403 and S404 The possibility information of bent type, it is possible thereby to which establishing out the participle obtained in the preservation history lyrics belongs to different types of songs Possibility information typelib.After the completion of typelib is established, it can be stored in server of dubbing in background music.
The foundation of typelib so that before above-mentioned preliminary screening is carried out, server of dubbing in background music can be dependent in typelib The participle of preservation belongs to the possibility information of different types of songs, directly determines the types of songs belonging to the lyrics to be dubbed in background music, letter The step of having changed determining types of songs further improves the efficiency dubbed in background music automatically.
The lyrics to be dubbed in background music illustrated in above example from composition form or granularity of division, can have participle, The two levels of word segment, and the embodiment of the present invention is provided dubs in background music generation method in addition to can be applied to this kind of wait to dub in background music The lyrics or this granularity of division for treating the lyrics of dubbing in background music, can also be applied to other composition forms the lyrics to be dubbed in background music or Situation applied to other granularity of division for treating the lyrics of dubbing in background music.Next, will be directed to the embodiment of the present invention how be applied to can The lyrics to be dubbed in background music of energy composition form or possible granularity of division illustrate.
Fig. 5 is a kind of method flow diagram of the generation method of dubbing in background music of lyrics provided in an embodiment of the present invention, applied to dubbing in background music Server, the server of dubbing in background music are included with music storehouse, first correspondence that segment and segment of dubbing in background music are included with music storehouse, In any one segment tool there are one the set of segments of dubbing in background music of at least one segment of dubbing in background music is included, herein in relation to server of dubbing in background music The explanation of correlated characteristic may refer to Fig. 1, and the associated description in embodiment corresponding to Fig. 2, which is not described herein again.
The method includes：
S501：It obtains and carries out the obtained at least one segment of word segmentation processing by treating the lyrics of dubbing in background music.
According to the granularity of division of word segmentation processing or the composition form of the lyrics to be dubbed in background music, can be obtained by the lyrics to be dubbed in background music To at least one segment, segment described here is related to granularity of division, can also be related to composition form.For example, dividing When granularity is thinner, which can be the form of participle, and when granularity of division is thicker, which can be the shape of word segment Formula.Such as when having less punctuate in the lyrics to be dubbed in background music, and not having specific paragraph relationship, which can be word segment Form, when punctuate is more in the lyrics to be dubbed in background music, and when paragraph divides clear, which can be the form of participle.Specifically will The lyrics to be dubbed in background music are divided into the segment of which kind of form, can accordingly be adjusted according to scene demand, computational accuracy etc., the present invention is to this It does not limit.
It should be noted that the lyrics to be dubbed in background music are divided into the segment of which kind of form also with matching in music storehouse in server of dubbing in background music The form of the segment stored is identical, thus could match segment of dubbing in background music by the segment marked off and the first correspondence Set.Such as with the segment stored in music storehouse be participle form, that lyrics to be dubbed in background music need by word segmentation processing obtain to A few participle.If both having included the form of participle with the segment stored in music storehouse, also include the form of word segment, that, which is treated, matches The segment that music and songs word segments can be related to above-mentioned granularity of division or lyrics composition form.
S502：With music storehouse according at least one segment search, institute is determined by first correspondence State the corresponding set of segments of dubbing in background music of at least one segment.
Since the segment marked off is identical with the segment preserved in music storehouse, therefore can realize through the first correspondence It determines each to divide the corresponding set of segments of dubbing in background music of segment.
S503：It is true from the corresponding set of segments of dubbing in background music of at least one segment according to the lyrics to be dubbed in background music Make the corresponding segment of dubbing in background music of each segment at least one segment.
S504：The lyrics to be dubbed in background music described in segment generation of dubbing in background music determined by splicing are dubbed in background music.
Since set of segments of dubbing in background music includes at least one segment of dubbing in background music, therefore each segment in the lyrics to be dubbed in background music can be passed through Component relationship determines the corresponding segment of dubbing in background music of each segment, specific method of determination and spelling from set of segments of dubbing in background music The mode of connecing can be according to the different and different of the concrete form of segment.
As it can be seen that it is dubbed in background music by being preset in server of dubbing in background music including the first correspondence between the segment that segments and dub in background music Library, when to acquisition after the lyrics of dubbing in background music are segmented to obtain at least one segment, can utilize at least one segment from With determining the corresponding set of segments of dubbing in background music of at least one segment according to the first correspondence, and treat according to this in music storehouse It dubs in background music the lyrics, will go out this by fragment assembly of determining to dub in background music from set of segments of dubbing in background music the lyrics to be dubbed in background music are corresponding to dub in background music, it can See, due to the corresponding segment of dubbing in background music of the segment that prestored, therefore can match that segment is corresponding to match after lyrics participle of dubbing in background music is treated Happy segment is waited to match so as to automatically generate out using the position of segment and segment in the lyrics to be dubbed in background music of dubbing in background music matched Music and songs word is dubbed in background music, and effectively increases the efficiency dubbed in background music for the lyrics.
Next by by the possibility form of segment, further the embodiment of the present invention is illustrated.
The first possible form, segment can be participles：
In this case, since participle is about the length of a word or a phrase in itself, therefore independent one segments Possibility as the lyrics to be dubbed in background music is little, generally requires at least multiple participles and forms the lyrics to be dubbed in background music.Therefore it is participle in segment In the case of, it can obtain by carrying out the obtained multiple participles of word segmentation processing to the lyrics to be dubbed in background music.
During the dubbing in background music of the lyrics to be dubbed in background music is spliced, position that can be according to participle in the lyrics to be dubbed in background music, directly Segment of dubbing in background music is corresponded to using the participle having determined to be spliced, to be spliced into, the lyrics to be dubbed in background music are corresponding to dub in background music；It can also be In the case that the lyrics to be dubbed in background music also have this level of word segment, after first showing that the son of word segment is dubbed in background music, further according to word Position of the segment in the lyrics to be dubbed in background music, being dubbed in background music by son is spliced into that the lyrics to be dubbed in background music are corresponding to dub in background music.
Also there is this level of word segment for the lyrics to be dubbed in background music, which includes at least one text Word slice section, a word segment includes at least one participle at least one word segment.
Determine participle it is corresponding dub in background music segment when, can according to it is multiple participle it is corresponding dub in background music set of segments with And the participle included by least one word segment, determine that the corresponding son of at least one word segment is dubbed in background music, The son that the corresponding segment of dubbing in background music of participle is obtained corresponding to this word segment wherein as included by a word segment is dubbed in background music. Specific method of determination may refer to the associated description in embodiment corresponding to Fig. 2, and which is not described herein again.
After obtaining the corresponding son of word segment and dubbing in background music, can respectively it be corresponded to by splicing at least one word segment The son lyrics to be dubbed in background music described in generation of dubbing in background music dub in background music.
Specific connecting method may refer to the associated description in embodiment corresponding to Fig. 2, and which is not described herein again.
Second of possible form, segment can be word segment：
Since word segment can include the content of certain length, therefore being possible to a word segment can treat as one Dub in background music the lyrics, in this case, when find this word segment it is corresponding dub in background music set of segments when, can will dub in background music segment collection Determined in conjunction dub in background music segment directly as this it is to be determined dub in background music dub in background music.
It, can be first from the corresponding segment collection of dubbing in background music of each word segment when the lyrics of dubbing in background music mark off multiple word segments Corresponding segment of dubbing in background music is determined in conjunction, further according to these positions of word segment in the lyrics to be dubbed in background music, is determined by splicing Go out that the lyrics to be dubbed in background music are corresponding to dub in background music.
Next the acquisition methods of dubbing in background music for the lyrics that the embodiment of the present invention is provided, the interaction end are discussed in detail from interaction end Can dub in background music relevant smart machine with the lyrics, the interaction and the friendship between user of the server that can realize and dub in background music Mutually.Fig. 6 is a kind of method flow diagram of the acquisition methods of dubbing in background music of lyrics provided in an embodiment of the present invention, and this method includes：
S601：The acquired lyrics to be dubbed in background music are sent to server of dubbing in background music, the lyrics to be dubbed in background music include at least one text Word slice section, a word segment includes at least one participle at least one word segment, and the server of dubbing in background music includes Described to include participle and the first correspondence of segment of dubbing in background music with music storehouse with music storehouse, wherein there are one wrap for any one participle tool Include the set of segments of dubbing in background music of at least one segment of dubbing in background music.
Interaction end can get the lyrics to be dubbed in background music, and the form for obtaining the lyrics to be dubbed in background music is not done in embodiments of the present invention Limit, can be by the lyrics to be dubbed in background music be manually entered or phonetic entry to the interaction end in.
