CN109640142A

CN109640142A - A kind of content recommendation method and device, equipment, storage medium

Info

Publication number: CN109640142A
Application number: CN201811575254.0A
Authority: CN
Inventors: 赵艳萍; 赵敏; 魏丽萍; 刘养亭
Original assignee: MIGU Digital Media Co Ltd; MIGU Culture Technology Co Ltd
Current assignee: MIGU Digital Media Co Ltd; MIGU Culture Technology Co Ltd
Priority date: 2018-12-21
Filing date: 2018-12-21
Publication date: 2019-04-16
Anticipated expiration: 2038-12-21
Also published as: CN109640142B

Abstract

The embodiment of the present application discloses a kind of content recommendation method and device, equipment, storage medium, wherein the described method includes: obtaining the voice messaging that the first user is used to indicate recommendation；If the voice print database of the voice messaging the first content of text corresponding with the voice messaging mismatches, the determining target object to match with first content of text, the target object is not first user；Commending contents are carried out to the target object.

Description

A kind of content recommendation method and device, equipment, storage medium

Technical field

The invention relates to electronic technology, a kind of content recommendation method and device, equipment, storage are related to, but are not limited to Medium.

Background technique

With the continuous development of science and technology, electronic technology has obtained development at full speed, and people can pass through various types Electronic equipment, enjoy the various conveniences of development in science and technology bring.For example, people can recommend application to issue for referring to by certain Show the phonetic order of recommendation, for example, user can issue " recommending a film " to recommendation application, " recommend a children's stories Story " etc. meets the phonetic order of user's current demand, which applies after receiving the phonetic order, can be to user The phonetic order of input carries out Application on Voiceprint Recognition, speech recognition and semantics recognition etc., and is user recommendation according to recognition result Content, to meet the needs of user.

Summary of the invention

In view of this, the embodiment of the present application is to solve the problems, such as present in the relevant technologies at least one and provide a kind of content Recommended method and device, equipment, storage medium.

The technical solution of the embodiment of the present application is achieved in that

In a first aspect, the embodiment of the present application provides a kind of content recommendation method, which comprises

Obtain the voice messaging that the first user is used to indicate recommendation；

If the voice print database of the voice messaging the first content of text corresponding with the voice messaging mismatches, determine The target object to match with first content of text, the target object are not first users；

Commending contents are carried out to the target object.

Second aspect, the embodiment of the present application provide a kind of content recommendation device, comprising:

Module is obtained, is configured to obtain the voice messaging that the first user is used to indicate recommendation；

Determining module, if the voice print database for being configured to the voice messaging the first text corresponding with the voice messaging Content mismatches, the determining target object to match with first content of text, and the target object is not first use Family；

Recommending module is configured to carry out commending contents to the target object.

The third aspect, the embodiment of the present application provide a kind of electronic equipment, including memory and processor, the memory are deposited The computer program that can be run on a processor is contained, the processor realizes above content recommended method when executing described program In step.

Fourth aspect, the embodiment of the present application provide a kind of computer readable storage medium, are stored thereon with computer program, The step in above content recommended method is realized when the computer program is executed by processor.

In the embodiment of the present application, a kind of content recommendation method is provided, in the method, is getting the first user for referring to After showing the voice messaging of recommendation, it is first determined the voice print database of the voice messaging corresponding with the voice messaging Whether one content of text matches；If the voice print database and first content of text mismatch, commending contents are not given The first user of the voice messaging is issued, but, in the target object (institute for determining to match with first content of text Stating target object not is first user) after, Xiang Suoshu target object carries out commending contents.

In this way, when carrying out commending contents, first, unwanted user can be given to avoid by commending contents, improve and recommend Accuracy rate；Second, it is mismatched in the voice print database of the voice messaging the first content of text corresponding with the voice messaging When, because the corresponding recommendation of the voice messaging not recommended to the first user for issuing the voice messaging, The power consumption of the corresponding first terminal of first user can be saved；Third, due to that can not only push away when carrying out commending contents The other users that can also be recommended to the first user for issuing the voice messaging except described first with open air are recommended, to expand The big application scenarios of commending contents.

Detailed description of the invention

Figure 1A is the composed structure schematic diagram of the embodiment of the present application network architecture；

Figure 1B is the implementation process schematic diagram of the embodiment of the present application content recommendation method；

Fig. 2 is the implementation process schematic diagram of another content recommendation method of the embodiment of the present application；

Fig. 3 A is the embodiment of the present application terminal interface display schematic diagram；

Fig. 3 B is another terminal interface display schematic diagram of the embodiment of the present application；

Fig. 4 is the composed structure schematic diagram of the embodiment of the present application content recommendation device；

Fig. 5 is a kind of hardware entities schematic diagram of the embodiment of the present application electronic equipment.

Specific embodiment

The technical solution of the application is further elaborated on reference to the accompanying drawings and examples.

The embodiment of the present application first provides a kind of network architecture of commending contents, and Figure 1A is the embodiment of the present application network architecture Composed structure schematic diagram, as shown in Figure 1A, the network architecture include terminal 111 to 11N and server 121；Wherein, terminal 111 It is interacted to network 131 is passed through between 11N and server 121；It is handed between terminal 111 and terminal 11N by network 131 Mutually.

The embodiment of the present application provides a kind of content recommendation method, and this method is applied to electronic equipment, and in general, electronics is set Standby can be various types of equipment with commending contents ability in the process of implementation, for example, the electronic equipment can be with The terminal 111 to 11N, the terminal may include mobile phone, tablet computer, desktop computer, personal digital assistant, navigator, Digital telephone, visual telephone, television set, sensing equipment etc..For another example, the electronic equipment is also possible to server 121.This method The function of being realized can realize that certain program code can be saved by the processor caller code in electronic equipment In computer storage medium, it is seen then that the electronic equipment includes at least pocessor and storage media.

