The content of the invention
It is a primary object of the present invention to provide a kind of video to dub switching method, television set and computer-readable storageMedium, it is desirable to provide a kind of method that achievable switching difference dubs version.
To achieve the above object, a kind of video provided by the invention dubs switching method, and the video dubs switchingMethod comprises the following steps：
Obtain the voice data of currently playing video；
Semantics recognition processing is carried out to the voice data, with corresponding to the speech audio in acquisition and the voice dataText information；
Obtained according to the text information and dub resource with what the speech audio in the voice data matched；
Confirm the broadcasting currently playing video playback to the selection for dubbing resource according to user dubs resource.
Preferably, it is described to dub money according to what text information acquisition matched with the speech audio in the voice dataThe step of source, includes：
The programme content title matched with the text information is obtained according to the text information；
Searched according to the programme content title and dub resource with what the speech audio in the voice data matched.
Preferably, described searched according to the programme content title is matched somebody with somebody with what the speech audio in the voice data matchedAlso include after the step of sound resource：
Dub resource the play time corresponding with the text information according to text information matching.
Preferably, it is described that the selection for dubbing resource is confirmed to play the currently playing video playback according to userThe step of dubbing resource also includes：
Match exist it is multiple dub resource when, according to the broadcasting time for dubbing resource to it is described dub resource rowSequence；
The selection that user is carried out according to multiple sequences for dubbing resource is received, and is entered according to the selection describedThe step of dubbing resource of the matching is played in currently playing video.
Preferably, it is described that the selection for dubbing resource is confirmed to play the currently playing video playback according to userAlso include after the step of dubbing resource：
The broadcasting time dubbed resource and be easily selected by a user broadcasting of currently playing video is collected, judges that the broadcasting time isIt is no to reach preset value；
When the broadcasting time reaches preset value, dub this to resource and be arranged to and the acquiescence for playing video matchingDub resource；
When user is again turned on the broadcasting video, resource is dubbed described in broadcasting.
Preferably, it is described played in the currently playing video matching the step of dubbing resource before also wrapInclude：
Obtain the accounts information of active user；
Judge whether the user is the user that meets preparatory condition according to the accounts information of the user；
When the user is the user for meeting preparatory condition, into the currently playing video play described in dubThe step of resource.
Preferably, the accounts information according to the user judges whether the user is the user for meeting preparatory conditionThe step of after also include：
When the user is the user for not meeting preparatory condition, there is provided a visualization interface, the visualization interface are usedIn showing and push registration message to the user.
Preferably, the accounts information according to the user judges whether the user is the user for meeting preparatory conditionThe step of after also include：
When the user is the user for not meeting preparatory condition, there is provided user interface chooses whether to pilot described for userResource is dubbed in matching；
The fragment for dubbing resource of the matching is played in the currently playing video.
In addition, to achieve the above object, the present invention also provides a kind of television set, and the television set includes：Memory, processingThe video that device and being stored in can be run on the memory and on the processor dubs changeover program, and the video is matched somebody with somebodyThe step of method as described above being realized when sound changeover program is by the computing device.
In addition, to achieve the above object, the present invention also provides a kind of computer-readable recording medium, described computer-readableThe changeover program of dubbing of video is stored with storage medium, dubbing for the video is realized such as when changeover program is executed by processorUpper described the step of dubbing switching method of video.
Video of the present invention dubs switching method by obtaining the voice data of currently playing video；To the voice dataSemantics recognition processing is carried out, with the text information corresponding to the speech audio in acquisition and the voice data；According to the textWord acquisition of information dubs resource with what the speech audio in the voice data matched；According to user to the choosing for dubbing resourceThat selects the confirmation broadcasting currently playing video playback dubs resource.By the way that The inventive method achieves match somebody with somebody to what is televisedSound switches so that the audio more diversity that television video plays, user can select different to dub version according to the hobby of oneselfOriginally currently viewing video was played.
Embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
A kind of video of present invention offer dubs switching method, and in one embodiment, referring to Fig. 1, the video is dubbedSwitching method comprises the following steps：
Step S10, obtain the voice data of currently playing video；
TV is a TV for being provided with Android system in the present embodiment.Specifically, the TV obtains currently playingVoice data, the TV that the voice data can be currently playing Web TV or live telecast is either locally storedVideo etc..
Step S20, semantics recognition processing is carried out to the voice data, to obtain and the voice sound in the voice dataText information corresponding to frequency；
Step S30, obtained according to the text information and dub resource with what the speech audio in the voice data matched；
In the technical scheme of the present embodiment, the TV is after currently playing voice data is got, by the audio numberAccording to audio management module is sent to, voice data is sent to predetermined software by audio management module.The predetermined software isThe software of voice recognition processing can be carried out to voice data, can identify that word corresponding to the speech audio in voice data is believedBreath.These predetermined softwares include Xun Fei speech synthesis softwares of University of Science and Technology, IBM speech recognition softwares etc..By these predetermined softwares,The voice data got is subjected to voice recognition processing, to identify text information corresponding to the speech audio of the voice data.Certainly may also be by setting the interface that internet voice can be called to identify engine in TV.
In the present embodiment, the interface of above-mentioned predetermined software is provided with TV, is interrogated and flown as University of Science and Technology using the predetermined softwareIllustrated exemplified by speech synthesis software.After the voice data got is received, University of Science and Technology's news fly this software in itself simultaneouslyThe content that can not be included to the voice data of acquisition parses.In the present embodiment, TV needs to be connected to internet.OneCan be that core builds home gateway by router in embodiment, and by radio connection known to WLAN etc. by TVBe connected with router communication, when TV has cable interface, can by it is logical directly use netting twine in a manner of make TV withRouter communication connects.Further, by way of directly using netting twine TV can also be made to be connected with router communication.EnterOne step, by connecting, after the Internet service provider such as mobile or telecommunications opens network service, it will be passed through to modulate by router and adjustedSection device is connected to internet, to realize the connection of TV and internet.In some other embodiment, TV is logical machineTop box is connected to internet, and TV is connected by wired or wireless way with set top box communication, and set top box can pass through companyThe router of modulation adjuster is connected to or directly by modulating adjuster to internet, to realize the connection of TV and internet.
Further, after the connection of TV and internet is realized, Xun Fei speech synthesis softwares of University of Science and Technology can establish withThe connection for the cloud server that Xun Fei companies of University of Science and Technology provide, now, Xun Fei speech synthesis softwares of University of Science and Technology are by acquired audio numberThe voice data from Xun Fei speech synthesis softwares of television University of Science and Technology is being received according to sending to cloud server, cloud serverAfterwards, this voice data can be parsed and identified, to obtain text information corresponding to the speech audio of the voice data,The text information of matching is sent to television processing, television and dubs money with the presence or absence of matching according to text information inquirySource.
Certainly, it is necessary to illustrate, in the above-described embodiments, if the speech recognition software energy being embedded in the TVAbove-mentioned operation individually is completed in television, then is parsed and identified etc. without transfer data to cloud server againOperation.
In other embodiments, the TV is direct by the voice data after currently playing voice data is gotCloud server is sent to by network.Cloud server is after the voice data that TV sends over is received, cloud serviceDevice calls the interface of internet voice identification engine, the voice data for being sended over TV by internet voice identification engineSpeech recognition is carried out, to obtain text information corresponding to the speech audio in the voice data, internet voice identification engineFurther the text information of matching is sent to cloud server, whether deposited according to the word information inquiry with server beyond the cloudsResource is dubbed in matching.
Step S40, the selection for dubbing resource is confirmed to play matching somebody with somebody for the currently playing video playback according to userSound resource.
In the present embodiment, when TV inquires dubbing in the presence of video information corresponding with the voice data identified matchingDuring resource, push this to user and dub resource, and user according to push selection mutually in requisition for broadcasting dub resource after, playIt is described to dub resource.
