CN108182942A - A kind of method and apparatus for supporting different virtual role interactions - Google Patents
A kind of method and apparatus for supporting different virtual role interactions Download PDFInfo
- Publication number
- CN108182942A CN108182942A CN201711464018.7A CN201711464018A CN108182942A CN 108182942 A CN108182942 A CN 108182942A CN 201711464018 A CN201711464018 A CN 201711464018A CN 108182942 A CN108182942 A CN 108182942A
- Authority
- CN
- China
- Prior art keywords
- role
- voice
- information
- voice messaging
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims description 28
- 230000003993 interaction Effects 0.000 title claims description 27
- 238000004891 communication Methods 0.000 claims description 105
- 238000012545 processing Methods 0.000 claims description 75
- 238000004458 analytical method Methods 0.000 claims description 9
- 230000002618 waking effect Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 description 6
- 230000001953 sensory effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000002452 interceptive effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 241001062009 Indigofera Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000010197 meta-analysis Methods 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/28—Constructional details of speech recognition systems
- G10L15/30—Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention discloses a kind of method and apparatus for supporting different virtual role interactions, described device includes terminal and server, and the terminal includes the first communication unit, first processing units, the first storage unit, voice playing unit；First storage unit includes multiple first memory modules；The server includes second processing unit, the second communication unit and the second storage unit, and second storage unit includes multiple second memory modules.By setting different storage units, the voice data of each role is disaggregatedly stored, simultaneously after each voice, historic role voice messaging in this voice communication data and the second memory module is subjected to statistical learning, generate new role's voice messaging, and new role's voice messaging is updated in the first memory module and the second memory module so that intelligent sound replys the demand for being more in line with user, effectively improves the sensory experience of user.
Description
Technical field
The present invention relates to field of human-computer interaction more particularly to a kind of method and apparatus for supporting different virtual role interactions.
Background technology
Intelligent sound interaction is that the international major company such as hot spot in recent years, Amazon, Google is successively all proposed intelligent language
Sound Related product (such as intelligent sound box) respectively marches the entrance of artificial intelligence field, domestic enterprise (such as Jingdone district, hundred as propulsion
Degree, Alibaba) etc. companies also have layout in intelligent sound field.However, up to now, either intelligent sound box is still
Other intelligent interaction products, mainly or based on single content reply or single role interaction.Although Google
The sound of different people can be identified, be treated same problem needle with the intelligent sound function of facing multiple users by being also proposed
Corresponding answer can be provided to different people to reply, but still be limited to for single role, if asking of being putd question to of multiple users
Inscribe it is multifarious, intelligent robot say reply voice often there is it is irrelevant, give an irrelevant answer phenomena such as, significantly impact
User's sensory experience.
Invention content
For this reason, it may be necessary to provide a kind of technical solution for supporting the interaction of different virtual roles, to solve in the prior art by
In intelligent sound interaction, role is single, function is limited, leads to that speech answering is inaccurate, poor user experience.
To achieve the above object, a kind of device for supporting different virtual role interactions, described device packet are inventor provided
Include terminal and server；The terminal is connect with server；The terminal includes the first communication unit, first processing units, the
One storage unit, voice playing unit；First storage unit includes multiple first memory modules；First processing units are distinguished
It is connect with the first communication unit, the first storage unit, voice playing unit；The server includes second processing unit, second
Communication unit and the second storage unit, the second processing unit are connect respectively with the second communication unit, the second storage unit, institute
The first communication unit and the connection of the second communication unit are stated, second storage unit includes multiple second memory modules；
The first processing units are used to obtain the key word information of multiple roles successively, according to key word information and role
The correspondence of information obtains the Role Information of the second communication unit transmission, and Role Information successively by the first communication unit
It stores into the first memory module, every one first memory module corresponds to one Role Information of storage；The key word information includes angle
Color ID, the Role Information include role's voice messaging；
The first processing units are additionally operable to obtain the first voice messaging input by user, and language is carried out to the first voice messaging
Sound identifies, determines the corresponding role ID of the voice messaging, obtains role's voice messaging of the role ID；And for receiving user
Second voice messaging of input, and semantic analysis is carried out to the second voice messaging, the second language is taken according to the generation of role's voice messaging
Message ceases corresponding speech answering information, and passes through voice playing unit and send out the speech answering information；
The first processing units are additionally operable to when receiving voice END instruction, this is sent by the first communication unit
Voice communication data to the second communication unit, the voice communication data include this call all second voice messagings and
Speech answering information；
The second processing unit is used for the historic role voice in this voice communication data and the second memory module
Information carries out statistical learning, generates new role's voice messaging, and new role's voice messaging is updated in the second memory module,
And for new role's voice messaging to be sent to the first communication unit by the second communication unit；
The first processing units are used to receive new role's voice messaging by the first communication unit, and be stored in
In corresponding first memory module.