Figure 1B is the implementation process schematic diagram of the embodiment of the present application content recommendation method, and as shown in Figure 1B, this method includes Step S101 to step S103:

S101, the voice messaging that the first user is used to indicate recommendation is obtained；

In practical applications, the first user of input voice information may not be the target object for finally needing to recommend.It lifts For example, Xiao Ming goes on business in other places, and the child Tong Tong of Xiao Ming is in, and Xiao Ming is virgin in order to fool the child for sleeping and being in, and Xiao Ming passes through mobile phone On recommendation application, input one section of phonetic order " recommend a children's story ", content carried out based on the phonetic order at this time and is pushed away The target object recommended is apparently not Xiao Ming, but the child Tong Tong of Xiao Ming, so, refer to here in the voice for getting Xiao Ming's input After order, first have to determine based on the phonetic order whether the feature of Xiao Ming with the content of text of the phonetic order " recommends one Children's story " matches, if it does not match, executing step S102；If matched, children's story is directly recommended small It is bright.

For another example, Xiao Ming wants that the grandfather to Henan native place recommends an opera, Xiao Ming is answered by the recommendation on mobile phone in Xi'an With inputting one section of phonetic order " playing one section of Henan opera for my grandfather ", it is clear that " broadcast for my grandfather with the content of text of phonetic order Put one section of Henan opera " target object that matches is the grandfather of Xiao Ming, rather than issue the Xiao Ming of phonetic order.

If the voice print database of S102, the voice messaging the first content of text corresponding with the voice messaging is not Match, the determining target object to match with first content of text, the target object is not first user；

In other embodiments, the target object is also possible to first user, that is to say, that when the vocal print number When according to matching with first content of text, it can determine that current target object to be recommended is first user, at this point, Commending contents can be carried out to first user.

It should be noted that the voice print database of the voice messaging the first content of text corresponding with the voice messaging is not Matching, typically refer to and the corresponding personal feature of voice print database personal feature corresponding with first content of text not Match, for example, obtaining corresponding personal feature is " male, 32 years old " by identifying to the voice print database, the voice letter Ceasing corresponding first content of text is " playing a children's story ", and by " playing a children's story " under normal circumstances, institute is right The personal feature answered should be children, rather than 32 years old male, so, " male, 32 years old " and " playing a children's story " The two is unmatched.

S103, Xiang Suoshu target object carry out commending contents.

In other embodiments, if the voice print database of the voice messaging the first text corresponding with the voice messaging Content matches, then target object is first user, directly can carry out commending contents to first user；If institute Voice print database the first content of text corresponding with the voice messaging for stating voice messaging mismatches, then target object can be institute State the object contact person to match in the address list of the first user with first content of text, accordingly, " Xiang Suoshu target pair As carrying out commending contents " it can be to send recommendation message to the object contact person, which is used to join to the target It is that people carries out commending contents, recommendation can be carried in the recommendation message, recommendation can be text information, multimedia Information can also be webpage link information, the webpage link information of multimedia messages of text information, wherein multimedia messages It can be picture, video or audio.

It should be noted that the executing subject of step S101 to step S103 can be first user corresponding first Terminal is also possible to server.If the executing subject of step S101 to step S103 are the first terminals, for step S103, the first terminal can pass through the logical of terminal corresponding with the target object after getting content to be recommended Letter connection, is sent to the corresponding terminal of the target object for the content to be recommended, in the corresponding terminal of the target object The upper display content to be recommended, to recommend the target object；In other embodiments, if the voice messaging Voice print database the first content of text corresponding with the voice messaging matches, and the first terminal can be believed based on the voice Breath, obtains content to be recommended, and then directly output shows the content to be recommended, to recommend first user.

If the executing subject of step S101 to step S103 are servers, for step S103, the server can It, will be described to be recommended by the communication connection of terminal corresponding with the target object with after getting content to be recommended Content is sent to the corresponding terminal of the target object, shows in the corresponding terminal of the target object described to be recommended interior Hold, to recommend the target object；In other embodiments, if the voice print database of the voice messaging and the voice Corresponding first content of text of information matches, and the server is getting content to be recommended, can be by with described first The content to be recommended is sent to the first terminal, institute is shown on the first terminal by the communication connection between terminal Content to be recommended is stated, to recommend first user.

In the embodiment of the present application, a kind of content recommendation method is provided, in the method, is used for getting the first user After the voice messaging for indicating recommendation, it is first determined the voice print database of the voice messaging is corresponding with the voice messaging Whether the first content of text matches；If the voice print database and first content of text mismatch, not by commending contents To issuing first user, but in the target object (target pair for determining to match with first content of text As not being first user) after, Xiang Suoshu target object carries out commending contents.

In this way, when carrying out commending contents, first, unwanted user can be given to avoid by commending contents, improve and recommend Accuracy rate；Second, it is mismatched in the voice print database of the voice messaging the first content of text corresponding with the voice messaging When, because the corresponding recommendation of the voice messaging not recommended to the first user for issuing the voice messaging, The power consumption of the corresponding first terminal of first user can be saved；Third, due to that can not only push away when carrying out commending contents The other users that can also be recommended to the first user for issuing the voice messaging except described first with open air are recommended, to expand The big application scenarios of commending contents, for example, Xiao Ming in Xi'an, wants that the grandfather to Henan native place recommends an opera, it is small at this time It is bright one section of phonetic order " playing one section of Henan opera for my grandfather " to be inputted by the recommendation application on mobile phone, thus reach to The purpose of grandfather's recommendation opera.

The embodiment of the present application provides another content recommendation method, and Fig. 2 is another content recommendation method of the embodiment of the present application Implementation process schematic diagram, as shown in Fig. 2, the method comprising the steps of S201 to step S206:

S201, the voice messaging that the first user is used to indicate recommendation is obtained；

S202, the voice print database for determining the voice messaging the first content of text corresponding with the voice messaging whether phase Matching；If so, executing step S203；Otherwise, step S204 is executed；

S203, it is based on the voice messaging, the first user of Xiang Suoshu carries out commending contents；

S204, determination first terminal currently corresponding with first user establish the second end of near field communication End；

Common near field communication the, for example, WLAN (Wireless based on IEEE 802.11b standard Fidelity, WiFi) direct-connected, bluetooth connection, ZigBee connection, infrared connection etc..