In other embodiments, when cloud server is inquired in the presence of matching with the video information when dubbing resource,Then this being dubbed to resource by network push to TV, and being switched to by TV in the service of dubbing, resource is dubbed to play this.
The present embodiment is by obtaining the voice data of currently playing video；The voice data is carried out at semantics recognitionReason, resource is dubbed with what the speech audio in the voice data matched to obtain；Institute is played in the currently playing videoThat states matching dubs resource.Realized by the method for the present embodiment and dub switching to what is televised so that television video is broadcastThe audio put more diversity, user can select different to dub version to play currently viewing regard according to the hobby of oneselfFrequently.
Or in another embodiment, by the way that the speech audio in voice data is converted into corresponding wordInformation, and corresponding video information is matched by text information, by matching somebody with somebody corresponding to the video information matching that parsesSound resource.Wherein, the video information includes currently playing video program content or video frequency program title and currently broadcastThe time point for the video frequency program put.
Specifically, in the above-mentioned speech audio carried out to voice data during voice recognition processing obtains the voice dataAfter corresponding text information, text information yet further is analyzed to obtain corresponding video by correspondingly handling software and believedBreath, resource is dubbed according to corresponding to matching video information.
Illustrated so that TV directly sends voice data to cloud server as an example.TV is sent by voice dataTo cloud server, cloud server calls the interface of internet voice identification engine, by speech recognition engine by audioData are converted into text information corresponding with the voice data, and internet voice identifies engine after identification converts, and will turnThe text information for changing completion is sent to cloud server, and cloud server further obtains and the text information according to text informationCorresponding video information.Namely in this manner, identified in internet voice and pass through cloud server between engine and TVAs intermediate processor；Television transmission data are stored at cloud server, and identify engine point by internet voiceData after analysis processing continue to transmit to cloud server, and the resource of dubbing of related matching also is stored in cloud serverOn.The data storage that in this way needs can be stored in server, greatly reduces the storage pressure of TV beyond the cloudsPower, memory space is added, improve the response speed of TV.
It should be noted that reference picture 2, above-mentioned steps S30 includes：
Step S301, the programme content title matched with the text information is obtained according to the text information；
Step S302, searched according to the programme content title and dubbed with what the speech audio in the voice data matchedResource.
In the present embodiment, after text information corresponding with the speech audio in the voice data is got, furtherThe programme content title matched with text information is obtained according to the word information searching.As currently playing programme content title isSong of Joy, then after text information corresponding to the speech audio of the program is got, according to the word information matches and the wordProgramme content title corresponding to information.Specifically, can be by the matching of text information and programme content lines, with text informationProgram of the lines of matching closest to as matching.And further according in programme content title lookup and the voice dataSpeech audio matching dub resource, i.e., according to the programme content title search existing for dub resource.
Corresponding programme content title is found by the text information that first passes through in the present embodiment, and further according to sectionMesh content name dubs resource corresponding to obtaining, and improves the accuracy of matching efficiency and matching.
In another embodiment, resource can be further dubbed according to matching the text information with the word to believePlay time corresponding to manner of breathing.
The present embodiment by text information match voice data in speech audio corresponding to the TV programme play whenBetween point, to facilitate subsequently when dub resource and playing, can directly play it is corresponding with the time point dub resource, improveThe compatible degree dubbed.
It should be noted that in another embodiment, also include after step S30：
Match exist it is multiple dub resource when, according to the broadcasting time for dubbing resource to it is described dub resource rowSequence；
The selection that user is carried out according to multiple sequences for dubbing resource is received, and step is entered according to the selectionS40。
In the present embodiment, when currently playing TV programme match when dubbing resource of multiple versions, such as star's version, doPlate, practical joke version etc. are laughed at, can be now ranked up according to the broadcasting time for dubbing resource matched, and will be sorted in TVShown on interface, so that user selects resource of dubbing interested as needed, and need what is played to match somebody with somebody in user's selectionAfter sound resource, play this and dub resource.