Interaction end is got after the lyrics of dubbing in background music, and can the lyrics to be dubbed in background music directly be sent to server of dubbing in background music.Also may be used Be receiving dub in background music server transmission for obtain wait the request for the lyrics of dubbing in background music when, which is sent to and is matched Happy server.
S602：From the server of dubbing in background music obtain described in the lyrics to be dubbed in background music are corresponding dubs in background music, described dub in background music is dubbed in background music to be described Fragment assembly of dubbing in background music in set of segments of dubbing in background music corresponding to participle in server lyrics to be dubbed in background music according to obtains, described to wait to match Set of segments of dubbing in background music corresponding to participle in music and songs word is matched for the server of dubbing in background music according to being searched first correspondence What music storehouse obtained.
The explanation of feature may refer to Fig. 1 in embodiment corresponding to Fig. 6, the associated description of embodiment corresponding to Fig. 2, here No longer repeat one by one.
In embodiments of the present invention, interaction end can realize the relevant treatment to dubbing in background music, and match for example, this is presented to user It is happy.Therefore server of dubbing in background music determine this wait the lyrics of dubbing in background music are corresponding dub in background music after, directly this can be dubbed in background music and be sent to interaction It holds or in the request dubbed in background music for acquisition for receiving interaction end transmission, by the corresponding hair of dubbing in background music of the lyrics to be dubbed in background music Give interaction end.
The lyrics to be dubbed in background music determined by server of dubbing in background music are corresponding dub in background music can there are one or it is multiple, for having This of acquisition can be dubbed in background music as undetermined and dubbed in background music by multiple situations about dubbing in background music, interaction end, from it is multiple it is undetermined dub in background music in select symbol Close the dubbing in background music as the lyrics to be dubbed in background music undetermined of dubbing in background music of requirement.Optionally, it is treated described in being obtained from the server of dubbing in background music The lyrics of dubbing in background music it is corresponding it is multiple it is undetermined dub in background music, undetermined dub in background music carries soundtrack information；It undetermined is dubbed in background music according to the multiple Soundtrack information selects the dubbing in background music as the lyrics to be dubbed in background music undetermined of dubbing in background music for meeting demand.
Soundtrack information can be for identifying the information undetermined dubbed in background music, and there is its corresponding letter of dubbing in background music in each Ledu undetermined of matching Breath, the soundtrack information can embody the correlated characteristic undetermined dubbed in background music to a certain extent.Specifically, soundtrack information can wrap Include the scoring undetermined dubbed in background music for carrying the soundtrack information and/or type of dubbing in background music.
Scoring can be used to indicate that undetermined dub in background music meets the degree of music principle, can be dependent in undetermined dub in background music piece of dubbing in background music Smooth degree between section obtains.Scoring is higher, illustrates that undetermined dub in background music more meets music principle, i.e. the tune undetermined dubbed in background music Can be more melodious, the better hearing enjoying of user can be brought.
Type of dubbing in background music can be used to indicate that affiliated types of songs undetermined of dubbing in background music, for example, it is undetermined dub in background music belong to rock music or Person's jazz etc..
Select to meet demand it is undetermined dub in background music as described wait the lyrics of dubbing in background music when dubbing in background music, can there are many mode, the A kind of mode can be selected using the scoring that soundtrack information includes as foundation, can be the highest work undetermined of dubbing in background music that will score For the lyrics to be dubbed in background music dub in background music or using score it is higher it is undetermined dub in background music as the lyrics to be dubbed in background music dub in background music or be can To set default value, using the dubbing in background music as the lyrics to be dubbed in background music undetermined of dubbing in background music higher than default value of scoring.
Type that the second way can be included using soundtrack information dub in background music is according to being selected, for example, user can be with According to multiple type of dubbing in background music undetermined dubbed in background music, select that like a type of is undetermined to dub in background music, using this it is undetermined dub in background music as The lyrics to be dubbed in background music are dubbed in background music.
Type that the third mode can be included with soundtrack information dub in background music and scoring are according to being selected, for example, right In same type of dubbing in background music it is undetermined dub in background music have multiple when, then multiple middle scoring undetermined of dubbing in background music highest can be treated as described in Dub in background music the lyrics dub in background music or for identical scoring it is undetermined dub in background music have multiple when, then user can be according to multiple undetermined The type of dubbing in background music dubbed in background music selects a type of the dubbing in background music as the lyrics to be dubbed in background music undetermined of dubbing in background music liked.
In view of it is possible that corresponding dub in background music of the lyrics to be dubbed in background music determined by server of dubbing in background music does not meet demand Situation for example, user wants to obtain rock genre and dub in background music, but does not have in the type of dubbing in background music dubbed in background music determined of server of dubbing in background music There is dubbing in background music for rock and roll class.In this case, what the server that illustrates to dub in background music was determined dubs in background music there are problem, needs server of dubbing in background music It re-starts determining.
Therefore, for acquisition dub in background music do not meet demand in the case of, interaction end can send anti-to the server of dubbing in background music Feedforward information, so that server lyrics to be dubbed in background music according to regenerating the feedback information are dubbed in background music.
Feedback information can be intended to indicate that the information for the demand that do not meet of dubbing in background music.In order to enable dubbing in background music for regenerating can Preferably meet demand, can include describing the information of the demand in the feedback information, for example, user wants to obtain rock and roll Class is dubbed in background music, and demand is that rock and roll class is dubbed in background music, then the information can be carried in feedback information.
It is more in line with by the feedback operation corresponding dub in background music of the lyrics to be dubbed in background music that server determines that can dub in background music The demand of user further improves the accuracy dubbed in background music for the lyrics to be dubbed in background music.
Fig. 7 is a kind of structure drawing of device of the generating means of dubbing in background music of lyrics provided in an embodiment of the present invention, applied to dubbing in background music Server, the server of dubbing in background music are included with music storehouse, first correspondence including participle and segment of dubbing in background music with music storehouse, In any one participle tool include acquiring unit there are one the set of segments of dubbing in background music of at least one segment of dubbing in background music, described device is included 701st, searching unit 702, determination unit 703 and generation unit 704：
The acquiring unit 701, for obtaining by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing, The lyrics to be dubbed in background music include at least one word segment, and a word segment is included at least at least one word segment One participle；
The searching unit 702, for, with music storehouse, utilizing described first to correspond to according to the multiple participle lookup and closing System determines the corresponding set of segments of dubbing in background music of each participle；
The determination unit 703, for according to the multiple participle corresponding dub in background music set of segments and described at least one Participle included by a word segment determines that the corresponding son of at least one word segment is dubbed in background music；
The generation unit 704 splices the corresponding son of at least one word segment and dubs in background music generation for passing through The lyrics to be dubbed in background music are dubbed in background music.
Optionally, the first word segment is one at least one word segment, and the determination unit is specifically used Corresponding set of segments of dubbing in background music is segmented included by the first word segment in obtaining；It is wrapped according in the first word segment The neighbouring relations of participle are included, determine selected from the corresponding set of segments of dubbing in background music of participle included by the first word segment match Smooth degree between happy segment；Smooth degree is met the one of preset condition and assembles happy segment as the first word segment Son dub in background music.
Optionally, the first participle and second is segmented as the adjacent participle in the first word segment, the determination unit Including computation subunit, obtain subelement and as subelement：
The computation subunit, for calculating in set of segments of dubbing in background music corresponding to the first participle in the first word segment The segment of dubbing in background music dubbed in background music in set of segments corresponding to segment and the second participle of dubbing in background music between splicing degree, obtain described first At least one splicing score between participle and the second participle；
It is described to obtain subelement, for institute to be calculated according to the splicing score of adjacent participle in the first word segment State at least one splicing total score of the first word segment；
It is described to be used as subelement, for being chosen from the splicing total score more than predetermined threshold value corresponding to a splicing total score One assembles happy segment dubs in background music as the son of the first word segment.
Optionally, the computation subunit is specifically used for obtaining one in set of segments of dubbing in background music corresponding to the first participle A segment of dubbing in background music obtains a segment of dubbing in background music in set of segments of dubbing in background music corresponding to second participle；The two are calculated to dub in background music Splicing degree between segment obtains the spelling determined between the first participle and the second participle by the two segments of dubbing in background music Connect score.
Optionally, the server of dubbing in background music further includes typelib, and the typelib includes participle and belongs to different song classes The possibility information of type, it is described to further include the second correspondence dubbed in background music between segment and types of songs with music storehouse；Described device Further include determining types of songs unit：