S205, from the corresponding second user of the second terminal, the determining mesh to match with first content of text Mark object；

It should be noted that from the corresponding second user of the second terminal, determining and first content of text When the target object to match, the number of identified target object can be 0, that is, in the second user, not with institute The target object that the first content of text matches is stated, at this point it is possible to which the response disappears to the first user returning response message Breath is for prompting first user to specify target object, to carry out content to target object specified by first user Recommend.

For example, the response message content returned to the corresponding first terminal of first user is " not finding will recommend Object, the target object to be recommended please be specify ", as shown in Figure 3A, show sound on the display interface 30 of the first terminal Answer message 301 " not finding the object to be recommended, the target object to be recommended please be specify ".The first terminal receives described the The target object of one user input, and the target object inputted to first user carries out commending contents.

It in other embodiments, can be corresponding according to the historical communication data of the second user for step S205 Second content of text, from the second user, the determining target object to match with first content of text；If described The historical communication data of each user in second user are mismatched with first content of text, and the first user of Xiang Suoshu returns Response message is returned, the response message is for prompting first user to specify target object.

S206, Xiang Suoshu target object carry out commending contents.

Still by taking the above-mentioned response message 301 for returning to first user as an example, first user can pass through click Region where the response message 301, to trigger the (tool of address list 302 that the first terminal shows first user Body is referring to Fig. 3 B), when the first terminal detects touch operation, the first terminal is determined corresponding to the touch operation Contact person, and the contact person is determined as the target object.In other embodiments, first user can also pass through Phonetic order is issued to specify the target object, specifies the mode of the target object not to be first user here It limits.

The embodiment of the present application provides another content recommendation method, the method comprising the steps of S301 to step S307:

S301, the voice messaging that the first user is used to indicate recommendation is obtained；

S302, the voice print database for determining the voice messaging the first content of text corresponding with the voice messaging whether phase Matching；If so, executing step S303；Otherwise, step S304 is executed；

S303, it is based on the voice messaging, the first user of Xiang Suoshu carries out commending contents；

The communications records of S304, the address list for obtaining first user and first user；

It should be noted that the address list can be in the corresponding first terminal of first user some apply it is logical News record, in general, the address list are the address list in the application in the first terminal with commending contents function.It is described In the communications records of first user, the contact person communicated within a preset time period with first user is at least had recorded Mark, number of communications corresponding with the identification of contacts.

S305, the communications records are based on, are determined from the address list of first user with first user's Number of communications is greater than the third user of the first preset threshold；

In other embodiments, the third user may include current and communication connection has been established in first user Contact person, for example, the current contact person with first user chat；Alternatively, the third user can also be described First user presets, for example, the household of oneself or good friend are set the third user by first user.

Here, without limitation for the number of the target object, it can be in the third user and believe with the voice The contact person that corresponding first content of text matches is ceased, for example, with the cohesion of first user is highest contacts People, the cohesion can be embodied with the number of communications between the contact person and first user.

S306, from the third user, determine the mesh that corresponding with the voice messaging the first content of text matches Mark object；

In other embodiments, can according to the corresponding third content of text of historical communication data of the third user, From the third user, the determining target object to match with first content of text.If in the third user The historical communication data of each user are mismatched with first content of text, Xiang Suoshu the first user returning response message, The response message is for prompting first user to specify target object.

S307, Xiang Suoshu target object carry out commending contents.

The embodiment of the present application provides another content recommendation method, the method comprising the steps of S401 to step S407:

S401, the voice messaging that the first user is used to indicate recommendation is obtained；

S402, the voice print database based on the voice messaging, determine fisrt feature information；The fisrt feature information is used for Characterize the personal feature of first user；

In practical applications, Application on Voiceprint Recognition can be carried out to the voice print database of the voice messaging, to obtain described the The sound characteristic of one user, the sound characteristic may include tone, tone color etc..Based on the sound characteristic of first user, Determine the fisrt feature information.For example, fisrt feature information database can be established based on different sound characteristics, it is described It include the corresponding relationship between the sound characteristic and the fisrt feature information in fisrt feature information database, it is right based on this It should be related to, can determine fisrt feature information corresponding to the sound characteristic of first user.In general, described first is special Reference breath includes the information such as gender, the age of first user.

S403, it is based on corresponding first content of text of the voice messaging, determines second feature information；The second feature Information is for characterizing personal feature corresponding with first content of text；

Similarly, the second feature information includes the information such as gender, age；It in other embodiments, can be to described Voice messaging carries out semantics recognition, to obtain first content of text；It is to be appreciated that the first different content of text, Corresponding second feature information may be far from each other.For example, first content of text is " playing a children's story ", Its corresponding second feature information may be children；For another example, first content of text is " playing one section of Henan opera ", corresponding Second feature information may be old man.That is, being based on first content of text, can determine and first text The characteristic information that content matches, for example, determining the information such as the age of recommended target object, gender.

In other embodiments, it can be based on first content of text, determine aiming field to be recommended；From preset Second feature information corresponding with the aiming field is inquired in database.It is to be appreciated that since the database is preparatory It establishes, therefore by inquiring the database, can quickly determine second feature information corresponding to the aiming field.? In practical application, second feature information database, the second feature information can be established based on the first different content of text The corresponding relationship between keyword and the second feature information in database including first content of text.Citing comes It says, as the following table 1 shows a kind of second feature information database.

Table 1

Keyword	Second feature information
		……	……
Children's story	3 years old by 7 years old
		Online game	18 years old or more
……	……

Assuming that the voice messaging based on first user, obtained content of text is " playing a children's story ", this When word segmentation processing first can be carried out to first content of text, obtain " playing ", "one", " children's story " these three words, Then, based on the lists of keywords in second feature information database shown in upper table 1, can determine " children's story " be to The aiming field of recommendation, then, the correspondence based on the characteristic information configured in " children's story " and second feature information database Relationship can be determined with " children's story " matched characteristic information to be " 3 years old to 7 years old ", hence, it can be determined that " playing one Second feature information corresponding to the content of text of children's story " is " 3 years old to 7 years old ".