In another embodiment, reference picture 3, the step S40 also include afterwards：
Step S50, the broadcasting time dubbed resource and be easily selected by a user broadcasting of currently playing video is collected, is broadcast described in judgementPut whether number reaches preset value；
Step S60, when the broadcasting time reaches preset value, dub this to resource and set and the broadcasting video matchingAcquiescence dub resource；
Step S70, when user is again turned on the broadcasting video, resource is dubbed described in broadcasting.
In the present embodiment, determine whether that the broadcasting video is currently easily selected by a user the broadcasting for dubbing resource time of broadcastingNumber, and judge whether the broadcasting time reaches preset value, when the broadcasting time reaches preset value, dub this to resource and setIt is set to and dubs resource with the acquiescence of the broadcasting video matching, and when next user is again turned on the broadcasting video, nowThen can directly play that broadcasting time reaches preset value dubs resource.
For example, when user currently selects " Song of Joy " of the video of broadcasting, now user have selected the choosing that switching is dubbed, then system dubs resource according to the selection of the option of user with " Song of Joy " currently playing fragment match, and gathers this and match somebody with somebodySound resource is easily selected by a user the number of broadcasting, when this is dubbed the number that resource is easily selected by a user broadcasting and reaches preset value, then willThis is dubbed resource and is arranged to dub resource with the acquiescence of the broadcasting video matching.When user is again turned on " Song of Joy " videoWhen, now directly play this and dub resource；Or when user selects unlatching to dub switching service, play this and dub resource.
The present embodiment judges institute by collecting the broadcasting time dubbed resource and be easily selected by a user broadcasting of currently playing videoState whether broadcasting time reaches preset value；When the broadcasting time reaches preset value, dub this to resource setting and broadcast with describedThe acquiescence for putting video matching dubs resource；When user is again turned on the broadcasting video, resource is dubbed described in broadcasting.Carried with thisHigh Consumer's Experience.
Further, referring to Fig. 4, in another embodiment of the invention, also including before the step S40：
Step S41, obtain the accounts information of active user；
Step S42, judge whether the user is the user that meets preparatory condition according to the accounts information of the user；
Step S43, when the user is the user for meeting preparatory condition, into step S40.
In the present embodiment, get matching dub resource after, further parse the voice data.Need what is illustratedThat the voice data includes the accounts information of active user, the accounts information include ID (Identification,Identity), the ID is used for the account of unique mark user, to judge whether user is paying customer according to the ID；The accountNumber information also includes mark ID, the software version number of the TV, hardware version numbers and the other information of user of TV；InstituteStating voice data also includes audio-frequency information, and the audio-frequency information includes speech audio, the length of the speech audio and voiceBuffer address of audio etc..
In the present embodiment, getting the ID number of user and judged to obtain whether active user is that payment is used according to the ID numberWhether can be that paying customer is operated accordingly according to it behind family.After active user is paying customer, then directly broadcastingWhat is be fitted on dubs resource.
When the user is not paid for user, then can to user push try out this match somebody with somebody sound function service, or toWhether family push registration message interface, registered for selection by the user.Specifically, can be designed according to actual conditions,This is not particularly limited.
The present embodiment by obtaining ID to judge to obtain whether the user is paying customer, so as to furtherTargetedly operated, improve Consumer's Experience, enrich performance.
Further, in another embodiment of the invention, also include after the step S42：
Step S44, when the user is the user for not meeting preparatory condition, there is provided a visualization interface, it is described visualChange interface to be used to showing and pushing registration message to the user.
In the present embodiment, when detecting that the user is not paid for user, then further generation registration disappears the TVInterface is ceased, and is shown in the display interface of TV.After the registration message that user shows on display interface is received, it can be used distantControl device and corresponding operation is selected according to the prompting of display interface；Or in other embodiments, the TV and some movements are wholeEnd is connected, and can receive the control of mobile terminal, can now use the corresponding operation of mobile terminal selection；Wherein, it is described mobile wholeEnd can be mobile phone, flat board etc..