The determining types of songs unit for searching the typelib according to the multiple participle, determines described to wait to match Types of songs belonging to music and songs word；
The searching unit includes screening subelement and determination subelement：
The screening subelement for according to second correspondence, is treated with described from described with being filtered out in music storehouse The corresponding segment of dubbing in background music of the types of songs belonging to the lyrics of dubbing in background music；
The determination subelement, for according to first correspondence, the song belonging to the lyrics to be dubbed in background music The corresponding set of segments of dubbing in background music of the multiple participle is determined in the corresponding segment of dubbing in background music of type.
Optionally, described device, which further includes, obtains historical song unit, participle unit, statistic unit, determines information unit With establish unit：
The acquisition historical song unit, for obtaining historical song, the historical song includes the history lyrics, history is matched Happy types of songs；
The participle unit, for being segmented to the history lyrics；
The statistic unit, for counting the occurrence number and third participle that third segments during participle With the correspondence of the types of songs of affiliated historical song, segmented out so that it is determined that going out the third described in same types of songs Existing number, the third participle are the participle obtained from the history lyrics；
The determining information unit, for according to third occurrence number of the participle during participle and same The number that the participle of third described in types of songs occurs determines that the third participle belongs to the possibility letter of different types of songs Breath；
It is described to establish unit, for belonging to the possibility of different types of songs according to the participle obtained in the history lyrics Information establishes the typelib.
Optionally, described device, which further includes, obtains historical song unit, participle unit, determines dub in background music slice unit and foundation Unit：
The acquisition historical song unit, for obtaining historical song, the historical song includes the history lyrics, history is matched Happy types of songs；
The participle unit, for being segmented to the history lyrics；
It is described to determine slice unit of dubbing in background music, for determining corresponding in the history is dubbed in background music dub in background music according to obtained participle Segment；
It is described to establish unit, for the first corresponding pass of segment of dubbing in background music in being dubbed in background music according to obtained participle with the history System and the history dub in background music in dubbing in background music segment and types of songs the second correspondence establish it is described with music storehouse.
Optionally, the acquiring unit is specifically used for carrying out word segmentation processing to the lyrics to be dubbed in background music of acquisition, obtains multiple points Word.
The explanation of the above-mentioned correlated characteristic about server of dubbing in background music may refer to Fig. 1, the correlation in embodiment corresponding to Fig. 2 Description, which is not described herein again.
As it can be seen that be preset in server of dubbing in background music including between the segment that segments and dub in background music the first correspondence with music storehouse, when After obtaining the multiple participles carried out acquired in word segmentation processing by treating the lyrics of dubbing in background music, multiple participle can be utilized from music storehouse It is middle that the corresponding set of segments of dubbing in background music of multiple participle is determined according to the first correspondence, later with the text of the lyrics to be dubbed in background music Word slice section is unit, and using corresponding set of segments of dubbing in background music is segmented in a word segment, it is right to be combined into a word segment institute The son answered is dubbed in background music, and the son of word segment is dubbed in background music and is spliced into dubbing in background music for the lyrics to be dubbed in background music, it is seen then that due to the participle pair that prestored The segment of dubbing in background music answered, therefore the corresponding segment of dubbing in background music of participle can be matched after lyrics participle of dubbing in background music is treated, so as to utilize The position of segment and participle in the lyrics to be dubbed in background music of dubbing in background music matched automatically generates out dubbing in background music for the lyrics to be dubbed in background music, and effectively carries The high efficiency dubbed in background music for the lyrics.
Fig. 8 is a kind of structure drawing of device of the generating means of dubbing in background music of lyrics provided in an embodiment of the present invention, applied to dubbing in background music Server, the server of dubbing in background music are included with music storehouse, first correspondence that segment and segment of dubbing in background music are included with music storehouse, In any one segment tool include acquiring unit there are one the set of segments of dubbing in background music of at least one segment of dubbing in background music, described device is included 801st, searching unit 802, determination unit 803 and concatenation unit 804：
The acquiring unit 801, it is obtained at least one by treating lyrics progress word segmentation processing of dubbing in background music for obtaining Segment；
The searching unit 802 for matching music storehouse according at least one segment search, passes through described first pair It should be related to and determine the corresponding set of segments of dubbing in background music of at least one segment；
The determination unit 803 for the lyrics to be dubbed in background music according to, is matched from least one segment is corresponding The corresponding segment of dubbing in background music of each segment at least one segment is determined in happy set of segments；
The concatenation unit 804 splices dubbing in background music for the lyrics to be dubbed in background music described in segment generation of dubbing in background music determined for passing through.
Optionally, the acquiring unit is specifically used for obtaining by obtained by carrying out the lyrics to be dubbed in background music word segmentation processing Multiple participles.
Optionally, the segment is word segment, and the lyrics to be dubbed in background music include at least one word segment.
Optionally, the determination unit be specifically used for according to it is the multiple participle it is corresponding dub in background music set of segments and Participle included by least one word segment determines that the corresponding son of at least one word segment is dubbed in background music, The son that the corresponding segment of dubbing in background music of participle is obtained corresponding to this word segment wherein as included by a word segment is dubbed in background music；
The concatenation unit is specifically used for dubbing in background music generation by splicing the corresponding son of at least one word segment The lyrics to be dubbed in background music are dubbed in background music.
The explanation of the above-mentioned correlated characteristic about server of dubbing in background music may refer to Fig. 1, the correlation in embodiment corresponding to Fig. 2 Description, which is not described herein again.
As it can be seen that it is dubbed in background music by being preset in server of dubbing in background music including the first correspondence between the segment that segments and dub in background music Library, when to acquisition after the lyrics of dubbing in background music are segmented to obtain at least one segment, can utilize at least one segment from With determining the corresponding set of segments of dubbing in background music of at least one segment according to the first correspondence, and treat according to this in music storehouse It dubs in background music the lyrics, will go out this by fragment assembly of determining to dub in background music from set of segments of dubbing in background music the lyrics to be dubbed in background music are corresponding to dub in background music, it can See, due to the corresponding segment of dubbing in background music of the segment that prestored, therefore can match that segment is corresponding to match after lyrics participle of dubbing in background music is treated Happy segment is waited to match so as to automatically generate out using the position of segment and segment in the lyrics to be dubbed in background music of dubbing in background music matched Music and songs word is dubbed in background music, and effectively increases the efficiency dubbed in background music for the lyrics.
Fig. 9 is a kind of structure drawing of device of the acquisition device of dubbing in background music of lyrics provided in an embodiment of the present invention, applied to interaction End, described device include transmitting element 901 and acquiring unit 902：
The transmitting element 901, for sending the acquired lyrics to be dubbed in background music, the lyrics to be dubbed in background music to server of dubbing in background music Including at least one word segment, a word segment includes at least one participle at least one word segment, described Server of dubbing in background music is included with music storehouse, described to include participle and the first correspondence of segment of dubbing in background music with music storehouse, wherein any one There are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music for participle tool；
The acquiring unit 902, it is described for the lyrics to be dubbed in background music to be corresponding dubs in background music described in being obtained from the server of dubbing in background music It dubs in background music and is spelled for the segment of dubbing in background music in set of segments of dubbing in background music corresponding to participle in server lyrics to be dubbed in background music according to of dubbing in background music It connects to obtain, set of segments of dubbing in background music corresponding to participle in the lyrics to be dubbed in background music is corresponded to for the server of dubbing in background music according to described first What relationship obtained described in searching with music storehouse.
Optionally, the acquiring unit includes obtaining subelement and selection subelement：
The acquisition subelement multiple undetermined is matched for the lyrics to be dubbed in background music described in being obtained from the server of dubbing in background music to be corresponding Happy, undetermined dub in background music carries soundtrack information；
The selection subelement, for according to the multiple soundtrack information undetermined dubbed in background music, selection to meet the undetermined of demand It dubs in background music dubbing in background music as the lyrics to be dubbed in background music.
Optionally, the soundtrack information includes carrying the scoring undetermined dubbed in background music of the soundtrack information and/or type of dubbing in background music.
Optionally, described device further includes feedback unit：
If what is obtained dubs in background music and do not meet demand, the feedback unit is triggered, the feedback unit, for the clothes of dubbing in background music Device of being engaged in sends feedback information, so that server lyrics to be dubbed in background music according to regenerating the feedback information are matched It is happy.
Optionally, the feedback information includes describing the information of the demand.
The explanation of the above-mentioned correlated characteristic about server of dubbing in background music may refer to Fig. 1, the correlation in embodiment corresponding to Fig. 2 Description, which is not described herein again.