Since the fisrt feature information is personal feature corresponding with the voice print database, the second feature information is Personal feature corresponding with first content of text, so, the second feature information and the fisrt feature information may It is identical, it is also possible to different.

It should be noted that in other embodiments, step S403 can also be first carried out, then step S402 is executed, also It is to say, the sequencing of described for determination fisrt feature information and the second feature information, here without limitation.

S404, according to the fisrt feature information and the second feature information, determine the vocal print number of the voice messaging According to whether being matched with first content of text；If so, executing step S405；Otherwise, step S406 is executed；

It is to be appreciated that determining the voice if the fisrt feature information and the second feature information match The voice print database of information the first content of text corresponding with the voice messaging matches, that is, current target object to be recommended It is the same person with the first user for issuing the voice messaging, executes step S405 at this time；If the fisrt feature information It is mismatched with the second feature information, determines the voice print database of the voice messaging the first text corresponding with the voice messaging This content mismatches, that is, current target object to be recommended and to issue the first user of the voice messaging be not the same person, Step S406 is executed at this time, carries out commending contents to the other users in addition to first user.

In general, if the fisrt feature information and the second feature information belong to same preset range, really The fixed fisrt feature information matches with the second feature information；, whereas if the fisrt feature information and described Two characteristic informations respectively fall in different preset ranges, determine that stating fisrt feature information and the second feature information mismatches. For example, the following table 2 shows preset range corresponding to different characteristic information.

Table 2

Scope identifier	Characteristic information
		……	……
4	3 years old to 7 years old, children
		5	30 years old to 40 years old, male
……	……

Based on table 2, it is assumed that the fisrt feature information is " male, 32 years old ", and the second feature information is " children ", from Can determine that " male, 32 years old " falls into preset range corresponding to scope identifier 4 in table 2, " children " fall into scope identifier 5 pair The preset range answered, it is seen then that the two characteristic informations fall into different preset ranges, so the with " male, 32 years old " One characteristic information and the second feature information with " children " are unmatched.

S405, it is based on the voice messaging, the first user of Xiang Suoshu carries out commending contents；

In other embodiments, when carrying out commending contents to first user, available first user's Like label, hobby label and the voice messaging based on first user are recommended more to meet for first user The content of the first user current demand.Wherein, the hobby label of first user can be used by learning described first The historical communication data at family obtain, for example, the video content of study user viewing, user browse webpage within a preset time period Content etc., to obtain the hobby label of first user.

The target object that S406, determination and first content of text match, the target object is not described first User；

S407, Xiang Suoshu target object carry out commending contents.

In content recommendation method provided by the embodiments of the present application, the language of recommendation is used to indicate in the first user of acquisition After message breath, fisrt feature information and second feature information are obtained, wherein the letter of fisrt feature described in the fisrt feature information The personal feature for characterizing first user is ceased, the second feature information is for characterizing and first content of text pair The personal feature answered；Whether matched by the determination fisrt feature information and the second feature information, to accurately determine Whether the voice print database of the voice messaging the first content of text corresponding with the voice messaging matches；That is, if first is special Reference breath and the second feature information match, and determine that the voice print database of the voice messaging is corresponding with the voice messaging First content of text be it is matched, at this time using first user as the target object, in Xiang Suoshu target target carries out Hold and recommends；If the fisrt feature information and the second feature information mismatch, the vocal print number of the voice messaging is determined According to the first content of text corresponding with the voice messaging mismatch, at this time will except described first use open air other users as The target object, Xiang Suoshu target object carry out commending contents.

In this way, by the characteristic information (the i.e. described fisrt feature information) of comparison first user and based on the voice Whether the corresponding obtained characteristic information of first content of text of information (the i.e. described second feature information) matches, thus accurately Determine whether first user is target object to be recommended, and then is the first content of text corresponding with the voice messaging The target object to match carries out commending contents, improves and recommends accuracy rate.

The embodiment of the present application provides another content recommendation method, including step S501 to step S508:

S501, the voice messaging that the first user is used to indicate recommendation is obtained；

If the voice print database of S502, the voice messaging the first content of text corresponding with the voice messaging is not Match, determines that first terminal currently corresponding with first user establishes the second terminal of near field communication；

S503, the historical communication data for obtaining the corresponding second user of the second terminal；

In practical applications, if the executing subject of step S501 to step S508 are first users corresponding first Terminal, the first terminal can be by the near field communication links between the second terminal, and Xiang Suoshu second is eventually End sends request message, with the historical communication data of second terminal described in request, for example, the historical communication data include Browse data, sharing data, screenshotss data etc..The second terminal is after receiving the request message, in authority checking By later, obtaining itself historical communication data, and these data are sent to by near field communication link described First terminal.

If the executing subject of step S501 to step S508 are servers, server can be used with described in direct request first The corresponding first terminal in family sends the identity of the second user to the server；The server receive it is described After the identity of second user, based on the identity of the second user, the historical communication of the second user is inquired Data, for example, the information such as video access record, web page access record.Alternatively, server can pass through first terminal and described the Two terminals establish indirect communication, to obtain the historical communication data stored in the second terminal (for example, browsing data, dividing Enjoy data, screenshotss data).

S504, corresponding second content of text of historical communication data according to the second user, from the second user In, the determining target object to match with first content of text；

In general, it if first content of text matches with second content of text, can determine described The corresponding second user of second content of text is the target object；, whereas if first content of text and described second Content of text mismatches, then can determine that the corresponding second user of second content of text is not the target object.It can be with Ground is understood, if second content of text is included the voice messaging corresponding first content of text, it is determined that described the Two content of text the first content of text corresponding with the voice messaging matches；Alternatively, can be in advance to the second user Historical communication data learnt, obtain the hobby label of the second user, then, the hobby based on the second user Label, if corresponding first content of text of the voice messaging and any hobby label of the second user match, Determine that the historical communication data second user of the second user the first content of text corresponding with the voice messaging matches. For example, the hobby label of the second user includes " gourmet ", " opera fan ", " love movement " etc., the voice messaging Corresponding first content of text is " pork braised in brown sauce ", then " pork braised in brown sauce " and " gourmet " phase in the hobby label of second user Match.