The TV then responds the operation after the operation of remote control or mobile terminal selection is received.Such as, Yong HuxuanSelect and be registered as paying customer, can now call third-party charging software, the id information of user is changed after user is paid, withUser is identified as paying customer.
Further, reference picture 5, in another embodiment, also include after the step S42：
Step S45, when the user is the user for not meeting preparatory condition, there is provided user interface chooses whether for userThat pilots the matching dubs resource；
Step S46, the fragment for dubbing resource of the matching is played in the currently playing video.
In the present embodiment, when the user is not paid for user, further generation is available for user to carry out experience on probationInterface, and being shown on the TV, so that the user dubs resource according to what prompting chose whether to pilot the matching.After user's selection experience on probation, resource is dubbed in the matching played in TV or cloud server memory storage.And in userExperience on probation terminate after, show registration message interface on display interface again, attract user to turn into paying customer.
The present invention further provides a kind of television set, the television set 100 includes：Memory, processor and it is stored in instituteThat states the video that can run on memory and on the processor dubs changeover program, and the video dubs changeover program quiltThe step of dubbing switching method of the video described in any of the above-described embodiment is realized during the computing device.
Specifically, the video realizes step when dubbing changeover program by the computing device：
Obtain the voice data of currently playing video；
Semantics recognition processing is carried out to the voice data, with corresponding to the speech audio in acquisition and the voice dataText information；
Obtained according to the text information and dub resource with what the speech audio in the voice data matched；
Confirm the broadcasting currently playing video playback to the selection for dubbing resource according to user dubs resource.
The present embodiment is by obtaining the voice data of currently playing video；The voice data is carried out at semantics recognitionReason, resource is dubbed with what the speech audio in the voice data matched to obtain；According to user to the choosing for dubbing resourceThat selects the confirmation broadcasting currently playing video playback dubs resource.Realized by the method for the present embodiment to televisingAudio conversion so that the audio more diversity that television video plays, user can select different dub according to the hobby of oneselfVersion plays currently viewing video.
Further, the step S30 includes：
Step S301, the programme content title matched with the text information is obtained according to the text information；
Step S302, searched according to the programme content title and dubbed with what the speech audio in the voice data matchedResource.
In the present embodiment, after text information corresponding with the speech audio in the voice data is got, furtherThe programme content title matched with text information is obtained according to the word information searching.As currently playing programme content title isSong of Joy, then after text information corresponding to the speech audio of the program is got, according to the word information matches and the wordProgramme content title corresponding to information.Specifically, can be by the matching of text information and programme content lines, with text informationProgram of the lines of matching closest to as matching.And further according in programme content title lookup and the voice dataSpeech audio matching dub resource, i.e., according to the programme content title search existing for dub resource.
Corresponding programme content title is found by the text information that first passes through in the present embodiment, and further according to sectionMesh content name dubs resource corresponding to obtaining, and improves the accuracy of matching efficiency and matching.
In another embodiment, resource can be further dubbed according to matching the text information with the word to believePlay time corresponding to manner of breathing.
The present embodiment by text information match voice data in speech audio corresponding to the TV programme play whenBetween point, to facilitate subsequently when dub resource and playing, can directly play it is corresponding with the time point dub resource, improveThe compatible degree dubbed.
In another embodiment, the changeover program of dubbing of the video is executed by processor following steps：
The broadcasting time dubbed resource and be easily selected by a user broadcasting of currently playing video is collected, judges that the broadcasting time isIt is no to reach preset value；
When the broadcasting time reaches preset value, dub this to resource and set the acquiescence with the broadcasting video matching to match somebody with somebodySound resource；
When user is again turned on the broadcasting video, resource is dubbed described in broadcasting.