As it can be seen that interaction end can be realized and dub in background music server interaction and the interaction between user, pass through interaction end Feedback corresponding dub in background music of the lyrics to be dubbed in background music that server is determined that can to dub in background music be more in line with the demand of user, further Improve the accuracy dubbed in background music for the lyrics to be dubbed in background music.
Figure 10 be a kind of system construction drawing of the scoring system of the lyrics provided in an embodiment of the present invention, the scoring system packet Include dub in background music server 1001 and interaction end 1002：
The server 1001 of dubbing in background music is included with music storehouse, and described corresponded to first of music storehouse including participle and segment of dubbing in background music is closed There are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music, the servers of dubbing in background music for system, wherein any one participle tool 1001, for obtaining by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing, the lyrics to be dubbed in background music include At least one word segment, a word segment includes at least one participle at least one word segment；According to described Multiple participles match music storehouse described in searching, and the corresponding segment collection of dubbing in background music of each participle is determined using first correspondence It closes；According to the multiple participle corresponding dub in background music set of segments and the participle included by least one word segment, really The corresponding son of at least one word segment is made to dub in background music；It is corresponded to respectively by splicing at least one word segment The son lyrics to be dubbed in background music described in generation of dubbing in background music dub in background music；
The interaction end 1002, for the server of dubbing in background music send it is acquired described in the lyrics to be dubbed in background music；From described The lyrics to be dubbed in background music are corresponding described in server of dubbing in background music acquisition dubs in background music.
It a kind of is compiled next, introducing with implementing the lyrics that are provided of the embodiment of the present invention relevant music of generating mode of dubbing in background music Device is collected, which can obtain the lyrics to be dubbed in background music, can also be corresponding to the lyrics to be dubbed in background music for server offer of dubbing in background music It dubs in background music into edlin.
The music editor can include having inputting the input interface of the lyrics to be dubbed in background music and for obtaining for opening up Show the editing interface dubbed in background music.
For the input interface for obtaining the lyrics to be dubbed in background music, the acquired lyrics to be dubbed in background music can be in the input interface The lyrics of input or the lyrics to be obtained after the voice of input is identified.
That is, user when using the music editor, can be manually entered music and songs to be matched in the input interface Word or the lyrics to be dubbed in background music having from other copy editor's tool, texts are imported by the input interface.Work as configuration When the smart machine for having the music editor has voice typing identification function, user can have the music and songs to be matched by broadcasting The audio file of word directly says the lyrics to be dubbed in background music, and the audio content of typing can be identified in smart machine, from In identify the lyrics to be dubbed in background music, and the input interface can obtain the lyrics to be dubbed in background music identified.Abundant obtaining is waited to dub in background music The mode of the lyrics improves the scope of application of the music editor.
Input interface is being got after the lyrics of dubbing in background music, since the music editor can be connected with server of dubbing in background music, therefore The lyrics to be dubbed in background music can be sent to server of dubbing in background music when needed.User can be by the work(that is set on the music editor Energy key generation output order, to indicate that the music editor sends the lyrics to be dubbed in background music to server of dubbing in background music.
When obtaining output order, the music editor is used to send to server of dubbing in background music and be obtained by the input interface The lyrics to be dubbed in background music taken, the server of dubbing in background music are included with music storehouse, described the first couple for including participle and segment of dubbing in background music with music storehouse It should be related to, wherein there are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music for any one participle tool.
The lyrics to be dubbed in background music, correlated characteristic with happy server of dubbing in background music explanation may refer to Fig. 1, it is real corresponding to Fig. 2 The associated description in example is applied, which is not described herein again.
When from it is described dub in background music server obtain described in wait the lyrics of dubbing in background music are corresponding dub in background music when, the music editor be used for will Corresponding dub in background music of the lyrics to be dubbed in background music is illustrated in the editing interface, and corresponding dub in background music of the lyrics to be dubbed in background music is matched to be described Fragment assembly of dubbing in background music in set of segments of dubbing in background music corresponding to participle in happy server lyrics to be dubbed in background music according to obtains, described to treat The set of segments of dubbing in background music corresponding to participle in the lyrics of dubbing in background music is dubbed in background music for the server of dubbing in background music according to being searched the first correspondence What library obtained.
The editing interface is dubbed in background music by displaying, can enable a user to observe the specifying information dubbed in background music, such as sound Symbol, beat, time shaft etc., and can be in this, as the editor basis dubbed in background music.It should be noted that the editing interface is in addition to can With the dubbing in background music by server transmission of dubbing in background music of displaying, it can also show that other are dubbed in background music, such as by knowing to institute's recording audio data What is do not obtained dubs in background music.
In order to realize the editor to dubbing in background music, optionally, the editing interface has Edit button, when receiving to institute When stating the triggering of Edit button, the editing interface is used to dubbing in background music of being shown being set as editable.In editable state Under, this, which is dubbed in background music, the edit operations such as can be deleted, increase, changing, to be finally reached the demand of user.
As it can be seen that by music editor the lyrics to be dubbed in background music can be obtained by input interface, and can will the lyrics be dubbed in background music Output to server of dubbing in background music, when receive server return of dubbing in background music for this wait the lyrics of dubbing in background music when dubbing in background music, can be by compiling It shows that this is dubbed in background music in editing interface, allows and dub in background music with editable possibility, improve user experience.
It a kind of is broadcast next, introducing with implementing the lyrics that are provided of the embodiment of the present invention relevant music of generating mode of dubbing in background music Device is put, which can obtain that the lyrics to be dubbed in background music are corresponding with its to dub in background music, and the two is mixed into song plays out.
The music player has mixed interface and broadcast interface.
The mixed interface is for the lyrics to be dubbed in background music and corresponding dub in background music of the lyrics to be dubbed in background music to be mixed to get Song, the lyrics to be dubbed in background music are corresponding to dub in background music as piece of dubbing in background music corresponding to participle in server lyrics to be dubbed in background music according to of dubbing in background music Fragment assembly of dubbing in background music in Duan Jihe obtains, and set of segments is dubbed in background music as the clothes of dubbing in background music corresponding to participle in the lyrics to be dubbed in background music Business device searches what is obtained with music storehouse according to the first correspondence, and the server of dubbing in background music is included with music storehouse, described to include with music storehouse Segment and dub in background music the first correspondence of segment, and wherein there are one include matching at least one segment of dubbing in background music for any one participle tool Happy set of segments.
The lyrics to be dubbed in background music, correlated characteristic with happy server of dubbing in background music explanation may refer to Fig. 1, it is real corresponding to Fig. 2 The associated description in example is applied, which is not described herein again.
The broadcast interface is used to play the song.
As it can be seen that the mixed interface can mix the lyrics to be dubbed in background music obtained in advance and corresponding dub in background music, mixing The mode present invention does not limit, and the lyrics have been obtained after mixing as the lyrics to be dubbed in background music, and music is the song dubbed in background music, should playing During song, it can only play this and dub in background music, and the lyrics (music and songs to be matched corresponding with playing time shaft of dubbing in background music is shown in relevant position In word with the lyrics of time shaft corresponding position), the lyrics to be dubbed in background music can also be exported when broadcasting is dubbed in background music by simulated sound simultaneously Simulated sound.User can see while broadcasting is dubbed in background music or hear whether the corresponding lyrics coordinate with dubbing in background music as a result, improve User experience.
One of ordinary skill in the art will appreciate that：Realizing all or part of step of above method embodiment can pass through The relevant hardware of program instruction is completed, and foregoing routine can be stored in a computer read/write memory medium, which exists During execution, step including the steps of the foregoing method embodiments is performed；And aforementioned storage medium can be in following media at least one Kind：Read-only memory (English：Read-only memory, abbreviation：ROM), RAM, magnetic disc or CD etc. are various to store The medium of program code.
It should be noted that each embodiment in this specification is described by the way of progressive, each embodiment it Between just to refer each other for identical similar part, the highlights of each of the examples are difference from other examples. For equipment and system embodiment, since it is substantially similar to embodiment of the method, so describe fairly simple, The relevent part can refer to the partial explaination of embodiments of method.Equipment and system embodiment described above is only schematic , wherein may or may not be as the unit that separating component illustrates physically separate, shown as unit Component may or may not be physical unit, you can be located at a place or can also be distributed to multiple networks On unit.Some or all of module therein can be selected according to the actual needs to realize the purpose of this embodiment scheme. Those of ordinary skill in the art are without creative efforts, you can to understand and implement.
The foregoing is only a preferred embodiment of the present invention, but protection scope of the present invention be not limited thereto, Any one skilled in the art in the technical scope disclosed by the present invention, the change or replacement that can be readily occurred in, It should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with scope of the claims Subject to.