Further, the speech answering information includes voice content information and voice configuration information, the speech play
Unit is used to play the voice content information according to the voice configuration information.
Further, default role is also stored in first memory module and wakes up information, the first processing units
Information is waken up for receiving role, when the role that receives of judgement, which wakes up information, wakes up information matches with default role, from the
Role's voice messaging is obtained in one memory module.
Further, the first processing units are additionally operable to receive role's operational order, according to role's operational order to the
Role's voice messaging in one memory module performs corresponding operating, and role's operational order includes role ID to be operated.
Inventor additionally provides a kind of method for supporting different virtual role interactions, and the method is applied to support different void
Intend the device of role's interaction, described device includes terminal and server；The terminal is connect with server；The terminal includes the
One communication unit, first processing units, the first storage unit, voice playing unit；First storage unit includes multiple the
One memory module；First processing units are connect respectively with the first communication unit, the first storage unit, voice playing unit；It is described
Server includes second processing unit, the second communication unit and the second storage unit, and the second processing unit is respectively with second
Communication unit, the connection of the second storage unit, first communication unit and the connection of the second communication unit, second storage unit
Including multiple second memory modules；It the described method comprises the following steps：
First processing units obtain the key word information of multiple roles successively, according to key word information and pair of Role Information
It should be related to, the Role Information that the second communication unit sends is obtained, and Role Information is stored to the successively by the first communication unit
In one memory module, every one first memory module corresponds to one Role Information of storage；The key word information includes role ID, described
Role Information includes role's voice messaging；
First processing units obtain the first voice messaging input by user, carry out speech recognition to the first voice messaging, really
The corresponding role ID of the fixed voice messaging obtains role's voice messaging of the role ID；And receive the second language input by user
Message ceases, and carries out semantic analysis to the second voice messaging, takes the second voice messaging corresponding according to the generation of role's voice messaging
Speech answering information, and pass through voice playing unit and send out the speech answering information；
First processing units send this voice communication number when receiving voice END instruction, by the first communication unit
According to the second communication unit, the voice communication data include all second voice messagings of this call and speech answering letter
Breath；
Second processing unit carries out the historic role voice messaging in this voice communication data and the second memory module
Statistical learning generates new role's voice messaging, and new role's voice messaging is updated in the second memory module and is used for
New role's voice messaging is sent to the first communication unit by the second communication unit；
First processing units receive new role's voice messaging by the first communication unit, and are stored in corresponding the
In one memory module.
Further, the speech answering information includes voice content information and voice configuration information, the method includes：
Voice playing unit plays the voice content information according to the voice configuration information.
Further, default role is also stored in first memory module and wakes up information, the method includes：
First processing units receive role and wake up information, are waken up when the role that judgement receives wakes up information with default role
During information matches, role's voice messaging is obtained from the first memory module.
Further, the method includes：
First processing units receive role's operational order, according to role's operational order to role's language in the first memory module
Message breath performs corresponding operating, and role's operational order includes role ID to be operated.
The prior art is different from, the method and apparatus of the support difference virtual role interaction described in above-mentioned technical proposal, institute
It states device and includes terminal and server, the terminal includes the first communication unit, first processing units, the first storage unit, language
Sound broadcast unit；First storage unit includes multiple first memory modules；The server includes second processing unit, the
Two communication units and the second storage unit, second storage unit include multiple second memory modules.It is different by setting
Storage unit disaggregatedly stores the voice data of each role, while after each voice, by this
Historic role voice messaging in voice communication data and the second memory module carries out statistical learning, generates new role's voice letter
Breath, and new role's voice messaging is updated in the first memory module and the second memory module so that intelligent sound is replied more
Meet the demand of user, effectively improve the sensory experience of user.
Description of the drawings
Fig. 1 is the schematic diagram of the device of support difference virtual role interaction that one embodiment of the invention is related to；
Fig. 2 is the terminal and the connection diagram in intelligent high in the clouds that one embodiment of the invention is related to；
Fig. 3 is the flow chart of the method for support difference virtual role interaction that one embodiment of the invention is related to；
Reference sign：
101st, terminal；
111st, the first communication unit；112nd, first processing units；113rd, the first storage unit；1131st, the first memory module；
114th, voice playing unit；
102nd, server；
121st, second processing unit；122nd, the second communication unit；123rd, the second storage unit；1231st, the second memory module.
Specific embodiment
For the technology contents of technical solution, construction feature, the objects and the effects are described in detail, below in conjunction with specific reality
It applies example and attached drawing is coordinated to be explained in detail.