If corresponding second content of text of the historical communication data of the second user is corresponding with the voice messaging First content of text matches, and determines that the second user is the target object.It should be noted that in the second user There may be multiple target objects.

In fact, what the historical communication data of the second user included is the daily behavioral data of the second user, So the historical communication data of the second user can embody the daily hobby of the second user.Therefore, here, base In in corresponding second content of text of historical communication data the first text corresponding with the voice messaging of the second user Hold and whether match, determines whether the second user is the target object to correspond to, can be improved determine the mesh in this way The accuracy rate for marking object is avoided as far as possible by commending contents to unwanted user, to improve user experience.

Certainly above-mentioned for how realizing according to corresponding second content of text of historical communication data of the second user, From the second user, the mode of the determining target object to match with first content of text is also applied for following reality The step S604 in example is applied, only the corresponding third content of text of the historical communication data based on third user and described One content of text determines target object.

After the determining target object to match with first content of text, first content of text can be based on With second content of text, content to be recommended is determined, for example, can be with S505 as follows to step S508 to determine State content to be recommended.

In content of text corresponding to S505, the historical communication data based on the target object and first text Hold, determines corresponding alternating content；

It is to be appreciated that being exactly here if the target object is determined from the second user Based on corresponding second content of text and first content of text, corresponding alternating content is determined, so as to more accurate Ground is that the target object recommends the current desired content asked；If the target object is determined from the third user Come, then being namely based on corresponding third content of text and first content of text here, determines corresponding alternating content.

S506, it is based on the alternating content, determines the candidate number of the alternating content；

S507, the alternating content that the candidate number is greater than the second preset threshold is determined as content to be recommended；

S508, the commending contents to be recommended are given to the target object.

It should be noted that a target object can be only determined when determining target object by step S504, it can also With the multiple target objects of determination, then in corresponding second content of text of historical communication data and described first according to target object It, can be first respectively according to the historical communication data pair of each target object when content of text to carry out commending contents for target object The content of text and first content of text answered, determine multiple alternating contents respectively, then, by the most candidate of candidate number Content is determined as the content this time recommended to target object.

For example, can determine 5 target objects, then in the history according to this 5 target objects when determining target object When the corresponding content of text of communication data and first content of text to carry out commending contents for target object, can first it distinguish According in the corresponding content of text of historical communication data and first text of each target object in 5 target objects Appearance determines multiple alternating contents respectively, for example, can be according to target object 1, target object 2 and the corresponding candidate of target object 4 Content is " sleeping beauty "；The corresponding alternating content of target object 3 is " The Frog Prince "；The corresponding alternating content of target object 5 For " lazy Jack ", then it can determine that alternating content is that " sleeping beauty ", " The Frog Prince " and " lazy Jack " respectively corresponds Candidate number be 3 times, 1 time and 1 time, at this point it is possible to which the most alternating content of candidate number " sleeping beauty " is determined as this The content recommended to all target objects.

The embodiment of the present application provides another content recommendation method, including step S601 to step S608:

S601, the voice messaging that the first user is used to indicate recommendation is obtained；

If the voice print database of S602, the voice messaging the first content of text corresponding with the voice messaging is not Match, obtains the address list and communications records of first user；

S603, the communications records are based on, it is determining logical with first user from the address list of first user Believe that number is greater than the third user of the first preset threshold；

S604, the corresponding third content of text of historical communication data according to the third user, from the third user In, the determining target object to match with first content of text；

In practical applications, if the executing subject of step S601 to step S608 are first users corresponding first Terminal, the first terminal can get the chat record of first user Yu the third user from communications applications, The historical communication data of the third user are extracted from the chat record；Alternatively, the first terminal is from communications applications In get the identity of the third user, the identity of third user is then sent to server, is obtained with request Take the historical communication data of third user.

If the executing subject of step S601 to step S608 are servers, if the server determines the voice Corresponding with the voice messaging the first content of text of the voice print database of information mismatches, and the server can request described the The corresponding first terminal of one user sends the identity of the third user to the server；The server is receiving After the identity for the third user that the first terminal is sent, based on the identity of the third user, inquiry The historical communication data of the third user, for example, the information such as video access record, web page access record.

It should be noted that the historical communication data refer to the communication of the third user within a preset time period Data.The historical communication data of the third user may include: the interaction between the third user and first user Information, the history of third user access data (such as broadcasting videograph, browsing webpage record etc.).

After the determining target object to match with first content of text, first content of text can be based on With the third content of text, content to be recommended is determined, for example, can be with S605 as follows to step S608 to determine State content to be recommended.

In content of text corresponding to S605, the historical communication data based on the target object and first text Hold, determines corresponding alternating content；

S606, it is based on the alternating content, determines the candidate number of the alternating content；

S607, the alternating content that the candidate number is greater than the second preset threshold is determined as content to be recommended；

S608, the commending contents to be recommended are given to the target object.

Currently, in recommendation service personalized recommendation can be carried out to the user according to the sound of user, such as in root When determining that the user is old man according to the sound of the user, the opera etc. that old man can be recommended to like to the user.But some In the case of, the user made a sound is not probably the target object this time recommended, for example, when a user is not for oneself But be other users come when issuing play command, if being directly to issue the user of the play command to carry out according to play command Commending contents, then may cause recommended and recommendation mismatches.

In order to effectively solve the above problems, the embodiment of the present application provides a kind of content recommendation method, can determine and this time push away Recommend corresponding true target object, i.e., the target object that the phonetic order issued with first user matches, and to this Target object carries out commending contents.This time recommend corresponding true target object due to that can determine, and carries out in corresponding Hold and recommend, therefore, the embodiment of the present application can effectively solve the problems, such as that above-mentioned recommended and recommendation are unmatched, to mention Rise the accuracy rate of commending contents.