Claims (31)

1. the generation method of dubbing in background music of a kind of lyrics, which is characterized in that applied to server of dubbing in background music, the server of dubbing in background music includes matching Music storehouse, described to include participle and the first correspondence of segment of dubbing in background music with music storehouse, wherein there are one include any one participle tool The set of segments of dubbing in background music of at least one segment of dubbing in background music, the method includes：
It obtains by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing, the lyrics to be dubbed in background music include at least one A word segment, a word segment includes at least one participle at least one word segment；
With music storehouse according to being searched the multiple participle, determine that each participle corresponds to respectively using first correspondence Set of segments of dubbing in background music；
According to the multiple participle corresponding dub in background music set of segments and the participle included by least one word segment, really The corresponding son of at least one word segment is made to dub in background music；
It is dubbed in background music by splicing at least one word segment corresponding son lyrics to be dubbed in background music described in generation of dubbing in background music.
2. according to the method described in claim 1, it is characterized in that, the first word segment is at least one word segment One, it is described corresponding to be dubbed in background music included by set of segments and at least one word segment according to the multiple participle Participle, determines that the corresponding son of at least one word segment is dubbed in background music, including：
Obtain the corresponding set of segments of dubbing in background music of the included participle of the first word segment；
According to the neighbouring relations of included participle in the first word segment, determine from included by the first word segment points Smooth degree in the corresponding set of segments of dubbing in background music of word between selected segment of dubbing in background music；
It smooth degree is met the one of preset condition assembles happy segment and dub in background music as the son of the first word segment.
3. according to the method described in claim 2, it is characterized in that, the first participle and the second participle are the first word segment In adjacent participle, the neighbouring relations according to included participle in the first word segment, determine from described first text Smooth degree in the corresponding set of segments of dubbing in background music of the included participle of word slice section between selected segment of dubbing in background music, including：
Calculate dub in background music segment and the second participle institute in set of segments of dubbing in background music corresponding to the first participle in the first word segment The splicing degree between segment of dubbing in background music in corresponding set of segments of dubbing in background music, obtains between the first participle and the second participle extremely Few splicing score；
At least the one of the first word segment is calculated according to the splicing score of adjacent participle in the first word segment A splicing total score；
It is described smooth degree met the one of predetermined threshold value assemble happy segment dub in background music as the son of the first word segment, wrap It includes：
One corresponding to a splicing total score is chosen from the splicing total score more than predetermined threshold value and assembles happy segment as described the The son of one word segment is dubbed in background music.
4. according to the method described in claim 3, it is characterized in that, described calculate first participle institute in the first word segment The spelling that dubbing in background music in corresponding set of segments of dubbing in background music is dubbed in background music corresponding to segment and the second participle between the segment of dubbing in background music in set of segments Degree is connect, obtains at least one splicing score between the first participle and the second participle, including：
A segment of dubbing in background music in set of segments of dubbing in background music corresponding to the first participle is obtained, is obtained corresponding to second participle A segment of dubbing in background music in set of segments of dubbing in background music；
The splicing degree between the two segments of dubbing in background music is calculated, obtains by the two being matched between the first participle and the second participle The splicing score that happy segment determines.
5. method according to any one of claims 1 to 4, which is characterized in that the server of dubbing in background music further includes type Library, the typelib includes the possibility information that participle belongs to different types of songs, described to further include segment of dubbing in background music with music storehouse The second correspondence between types of songs；Match music storehouse according to being searched the multiple participle described, determine described Before the corresponding set of segments of dubbing in background music of multiple participles, further include：
The typelib is searched according to the multiple participle, determines the types of songs belonging to the lyrics to be dubbed in background music；
It is described to match music storehouse according to being searched the multiple participle, determine the multiple participle using first correspondence Corresponding set of segments of dubbing in background music, including：
According to second correspondence, from the types of songs pair with belonging to filtering out in music storehouse with the lyrics to be dubbed in background music The segment of dubbing in background music answered；
According to first correspondence, determined from segment of dubbing in background music corresponding with the types of songs belonging to the lyrics to be dubbed in background music Go out the corresponding set of segments of dubbing in background music of the multiple participle.
6. it according to the method described in claim 5, it is characterized in that, further includes：
Historical song is obtained, the historical song includes the history lyrics, history matches happy types of songs；
The history lyrics are segmented；
Count the occurrence number and the song class of third participle and affiliated historical song that third segments during participle The correspondence of type, so that it is determined that go out the number that third participle occurs described in same types of songs, the third participle For the participle obtained from the history lyrics；
Occurrence number during participle is segmented according to the third and the third described in same types of songs segments out Existing number determines that the third participle belongs to the possibility information of different types of songs；
The possibility information that participle according to being obtained in the history lyrics belongs to different types of songs establishes the typelib.
7. it according to the method described in claim 5, it is characterized in that, further includes：
Historical song is obtained, the historical song includes the history lyrics, history matches happy types of songs；
The history lyrics are segmented；
Participle according to obtaining determines the corresponding segment of dubbing in background music in the history is dubbed in background music；
Match during participle according to obtaining dubs in background music the first correspondence of segment in dubbing in background music with the history and the history is dubbed in background music Second correspondence of happy segment and types of songs is established described with music storehouse.
8. according to the method described in claim 1, it is characterized in that, described obtain carries out word segmentation processing by treating the lyrics of dubbing in background music Acquired multiple participles, including：
Word segmentation processing is carried out to the lyrics to be dubbed in background music of acquisition, obtains multiple participles.
9. the generating means of dubbing in background music of a kind of lyrics, which is characterized in that applied to server of dubbing in background music, the server of dubbing in background music includes matching Music storehouse, described to include participle and the first correspondence of segment of dubbing in background music with music storehouse, wherein there are one include any one participle tool The set of segments of dubbing in background music of at least one segment of dubbing in background music, it is single that described device includes acquiring unit, searching unit, determination unit and generation Member：
The acquiring unit, it is described to treat for obtaining by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing The lyrics of dubbing in background music include at least one word segment, and a word segment includes at least one point at least one word segment Word；
The searching unit, for, with music storehouse, being determined according to the multiple participle lookup using first correspondence Go out the corresponding set of segments of dubbing in background music of each participle；
The determination unit, for according to corresponding set of segments and at least one letter plate of dubbing in background music of the multiple participle Participle included by section, determines that the corresponding son of at least one word segment is dubbed in background music；
The generation unit, for pass through splice the corresponding son of at least one word segment dub in background music generation described in wait to match Music and songs word is dubbed in background music.
10. device according to claim 9, which is characterized in that the first word segment is at least one word segment In one, the determination unit is specifically used for obtaining the corresponding segment collection of dubbing in background music of the included participle of the first word segment It closes；According to the neighbouring relations of included participle in the first word segment, determine from included by the first word segment points Smooth degree in the corresponding set of segments of dubbing in background music of word between selected segment of dubbing in background music；Smooth degree is met the one of preset condition The son that happy segment is assembled as the first word segment is dubbed in background music.