Referring to Fig. 1, the schematic diagram of the device of support difference virtual role interaction being related to for one embodiment of the invention.Institute
It states device and includes terminal 101 and server 102；The terminal 101 is connect with server 102；It is logical that the terminal 101 includes first
Believe unit 111, first processing units 112, the first storage unit 113, voice playing unit 114；First storage unit 113
Including multiple first memory modules 1131；First processing units 112 respectively with the first communication unit 111, the first storage unit
113rd, voice playing unit 114 connects；The server 102 includes second processing unit 121, the second communication unit 122 and the
Two storage units 123, the second processing unit 121 are connect respectively with the second communication unit 122, the second storage unit 123, institute
It states the first communication unit 111 and the second communication unit 122 connects, second storage unit 123 includes the multiple second storage moulds
Block 1231.
The terminal is the electronic equipment with data processing function and voice play function, such as mobile phone, tablet, personal meter
Calculation machine etc., the server can select cloud server.First communication unit and the second communication unit can pass through indigo plant
Tooth, WIFI etc. communicate.The first processing units and second processing unit are the electronic component with data processing function,
Such as CPU, DSP, MCU processor.First storage unit and the second storage unit are the electronics with data storage function
Element, such as hard disk, floppy disk, USB flash disk, the first memory module are the submodule of the first storage unit, and the second memory module is second
The submodule of storage unit, voice playing unit can be microphone.
The first processing units 112 are used to obtain the key word information of multiple roles successively, according to key word information with
The correspondence of Role Information, the Role Information sent by the first communication unit 111 the second communication unit 122 of acquisition, and according to
Secondary Role Information is stored into the first memory module 1131, and every one first memory module should store a Role Information to 1131.It closes
Keyword information can be by receiving information acquisition input by user or providing several alternative key word informations to user
Selection is determined by detecting the selection instruction of user.
The Role Information includes role's voice messaging, and server can be previously stored many different key word informations and correspond to
Speech robot people initial speech packet.The key word information is given including type belonging to role ID, role etc., such as user
Role 1 is named as " Xiao Ming ", and sets the type belonging to it as " being good at sports genre ", then first processor will be from server
" being good at sports genre " relevant robot voice packet is got, and is stored in 1 corresponding first memory module of role
In 1131, such as in the first memory module A that can be stored in Fig. 1.Certainly, user can also be named as " small to role 2
It is red ", and the type belonging to it is set as " being good at literature type ", then first processor will get from server and " be good at text
The relevant robot initial voice packet of type ", and be stored in 2 corresponding first memory module 1131 of role, such as
It can be stored in the first memory module B in Fig. 1.Similarly, user can obtain different intelligence by the above method from server
The initial speech packet of energy interactive voice robot, and classify and stored, to call acquisition at any time.
The first processing units 112 are additionally operable to obtain the first voice messaging input by user, to the first voice messaging into
Row speech recognition determines the corresponding role ID of the voice messaging, obtains role's voice messaging of the role ID；And for receiving
Second voice messaging input by user, and semantic analysis is carried out to the second voice messaging, take the according to the generation of role's voice messaging
The corresponding speech answering information of two voice messagings, and pass through voice playing unit and send out the speech answering information.Described first
Voice messaging is the voice messaging for waking up corresponding role, can be the key word information that user has been previously set, such as user can
With by the way that terminal call " Xiao Ming ", terminal will be transferred after analyzing voice input by user in the first memory module A
Data so as to subsequent calls.Certainly, the problem of user can also be by puing question to some targeted, such as when the first processing is single
When the first voice messaging that meta analysis goes out user includes the relevant vocabulary of sport or problem, role 1 can also be automatically waken up,
It is interacted so as to more targeted with user.Second voice messaging is the voice data that is normally carried out of user, and user can be with
Interacted in a manner of puing question to or chatting with terminal, first processing units by the second voice messaging carry out semantic analysis,
The corresponding speech answering information of second voice messaging is taken, and pass through voice playing unit and send out institute according to the generation of role's voice messaging
Predicate sound return information.Due to being classified to role so that more specific aim when each role answers user's query,
The content of answer is more in line with user demand, effectively improves user experience.
The first processing units 112 are additionally operable to when receiving voice END instruction, are sent out by the first communication unit 111
This voice communication data is given to the second communication unit, the voice communication data include all second voices letter of this call
Breath and speech answering information；The second processing unit 122 is used for this voice communication data and the second memory module
Historic role voice messaging in 1231 carries out statistical learning, generates new role's voice messaging, and new role's voice messaging
It is updated in the second memory module 1231 and for new role's voice messaging to be sent to by the second communication unit 122
First communication unit 111；The first processing units 112 are used to receive new role's voice letter by the first communication unit 111
Breath, and be stored in corresponding first memory module 1131.