In the present embodiment, determine whether that the broadcasting video is currently easily selected by a user the broadcasting for dubbing resource time of broadcastingNumber, and judge whether the broadcasting time reaches preset value, when the broadcasting time reaches preset value, dub this to resource and setIt is set to and dubs resource with the acquiescence of the broadcasting video matching, and when next user is again turned on the broadcasting video, nowThen can directly play that broadcasting time reaches preset value dubs resource.
For example, when user currently selects " Song of Joy " of the video of broadcasting, now user have selected the choosing that switching is dubbed, then system dubs resource according to the selection of the option of user with " Song of Joy " currently playing fragment match, and gathers this and match somebody with somebodySound resource is easily selected by a user the number of broadcasting, when this is dubbed the number that resource is easily selected by a user broadcasting and reaches preset value, then willThis is dubbed resource and is arranged to dub resource with the acquiescence of the broadcasting video matching.When user is again turned on " Song of Joy " videoWhen, now directly play this and dub resource；Or when user selects unlatching to dub switching service, play this and dub resource.
The present embodiment judges institute by collecting the broadcasting time dubbed resource and be easily selected by a user broadcasting of currently playing videoState whether broadcasting time reaches preset value；When the broadcasting time reaches preset value, dub this to resource setting and broadcast with describedThe acquiescence for putting video matching dubs resource；When user is again turned on the broadcasting video, resource is dubbed described in broadcasting.Carried with thisHigh Consumer's Experience.
Further, the changeover program of dubbing of the video is executed by processor following steps：
Obtain the accounts information of active user；
Judge whether the user is the user that meets preparatory condition according to the accounts information of the user；
When the user is the user for meeting preparatory condition, obtained and the audio number into according to the text informationThe step of dubbing resource of speech audio matching in.
The present embodiment by obtaining ID to judge to obtain whether the user is paying customer, so as to furtherTargetedly operated, improve Consumer's Experience, enrich performance.
Further, the changeover program of dubbing of the video is executed by processor following steps：
When the user is the user for not meeting preparatory condition, there is provided a visualization interface, the visualization interface are usedIn showing and push registration message to the user.
In the present embodiment, when detecting that the user is not paid for user, then further generation registration disappears the TVInterface is ceased, and is shown in the display interface of TV.After the registration message that user shows on display interface is received, it can be used distantControl device and corresponding operation is selected according to the prompting of display interface；Or in other embodiments, the TV and some movements are wholeEnd is connected, and can receive the control of mobile terminal, can now use the corresponding operation of mobile terminal selection；Wherein, it is described mobile wholeEnd can be mobile phone, flat board etc..
The TV then responds the operation after the operation of remote control or mobile terminal selection is received.Such as, Yong HuxuanSelect and be registered as paying customer, can now call third-party charging software, the id information of user is changed after user is paid, withUser is identified as paying customer.
Further, the changeover program of dubbing of the video is executed by processor following steps：
When the user is the user for not meeting preparatory condition, there is provided user interface chooses whether to pilot described for userResource is dubbed in matching；
The fragment for dubbing resource of the matching is played in the currently playing video.
In the present embodiment, when the user is not paid for user, further generation is available for user to carry out experience on probationInterface, and being shown on the TV, so that the user dubs resource according to what prompting chose whether to pilot the matching.After user's selection experience on probation, resource is dubbed in the matching played in TV or cloud server memory storage.And in userExperience on probation terminate after, show registration message interface on display interface again, attract user to turn into paying customer.
The present invention also provides a kind of computer-readable recording medium, and calculating is stored with the computer-readable recording mediumMachine program, dubbing for the video realize that dubbing for the video described in any one is cut as described above when changeover program is executed by processorThe step of changing method.
The preferred embodiments of the present invention are these are only, are not intended to limit the scope of the invention, it is every to utilize this hairThe equivalent structure or equivalent flow conversion that bright specification and accompanying drawing content are made, or directly or indirectly it is used in other related skillsArt field, is included within the scope of the present invention.