11. device according to claim 10, which is characterized in that the first participle and second is segmented as first letter plate Adjacent participle in section, the determination unit include computation subunit, obtain subelement and as subelement：
The computation subunit, for calculating matching in set of segments of dubbing in background music corresponding to the first participle in the first word segment The splicing degree between segment of dubbing in background music in set of segments of dubbing in background music corresponding to happy segment and the second participle, obtains the first participle And second participle between at least one splicing score；
It is described to obtain subelement, for being calculated described according to the splicing score of adjacent participle in the first word segment At least one splicing total score of one word segment；
It is described to be used as subelement, for choosing one group corresponding to a splicing total score from the splicing total score more than predetermined threshold value Segment of dubbing in background music is dubbed in background music as the son of the first word segment.
12. according to the devices described in claim 11, which is characterized in that the computation subunit is specifically used for obtaining described first A segment of dubbing in background music in set of segments of dubbing in background music corresponding to participle is obtained in set of segments of dubbing in background music corresponding to second participle One segment of dubbing in background music；The splicing degree between the two segments of dubbing in background music is calculated, is obtained between the first participle and the second participle The splicing score determined by the two segments of dubbing in background music.
13. according to the device described in claim 9-12 any one, which is characterized in that the server of dubbing in background music further includes type Library, the typelib includes the possibility information that participle belongs to different types of songs, described to further include segment of dubbing in background music with music storehouse The second correspondence between types of songs；Described device further includes determining types of songs unit：
The determining types of songs unit for searching the typelib according to the multiple participle, determines the music and songs to be matched Types of songs belonging to word；
The searching unit includes screening subelement and determination subelement：
The screening subelement for according to second correspondence, is waited to dub in background music with filtering out in music storehouse from described with described The corresponding segment of dubbing in background music of types of songs belonging to the lyrics；
The determination subelement, for according to first correspondence, the types of songs belonging to the lyrics to be dubbed in background music The corresponding set of segments of dubbing in background music of the multiple participle is determined in corresponding segment of dubbing in background music.
14. device according to claim 13, which is characterized in that described device, which further includes, to be obtained historical song unit, divides Word unit determines to dub in background music and slice unit and establishes unit：
The acquisition historical song unit, for obtaining historical song, the historical song includes the history lyrics, history with happy Types of songs；
The participle unit, for being segmented to the history lyrics；
It is described to determine slice unit of dubbing in background music, for determining the corresponding piece of dubbing in background music in the history is dubbed in background music according to obtained participle Section；
It is described to establish unit, for the first correspondence of segment of dubbing in background music in being dubbed in background music according to obtained participle with the history, with And the history dub in background music in dub in background music segment and types of songs the second correspondence establish it is described with music storehouse.
15. the generation method of dubbing in background music of a kind of lyrics, which is characterized in that applied to server of dubbing in background music, the server of dubbing in background music includes With music storehouse, first correspondence that segment and segment of dubbing in background music are included with music storehouse, wherein there are one wrap for any one segment tool The set of segments of dubbing in background music of at least one segment of dubbing in background music is included, the method includes：
It obtains and carries out the obtained at least one segment of word segmentation processing by treating the lyrics of dubbing in background music；
With music storehouse according at least one segment search, determined by first correspondence described at least one The corresponding set of segments of dubbing in background music of segment；
According to the lyrics to be dubbed in background music, determined from the corresponding set of segments of dubbing in background music of at least one segment it is described to The corresponding segment of dubbing in background music of each segment in a few segment；
The lyrics to be dubbed in background music described in segment generation of dubbing in background music determined by splicing are dubbed in background music.
16. according to the method for claim 15, which is characterized in that the segment is participle, and described obtain is matched by treating Music and songs word carries out the obtained at least one segment of word segmentation processing, including：
It obtains by carrying out the obtained multiple participles of word segmentation processing to the lyrics to be dubbed in background music.
17. according to the method for claim 15, which is characterized in that the segment be word segment, the lyrics to be dubbed in background music Including at least one word segment.
18. according to the method for claim 16, which is characterized in that the lyrics to be dubbed in background music include at least one letter plate Section, a word segment includes at least one participle at least one word segment, then music and songs to be matched described in the basis Word is determined each to segment at least one participle from least one segment in corresponding set of segments of dubbing in background music Corresponding segment of dubbing in background music, including：
According to corresponding set of segments and point included by least one word segment of dubbing in background music of the multiple participle Word determines that the corresponding son of at least one word segment is dubbed in background music, wherein the participle point as included by a word segment The son that not corresponding segment of dubbing in background music is obtained corresponding to this word segment is dubbed in background music；
The lyrics to be dubbed in background music described in segment generation of dubbing in background music determined by splicing are dubbed in background music, including：
It is dubbed in background music by splicing at least one word segment corresponding son lyrics to be dubbed in background music described in generation of dubbing in background music.
19. the generating means of dubbing in background music of a kind of lyrics, which is characterized in that applied to server of dubbing in background music, the server of dubbing in background music includes With music storehouse, first correspondence that segment and segment of dubbing in background music are included with music storehouse, wherein there are one wrap for any one segment tool The set of segments of dubbing in background music of at least one segment of dubbing in background music is included, described device includes acquiring unit, searching unit, determination unit and splicing Unit：
The acquiring unit carries out the obtained at least one segment of word segmentation processing for obtaining by treating the lyrics of dubbing in background music；
The searching unit for matching music storehouse according at least one segment search, passes through first correspondence Determine the corresponding set of segments of dubbing in background music of at least one segment；
The determination unit, for the lyrics to be dubbed in background music according to, from the corresponding segment of dubbing in background music of at least one segment The corresponding segment of dubbing in background music of each segment at least one segment is determined in set；
The concatenation unit splices dubbing in background music for the lyrics to be dubbed in background music described in segment generation of dubbing in background music determined for passing through.
20. device according to claim 19, which is characterized in that the determination unit is specifically used for according to the multiple point Word is corresponding to dub in background music set of segments and the participle included by least one word segment, determines described at least one The corresponding son of a word segment is dubbed in background music, wherein the corresponding segment of dubbing in background music of participle obtains as included by a word segment Son corresponding to this word segment is dubbed in background music；
The concatenation unit is specifically used for dubbing in background music described in generation by splicing the corresponding son of at least one word segment The lyrics to be dubbed in background music are dubbed in background music.
21. a kind of acquisition methods of dubbing in background music of lyrics, which is characterized in that applied to interaction end, the method includes：
The acquired lyrics to be dubbed in background music are sent to server of dubbing in background music, the lyrics to be dubbed in background music include at least one word segment, institute It states a word segment at least one word segment and includes at least one participle, the server of dubbing in background music is included with music storehouse, institute It states and includes participle and the first correspondence of segment of dubbing in background music with music storehouse, wherein any one participle tool is at least one there are one including The set of segments of dubbing in background music for segment of dubbing in background music；