In the embodiment of the present application, the method may include following steps S701 to step S703:

S701, determine whether the first user for issuing phonetic order is this time to recommend corresponding target object；If so, holding Row step S702；Otherwise, step S703 is executed；

It should be noted that the phonetic order is voice messaging described in above-described embodiment, first user The user of recommendation is as indicated described in above-described embodiment.

In the embodiment of the present application, the first terminal that first user is held can receive first user and issue Phonetic order, and the phonetic order is sent to server, server receives the voice that the first terminal uploads and refers to After order, the corresponding voice print database of the phonetic order can be analyzed, and issue the voice based on the analysis results to determine The first user's characteristic information (the i.e. described fisrt feature information) of first user of instruction, such as age, gender；Then, it services Device can carry out speech recognition to the phonetic order, obtain speech recognition result (i.e. described first content of text), server can It, can be true if successful match to match the user's characteristic information of first user with institute speech recognition result Fixed first user is that this recommends corresponding target object；If it fails to match, it can determine that first user is not This time recommend corresponding target object.

In one example, the user's characteristic information of first user is being matched with institute speech recognition result When, it can be first based on institute's speech recognition result, determine corresponding second user characteristic information (the i.e. institute of institute's speech recognition result State second feature information), for example, word segmentation processing first can be carried out to institute's speech recognition result, multiple keywords are obtained, so Afterwards, the word to match with any keyword of the multiple keyword is therefrom determined whether based on preset dictionary, if so, The second user characteristic information can be then determined according to these keywords and preset dictionary.Wherein, in preset dictionary It may include the corresponding relationship of word and user's characteristic information.Determining with after the second user characteristic information, server can be with Determine whether first user's characteristic information and the second user characteristic information are consistent, if unanimously, can determine matching Success；If inconsistent, it can determine that it fails to match.

In one example, first user's characteristic information may include: male, and 30 years old or so；The recognition result It can be with are as follows: " playing a children's story ", then server is by first user's characteristic information and institute's speech recognition result When being matched, word segmentation processing first can be carried out to institute's speech recognition result, obtain " playing ", "one", " children's story " Then these three words have in these three words that server can be determined based on the tables of data in dictionary as shown in table 3 Word (the i.e. described aiming field) relevant to user's characteristic information preset in tables of data is " children's story ", and is determined and " child The matched user's characteristic information of words story " is " 3 years old to 7 years old ".Server can determine that first user's characteristic information is 30 It is year, whether consistent with the second user characteristic information " 3 years old to 7 years old ", since two user's characteristic informations are inconsistent, because This, can determine that it fails to match.

S702, commending contents are carried out for first user according to the corresponding recommendation of the phonetic order；

In other embodiments, however, it is determined that first user is that this recommends corresponding target object, then server can To determine the user's characteristic information of first user according to the corresponding voice print database of the phonetic order, it is then based on described The user's characteristic information and institute's speech recognition result of first user carries out commending contents for first user.

S703, from there is currently second user in determine target object, and to the target object carry out commending contents.

It should be noted that in the present embodiment, however, it is determined that first user is not this time to recommend corresponding target pair As, then server can from there is currently second user in determine target object, and to the target object carry out content push away It recommends.

In one example, server can be whole by the corresponding first terminal determination of first user and described first The current second terminal for establishing communication connection in end, wherein the second terminal can first pass through in advance WiFi connection, bluetooth connection, The modes such as close range wireless communication (Near Field Communication, NFC) connection, infrared connection and the first terminal Establish communication connection.After determining the second terminal, server can by the first terminal come with the second terminal Communication connection is established, and obtains the user's operation data (the i.e. described historical communication data) stored in the second terminal, it is such as clear Look at data, sharing data, screenshotss data etc., then, server can grasp institute's speech recognition result and first user It is matched as data, if successful match, can determine that the corresponding second user of the second terminal is the mesh that this is recommended Mark object；If it fails to match, server can obtain the use that can be used for identifying the second user from the second terminal The user identity information of family identity, such as cell-phone number, login account information, then, according to first user identity information, from The first user's operation data are obtained on the server of the second terminal access, then, server can be by the voice Recognition result is matched with the operation data, and the corresponding second user of the operation data of successful match is determined as this The target object of recommendation.

For example, server can be determined by the first terminal establishes the end that WiFi communication is connect with the first terminal A is held, after determining terminal A, server can obtain the user's operation of terminal A storage by the first terminal from terminal A Data, as video-see records, wherein may include the record for watching children's story in video-see record, then at this point, service Device can be by institute's speech recognition result " playing a children's story ", the user's operation data stored with the terminal A got It is matched, and after successful match, the second user of using terminal A is determined as to the target object this time recommended.

After the target object for determining this time to recommend in multiple second users, server can be according to target pair The phonetic order that the user's operation data of elephant and first user issue carries out commending contents for the target object.

For example, the user's operation data of the second user may include: to play children's story " Cinderella " and " Bai Xue Princess ", then when the phonetic order that first user issues are as follows: when playing a children's story, server can be to described the Two users recommend the children's story " sleeping beauty " to match with children's story " Cinderella " and " Snow White ".

It should be noted that in the present embodiment, server, can only really when determining target object from second user A fixed target object, can also determine multiple target objects, then according to the user's operation data of target object and first It, can be first respectively according to the user of each target object when the phonetic order that user issues to carry out commending contents for target object Operation data and phonetic order determine multiple alternating contents respectively, and then, the most alternating content of candidate number determines thus The secondary content recommended to target object.

For example, server can determine 5 target objects, then in basis when determining target object from second user The phonetic order that the user's operation data of this 5 target objects and the first user issue to carry out commending contents for target object When, it can be first true according to the user's operation data of each target object in 5 target objects and phonetic order difference respectively Fixed multiple alternating contents such as can determine that alternating content is " sleeping beauty " according to the 1st, 2 and 4 target object；According to the 3rd A target object determines that alternating content is " The Frog Prince "；According to the 5th target object, determine that alternating content is " lazy outstanding person Gram ", then it can determine that the corresponding candidate number of 3 alternating contents is respectively 3 times, 1 time and 1 time, at this point it is possible to by candidate The most alternating content of number " sleeping beauty " is determined as the content this time recommended to target object.