From the server of dubbing in background music obtain described in the lyrics to be dubbed in background music are corresponding dubs in background music, it is described dub in background music for it is described dub in background music server according to Fragment assembly of dubbing in background music in set of segments of dubbing in background music corresponding to participle in the lyrics to be dubbed in background music obtains, and divides in the lyrics to be dubbed in background music Set of segments of dubbing in background music corresponding to the word server of dubbing in background music is obtained according to being searched first correspondence with music storehouse.
22. according to the method for claim 21, which is characterized in that described to wait to dub in background music described in server acquisition from described dub in background music The lyrics are corresponding to dub in background music, including：
From the server of dubbing in background music obtain described in the lyrics to be dubbed in background music it is corresponding it is multiple it is undetermined dub in background music, it is described it is undetermined dub in background music to carry match Happy information；
According to the multiple soundtrack information undetermined dubbed in background music, select to meet the undetermined of demand and dub in background music as the lyrics to be dubbed in background music It dubs in background music.
23. according to the method for claim 22, which is characterized in that the soundtrack information includes carrying the soundtrack information The scoring undetermined dubbed in background music and/or type of dubbing in background music.
24. according to the method for claim 21, which is characterized in that wait to match described in server acquisition from described dub in background music described Music and songs word is corresponding dub in background music after, further include：
If what is obtained dubs in background music and do not meet demand, feedback information is sent to the server of dubbing in background music so that the server according to The feedback information regenerates dubbing in background music for the lyrics to be dubbed in background music.
25. according to the method for claim 24, which is characterized in that the feedback information includes the letter for describing the demand Breath.
26. a kind of acquisition device of dubbing in background music of lyrics, which is characterized in that applied to interaction end, described device include transmitting element and Acquiring unit：
The transmitting element, for sending the acquired lyrics to be dubbed in background music to server of dubbing in background music, the lyrics to be dubbed in background music are included extremely Lack a word segment, a word segment includes at least one participle, the clothes of dubbing in background music at least one word segment Device be engaged in including with music storehouse, described to include participle and the first correspondence of segment of dubbing in background music with music storehouse, wherein any one segments tool There are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music；
The acquiring unit, for the lyrics to be dubbed in background music to be corresponding dubs in background music described in being obtained from the server of dubbing in background music, it is described dub in background music for Fragment assembly of dubbing in background music in set of segments of dubbing in background music corresponding to participle in server lyrics to be dubbed in background music according to of dubbing in background music obtains, Set of segments of dubbing in background music corresponding to participle in the lyrics to be dubbed in background music is looked into for the server of dubbing in background music according to first correspondence It is obtained described in looking for music storehouse.
27. device according to claim 26, which is characterized in that the acquiring unit includes obtaining subelement and selection Unit：
The acquisition subelement, for the lyrics to be dubbed in background music described in being obtained from the server of dubbing in background music it is corresponding it is multiple it is undetermined dub in background music, Undetermined dub in background music carries soundtrack information；
The selection subelement, for according to the multiple soundtrack information undetermined dubbed in background music, selection to meet the undetermined of demand and dubs in background music As dubbing in background music for the lyrics to be dubbed in background music.
28. a kind of scoring system of the lyrics, which is characterized in that the scoring system includes dub in background music server and interaction end：
The server of dubbing in background music is included with music storehouse, first correspondence including participle and segment of dubbing in background music with music storehouse, wherein There are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music, the servers of dubbing in background music, for obtaining for any one participle tool It takes by treating the multiple participles dubbed in background music acquired in lyrics progress word segmentation processing, the lyrics to be dubbed in background music include at least one word Segment, a word segment includes at least one participle at least one word segment；It is searched according to the multiple participle It is described to match music storehouse, determine the corresponding set of segments of dubbing in background music of each participle using first correspondence；According to described Multiple participles corresponding dub in background music set of segments and the participle included by least one word segment, determine it is described at least The corresponding son of one word segment is dubbed in background music；It is dubbed in background music generation by splicing the corresponding son of at least one word segment The lyrics to be dubbed in background music are dubbed in background music；
The interaction end, for the server of dubbing in background music send it is acquired described in the lyrics to be dubbed in background music；From the service of dubbing in background music The lyrics to be dubbed in background music are corresponding described in device acquisition dubs in background music.
29. a kind of music editor, which is characterized in that the music editor has inputs the lyrics to be dubbed in background music for obtaining Input interface and the editing interface dubbed in background music for displaying；
For obtaining the lyrics to be dubbed in background music, the acquired lyrics to be dubbed in background music input the input interface in the input interface The lyrics or the lyrics to be obtained after the voice of input is identified；
When obtaining output order, the music editor is used to send what is obtained by the input interface to server of dubbing in background music The lyrics to be dubbed in background music, the server of dubbing in background music are included with music storehouse, and described corresponded to first of music storehouse including participle and segment of dubbing in background music is closed There are one the set of segments of dubbing in background music for including at least one segment of dubbing in background music for system, wherein any one participle tool；
When from it is described dub in background music server obtain described in wait the lyrics of dubbing in background music are corresponding dub in background music when, the music editor be used for will described in Corresponding dub in background music of the lyrics to be dubbed in background music is illustrated in the editing interface, and the lyrics to be dubbed in background music are corresponding to dub in background music as the clothes of dubbing in background music Fragment assembly of dubbing in background music in set of segments of dubbing in background music corresponding to being engaged in segmenting in device lyrics to be dubbed in background music according to obtains, described to wait to dub in background music Set of segments of dubbing in background music corresponding to participle in the lyrics is obtained for the server of dubbing in background music according to being searched the first correspondence with music storehouse It arrives.
30. music editor according to claim 29, which is characterized in that the editing interface has Edit button, when When receiving the triggering to the Edit button, the editing interface is used to dubbing in background music of being shown being set as editable；
What the editing interface was shown dub in background music to further include is dubbed in background music by what is identified to institute's recording audio data.
31. a kind of music player, which is characterized in that the music player has mixed interface and broadcast interface；
The mixed interface for the lyrics to be dubbed in background music and corresponding dub in background music of the lyrics to be dubbed in background music to be carried out being mixed to get song, The lyrics to be dubbed in background music are corresponding to dub in background music as segment collection of dubbing in background music corresponding to participle in server lyrics to be dubbed in background music according to of dubbing in background music Fragment assembly of dubbing in background music in conjunction obtains, and set of segments is dubbed in background music as the server of dubbing in background music corresponding to participle in the lyrics to be dubbed in background music Search what is obtained with music storehouse according to the first correspondence, the server of dubbing in background music is included with music storehouse, described to include participle with music storehouse With the first correspondence of segment of dubbing in background music, wherein there are one the pieces of dubbing in background music for including at least one segment of dubbing in background music for any one participle tool Duan Jihe；
The broadcast interface is used to play the song.
CN201611264888.5A 2016-12-30 2016-12-30 Lyric score generation method and related device Active CN108268530B (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201611264888.5A CN108268530B (en) 2016-12-30 2016-12-30 Lyric score generation method and related device
TW106126946A TW201824249A (en) 2016-12-30 2017-08-09 Method for generating music to accompany lyrics and related apparatus
PCT/CN2017/117358 WO2018121368A1 (en) 2016-12-30 2017-12-20 Method for generating music to accompany lyrics and related apparatus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611264888.5A CN108268530B (en) 2016-12-30 2016-12-30 Lyric score generation method and related device