Based on embodiment above-mentioned, the embodiment of the present application provides a kind of content recommendation device, which includes included Each unit included by each module and each module can be realized by the processor in electronic equipment；Certainly can also pass through Specific logic circuit is realized；In the process of implementation, processor can for central processing unit (CPU), microprocessor (MPU), Digital signal processor (DSP) or field programmable gate array (FPGA) etc..

Fig. 4 is the composed structure schematic diagram of the embodiment of the present application content recommendation device, as shown in figure 3, described device 400 is wrapped It includes and obtains module 401, the first determining module 402 and recommending module 403；Wherein,

Module 401 is obtained, is configured to obtain the voice messaging that the first user is used to indicate recommendation；

First determining module 402, if the voice print database for being configured to the voice messaging is corresponding with the voice messaging First content of text mismatches, and the determining target object to match with first content of text, the target object is not institute State the first user；

Recommending module 403 is configured to carry out commending contents to the target object.

In other embodiments, the first determining module 402, comprising:

First determination unit is configured that and determines that currently first terminal corresponding with first user is established closely logical Believe the second terminal of connection；

Second determination unit is configured that from the corresponding second user of the second terminal, is determined and first text The target object that content matches.

In other embodiments, the first determining module 402, comprising:

Acquiring unit is configured that the address list and communications records for obtaining first user；

Third determination unit is configured that based on the communications records, from the address list of first user, determine with The number of communications of first user is greater than the third user of the first preset threshold；

4th determination unit is configured that from the third user, determines the mesh to match with first content of text Mark object.

In other embodiments, before triggering the determining module 402, described device 400 further include: second determines mould Block 404, second determining module 404 are configured that the voice print database for determining the voice messaging, in first text Hold and whether matches.

In other embodiments, second determining module 404, comprising:

5th determination unit is configured that the voice print database based on the voice messaging, determines fisrt feature information；It is described Fisrt feature information is used to characterize the personal feature of first user；

6th determination unit is configured that based on first content of text, determines second feature information；Described second is special Reference breath is for characterizing personal feature corresponding with first content of text；

7th determination unit is configured that according to the fisrt feature information and the second feature information, determines institute's predicate Whether the voice print database of message breath, match with first content of text.

In other embodiments, second determination unit is configured that the historical communication data according to the second user Corresponding second content of text, from the second user, the determining target object to match with first content of text.

In other embodiments, the 4th determination unit is configured that the historical communication data according to the third user Corresponding third content of text, from the third user, the determining target object to match with first content of text.

In other embodiments, described device 400 further includes cue module 405, and cue module 405 is configured that if institute The historical communication data for stating second user or each user in the third user and first content of text are not Match, Xiang Suoshu the first user returning response message, the response message is for prompting first user to specify target object.

In other embodiments, the recommending module 403, comprising:

8th determination unit is configured that based on first content of text and second content of text, determines to be recommended Content；Alternatively, being based on first content of text and the third content of text, content to be recommended is determined；

Recommendation unit is configured that the commending contents to be recommended to the target object.

In other embodiments, the 8th determination unit is configured that

Content of text corresponding to historical communication data based on the target object and first content of text determine Corresponding alternating content；

Based on the alternating content, the candidate number of the alternating content is determined；

The alternating content that the candidate number is greater than the second preset threshold is determined as the content to be recommended.

The description of apparatus above embodiment, be with the description of above method embodiment it is similar, have same embodiment of the method Similar beneficial effect.For undisclosed technical detail in the application Installation practice, the application embodiment of the method is please referred to Description and understand.

It should be noted that in the embodiment of the present application, if realizing that above-mentioned content pushes away in the form of software function module Method is recommended, and when sold or used as an independent product, also can store in a computer readable storage medium.Base In such understanding, part that the technical solution of the embodiment of the present application substantially in other words contributes to the relevant technologies can be with The form of software product embodies, which is stored in a storage medium, including some instructions to So that electronic equipment (can be mobile phone, tablet computer, desktop computer, personal digital assistant, navigator, digital telephone, video electricity Words, television set, sensing equipment etc., alternatively, being also possible to server) execute each embodiment the method for the application whole or Part.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (Read Only Memory, ROM), magnetic disk Or the various media that can store program code such as CD.In this way, the embodiment of the present application is not limited to any specific hardware It is combined with software.

Accordingly, the embodiment of the present application provides a kind of electronic equipment, and Fig. 5 is that one kind of the embodiment of the present application electronic equipment is hard Part entity schematic diagram, as shown in figure 5, the hardware entities of the electronic equipment 500 include: including memory 501 and processor 502, The memory 501 is stored with the computer program that can be run on processor 502, and the processor 502 executes described program The step in content recommendation method provided in Shi Shixian above-described embodiment.

Memory 501 is configured to store the instruction and application that can be performed by processor 502, can also cache device to be processed 502 and electronic equipment 500 in each module it is to be processed or processed data (for example, image data, audio data, voice Communication data and video communication data), flash memory (FLASH) or Random Communication memory (Random Access can be passed through Memory, RAM) it realizes.

Accordingly, the embodiment of the present application provides a kind of computer readable storage medium, is stored thereon with computer program, should Step in the content recommendation method provided in above-described embodiment is provided when computer program is executed by processor.

It need to be noted that: the description of medium stored above and apparatus embodiments, with retouching for above method embodiment It is similar for stating, and has with embodiment of the method similar beneficial effect.For in the application storage medium and apparatus embodiments not The technical detail of disclosure please refers to the description of the application embodiment of the method and understands.

It should be understood that " one embodiment " or " embodiment " that specification is mentioned in the whole text mean it is related with embodiment A particular feature, structure, or characteristic includes at least one embodiment of the application.Therefore, occur everywhere in the whole instruction " in one embodiment " or " in one embodiment " not necessarily refer to identical embodiment.In addition, these specific features, knot Structure or characteristic can combine in any suitable manner in one or more embodiments.It should be understood that in the various implementations of the application In example, magnitude of the sequence numbers of the above procedures are not meant that the order of the execution order, the execution sequence Ying Yiqi function of each process It can be determined with internal logic, the implementation process without coping with the embodiment of the present application constitutes any restriction.Above-mentioned the embodiment of the present application Serial number is for illustration only, does not represent the advantages or disadvantages of the embodiments.

It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row His property includes, so that the process, method, article or the device that include a series of elements not only include those elements, and And further include other elements that are not explicitly listed, or further include for this process, method, article or device institute it is intrinsic Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do There is also other identical elements in the process, method of element, article or device.

In several embodiments provided herein, it should be understood that disclosed device and method can pass through it Its mode is realized.Apparatus embodiments described above are merely indicative, for example, the division of the unit, only A kind of logical function partition, there may be another division manner in actual implementation, such as: multiple units or components can combine, or It is desirably integrated into another system, or some features can be ignored or not executed.In addition, shown or discussed each composition portion Mutual coupling or direct-coupling or communication connection is divided to can be through some interfaces, the INDIRECT COUPLING of equipment or unit Or communication connection, it can be electrical, mechanical or other forms.

Above-mentioned unit as illustrated by the separation member, which can be or may not be, to be physically separated, aobvious as unit The component shown can be or may not be physical unit；Both it can be located in one place, and may be distributed over multiple network lists In member；Some or all of units can be selected to achieve the purpose of the solution of this embodiment according to the actual needs.

In addition, each functional unit in each embodiment of the application can be fully integrated in one processing unit, it can also To be each unit individually as a unit, can also be integrated in one unit with two or more units；It is above-mentioned Integrated unit both can take the form of hardware realization, can also realize in the form of hardware adds SFU software functional unit.

Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above method embodiment can pass through The relevant hardware of program instruction is completed, and program above-mentioned can store in computer-readable storage medium, which exists When execution, step including the steps of the foregoing method embodiments is executed；And storage medium above-mentioned includes: movable storage device, read-only deposits The various media that can store program code such as reservoir (Read Only Memory, ROM), magnetic or disk.

If alternatively, the above-mentioned integrated unit of the application is realized in the form of software function module and as independent product When selling or using, it also can store in a computer readable storage medium.Based on this understanding, the application is implemented The technical solution of example substantially in other words can be embodied in the form of software products the part that the relevant technologies contribute, The computer software product is stored in a storage medium, including some instructions use so that electronic equipment (can be mobile phone, Tablet computer, desktop computer, personal digital assistant, navigator, digital telephone, visual telephone, television set, sensing equipment etc., alternatively, Can also be with server) execute each embodiment the method for the application all or part.And storage medium above-mentioned includes: to move The various media that can store program code such as dynamic storage equipment, ROM, magnetic or disk.

The above, only presently filed embodiment, but the protection scope of the application is not limited thereto, it is any to be familiar with Those skilled in the art within the technical scope of the present application, can easily think of the change or the replacement, and should all cover Within the protection scope of the application.Therefore, the protection scope of the application should be based on the protection scope of the described claims.

Claims

1. a kind of content recommendation method, which is characterized in that the described method includes:

If the voice print database of the voice messaging the first content of text corresponding with the voice messaging mismatches, determining and institute The target object that the first content of text matches is stated, the target object is not first user；

Commending contents are carried out to the target object.

2. the method according to claim 1, wherein the mesh that the determination and first content of text match Mark object, comprising:

Determine that first terminal currently corresponding with first user establishes the second terminal of near field communication；From described In the corresponding second user of two terminals, the determining target object to match with first content of text；Alternatively,

Obtain the address list and communications records of first user；Based on the communications records, from the communication of first user In record, the determining third user for being greater than the first preset threshold with the number of communications of first user；From the third user, The determining target object to match with first content of text.

3. the method according to claim 1, wherein matching in the determination and first content of text Before target object, the method also includes:

Whether the voice print database for determining the voice messaging matches with first content of text；

Whether the voice print database of the determination voice messaging, match with first content of text, comprising:

Based on the voice print database of the voice messaging, fisrt feature information is determined；The fisrt feature information is described for characterizing The personal feature of first user；

Based on first content of text, second feature information is determined；The second feature information is for characterizing and described first The corresponding personal feature of content of text；

According to the fisrt feature information and the second feature information, the voice print database of the voice messaging is determined, and it is described Whether the first content of text matches.

4. according to the method described in claim 2, it is characterized in that, from the second user or the third user, really The fixed target object to match with first content of text, comprising:

According to corresponding second content of text of the historical communication data of the second user, from the second user, determine with The target object that first content of text matches；Alternatively,

According to the corresponding third content of text of the historical communication data of the third user, from the third user, determine with The target object that first content of text matches.

5. according to the method described in claim 4, it is characterized in that, the method also includes:

If the second user or the historical communication data of each user in the third user and first text Content mismatches, Xiang Suoshu the first user returning response message, and the response message is for prompting first user specified Target object.

6. according to the method described in claim 5, it is characterized in that, described carry out commending contents to the target object, comprising:

Based on first content of text and second content of text, content to be recommended is determined；Alternatively, based on first text This content and the third content of text, determine content to be recommended；

Give the commending contents to be recommended to the target object.

7. according to the method described in claim 6, it is characterized in that, determination content to be recommended, comprising:

Content of text corresponding to historical communication data based on the target object and first content of text are determined and are corresponded to Alternating content；

8. a kind of content recommendation device characterized by comprising

Determining module, if the voice print database for being configured to the voice messaging the first content of text corresponding with the voice messaging It mismatches, the determining target object to match with first content of text, the target object is not first user；

9. a kind of electronic equipment, including memory and processor, the memory are stored with the calculating that can be run on a processor Machine program, which is characterized in that the processor realizes any one of claim 1 to 7 commending contents when executing described program Step in method.

10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program quilt The step in any one of claim 1 to 7 content recommendation method is realized when processor executes.