Publications (2)

Publication Number Publication Date
CN108268530A true CN108268530A (en) 2018-07-10
CN108268530B CN108268530B (en) 2022-04-29

Family

ID=62711084

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611264888.5A Active CN108268530B (en) 2016-12-30 2016-12-30 Lyric score generation method and related device

Country Status (3)

Country Link
CN (1) CN108268530B (en)
TW (1) TW201824249A (en)
WO (1) WO2018121368A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109166564A (en) * 2018-07-19 2019-01-08 平安科技（深圳）有限公司 For the method, apparatus and computer readable storage medium of lyrics text generation melody
CN109309863A (en) * 2018-08-01 2019-02-05 李春莲 Movie contents matching mechanism for seedlings
TWI713958B (en) * 2018-12-22 2020-12-21 淇譽電子科技股份有限公司 Automated songwriting generation system and method thereof
WO2021147420A1 (en) * 2020-01-22 2021-07-29 华为技术有限公司 Audio generation method, related apparatus and storage medium

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109086408B (en) * 2018-08-02 2022-10-28 腾讯科技（深圳）有限公司 Text generation method and device, electronic equipment and computer readable medium
CN112669849A (en) * 2020-12-18 2021-04-16 百度国际科技（深圳）有限公司 Method, apparatus, device and storage medium for outputting information
TWI784434B (en) * 2021-03-10 2022-11-21 國立清華大學 System and method for automatically composing music using approaches of generative adversarial network and adversarial inverse reinforcement learning algorithm
CN113377992A (en) * 2021-06-21 2021-09-10 腾讯音乐娱乐科技（深圳）有限公司 Song segmentation method, device and storage medium

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2206741A1 (en) * 1997-06-02 1998-12-02 Mitac Inc. Method and apparatus for generating musical accompaniment signals, and method and device for generating a video output in a musical accompaniment apparatus
CN101271457A (en) * 2007-03-21 2008-09-24 中国科学院自动化研究所 Music retrieval method and device based on rhythm
US20140121797A1 (en) * 2012-07-16 2014-05-01 SongFlutter, Inc. System and Method for Combining a Song and Non-Song Musical Content
CN103890838A (en) * 2011-06-10 2014-06-25 X-系统有限公司 Method and system for analysing sound
CN104078035A (en) * 2013-07-02 2014-10-01 深圳市腾讯计算机系统有限公司 Music playing method and device
CN104391980A (en) * 2014-12-08 2015-03-04 百度在线网络技术（北京）有限公司 Song generating method and device
CN105070283A (en) * 2015-08-27 2015-11-18 百度在线网络技术（北京）有限公司 Singing voice scoring method and apparatus
CN105788589A (en) * 2016-05-04 2016-07-20 腾讯科技（深圳）有限公司 Audio data processing method and device
US20160239876A1 (en) * 2012-07-16 2016-08-18 Brian K. ALES Musically contextual audio advertisements
CN105931625A (en) * 2016-04-22 2016-09-07 成都涂鸦科技有限公司 Rap music automatic generation method based on character input
CN106057208A (en) * 2016-06-14 2016-10-26 科大讯飞股份有限公司 Audio correction method and device

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103839559B (en) * 2012-11-20 2017-07-14 华为技术有限公司 Audio file manufacture method and terminal device

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2206741A1 (en) * 1997-06-02 1998-12-02 Mitac Inc. Method and apparatus for generating musical accompaniment signals, and method and device for generating a video output in a musical accompaniment apparatus
CN101271457A (en) * 2007-03-21 2008-09-24 中国科学院自动化研究所 Music retrieval method and device based on rhythm
CN103890838A (en) * 2011-06-10 2014-06-25 X-系统有限公司 Method and system for analysing sound
US20140121797A1 (en) * 2012-07-16 2014-05-01 SongFlutter, Inc. System and Method for Combining a Song and Non-Song Musical Content
US20160239876A1 (en) * 2012-07-16 2016-08-18 Brian K. ALES Musically contextual audio advertisements
CN104078035A (en) * 2013-07-02 2014-10-01 深圳市腾讯计算机系统有限公司 Music playing method and device
CN104391980A (en) * 2014-12-08 2015-03-04 百度在线网络技术（北京）有限公司 Song generating method and device
CN105070283A (en) * 2015-08-27 2015-11-18 百度在线网络技术（北京）有限公司 Singing voice scoring method and apparatus
CN105931625A (en) * 2016-04-22 2016-09-07 成都涂鸦科技有限公司 Rap music automatic generation method based on character input
CN105788589A (en) * 2016-05-04 2016-07-20 腾讯科技（深圳）有限公司 Audio data processing method and device
CN106057208A (en) * 2016-06-14 2016-10-26 科大讯飞股份有限公司 Audio correction method and device

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
CHRISTIAN DITTMAR 等: "Towards lyrics spotting in the SyncGlobal project", 《2012 3RD INTERNATIONAL WORKSHOP ON COGNITIVE INFORMATION PROCESSING (CIP)》 *
江赟: "楚调"唐音"歌吟源流及其基本特点", 《江西科技师范大学学报》 *
纪欢格: "影视配乐的创作手法与录制", 《广东技术师范学院学报》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109166564A (en) * 2018-07-19 2019-01-08 平安科技（深圳）有限公司 For the method, apparatus and computer readable storage medium of lyrics text generation melody
CN109309863A (en) * 2018-08-01 2019-02-05 李春莲 Movie contents matching mechanism for seedlings
TWI713958B (en) * 2018-12-22 2020-12-21 淇譽電子科技股份有限公司 Automated songwriting generation system and method thereof
WO2021147420A1 (en) * 2020-01-22 2021-07-29 华为技术有限公司 Audio generation method, related apparatus and storage medium

Also Published As

Publication number Publication date
WO2018121368A1 (en) 2018-07-05
TW201824249A (en) 2018-07-01
CN108268530B (en) 2022-04-29

Similar Documents

Publication Publication Date Title
CN108268530A (en) Dub in background music generation method and the relevant apparatus of a kind of lyrics
Goto et al. Songle: A Web Service for Active Music Listening Improved by User Contributions.
US10229669B2 (en) Apparatus, process, and program for combining speech and audio data
CN103597543B (en) Semantic audio track mixer
CN111512359A (en) Modular automatic music production server
CN108305643A (en) The determination method and apparatus of emotion information
CN110032630A (en) Talk about art recommendation apparatus, method and model training equipment
CN108810446A (en) A kind of label generating method of video conference, device, equipment and medium
Hoover et al. Interactively evolving harmonies through functional scaffolding
CN108830467A (en) Evaluation method, device, equipment and the storage medium of reading behavior
CN108255840A (en) A kind of recommendation method and system of song
Vitevitch et al. Path-length and the misperception of speech: Insights from network science and psycholinguistics
Fan et al. Ranking-Based Emotion Recognition for Experimental Music.
Thorogood et al. Computationally Created Soundscapes with Audio Metaphor.
CN108304494A (en) Document classification processing method, device and terminal, server, storage medium
CN107770235A (en) One kind bucket song service implementing method and system
Chu et al. An empirical study on how people perceive AI-generated music
CN109584859A (en) Phoneme synthesizing method and device
Chowdhury et al. Tracing back music emotion predictions to sound sources and intuitive perceptual qualities
Imasato et al. Using a language model to generate music in its symbolic domain while controlling its perceived emotion
CN109471951A (en) Lyrics generation method, device, equipment and storage medium neural network based
Nadeem et al. Let's make some music
Vanka et al. Adoption of AI Technology in the Music Mixing Workflow: An Investigation
CN110889787A (en) Music teaching auxiliary system
CN111310453A (en) User theme vectorization representation method and system based on deep learning

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
REG Reference to a national code

Ref country code: HK

Ref legal event code: DE

Ref document number: 1257719

Country of ref document: HK

GR01 Patent grant
GR01 Patent grant