Voice END instruction can terminate to trigger during this voice communication by user, touch when can also be by closing terminal
Hair.As shown in Fig. 2, the voice data of the role A in terminal is stored in role's A service clouds in intelligent high in the clouds (being equivalent to server)
(being equivalent to the second memory module), each user terminate all voice data with after role's A voice communications, this is conversed
(including receiving voice data and speech answering voice input by user) compressed package sent is to cloud intelligence high in the clouds, intelligent cloud
The data stored in voice data and role's A service clouds can be carried out comprehensive analysis by end, generate new role's voice messaging, and will
The new corresponding voice data of role A is updated in the second memory module.Specifically, server can believe historic role voice
Breath (has been previously stored in this corresponding voice messaging of call role in the second memory module) before this voice communication
The weighted value different from this voice communication data configuration, so as to which comprehensive analysis obtains new voice messaging, weighted value can root
It is determined according to timestamp, usually, the weighted values of the last voice communication data is higher than the weighted value conversed more forward the time.
Such as putd question to (problem is same) for best 10 times of user, the voice data that preceding 5 role A are answered is " D ", latter 5 times
The voice data that role A is answered is " E ", then in the voice messaging of role A in the updated, since number is identical and " E "
Weighted value bigger, thus when user puts question to same problem again, " E " will be selected to be played out as speech answering data.Again
Such as an identical problem, the voice data that preceding 9 role A are answered is " D ", the voice number of rear 1 role A answers
According to for " E ", although the weighted value bigger of the answer of last time, since the answers of first 9 times are all " D ", do not carried in user
Go out clear and definite negative acknowledge it is incorrect under the premise of, undoubtedly " D " is more suitable for the answer of this enquirement, thus angle in the updated
In the voice messaging of color A, " D " can be selected to be played out as speech answering information for this problem.The mode of broadcasting includes
Receiver plays, microphone plays etc..In this way, the real-time adjustment by server calculates, speech robot people is made by intelligence learning
Obtaining language return information more allows user to be satisfied with, and promotes user's sensory experience.
In certain embodiments, the speech answering information includes voice content information and voice configuration information, institute's predicate
Sound broadcast unit is used to play the voice content information according to the voice configuration information.The voice configuration information includes language
Gas information, timbre information etc..User can set in the key word information for setting certain role when the role speaks together
Tone color, tone etc. so that speech answering information is played out with preset voice configuration information, promote user experience.When
So, user can also input the tone of voice to user and intonation is analyzed, and the corresponding tone and intonation is selected to play
Voice content information, such as the second voice messaging tone input by user are higher, then the tone that tone can be selected more gentle
Voice content information is played, so as to effectively promote user experience.
In certain embodiments, default role is also stored in first memory module and wakes up information, at described first
Reason unit wakes up information for receiving role, and information matches are waken up with default role when the role that judgement receives wakes up information
When, role's voice messaging is obtained from the first memory module.Other than role ID, user can set according to actual needs
Role wakes up information, and interactive voice is carried out so as to quickly and conveniently wake up role.Default role, which wakes up information, can include word
Information, voice messaging, face information etc..
In certain embodiments, the first processing units are additionally operable to receive role's operational order, are referred to according to role's operation
It enables and corresponding operating is performed to role's voice messaging in the first memory module, role's operational order includes angle to be operated
Color ID.Role's operational order includes the instructions such as increasing role, delete, changing, looking into, consequently facilitating carrying out pipe to multiple roles
Reason.
As shown in figure 3, inventor providing a kind of method for supporting different virtual role interactions, the method is applied to branch
The device of different virtual role interactions is held, described device includes terminal and server；The terminal is connect with server；The end
End includes the first communication unit, first processing units, the first storage unit, voice playing unit；The first storage unit packet
Include multiple first memory modules；First processing units connect respectively with the first communication unit, the first storage unit, voice playing unit
It connects；The server includes second processing unit, the second communication unit and the second storage unit, the second processing unit difference
It is connect with the second communication unit, the second storage unit, first communication unit and the connection of the second communication unit, described second deposits
Storage unit includes multiple second memory modules；It the described method comprises the following steps：
The key word information that step S301 first processing units obtain multiple roles successively is initially entered, is believed according to keyword
The correspondence of breath and Role Information obtains the Role Information of the second communication unit transmission by the first communication unit, and successively
Role Information is stored into the first memory module, and every one first memory module corresponds to one Role Information of storage；The keyword letter
Breath includes role ID, and the Role Information includes role's voice messaging；
It then enters step S302 first processing units and obtains the first voice messaging input by user, to the first voice messaging
Speech recognition is carried out, the corresponding role ID of the voice messaging is determined, obtains role's voice messaging of the role ID；And it receives and uses
Second voice messaging of family input, and semantic analysis is carried out to the second voice messaging, take second according to the generation of role's voice messaging
The corresponding speech answering information of voice messaging, and pass through voice playing unit and send out the speech answering information；
S303 first processing units are then entered step when receiving voice END instruction, are sent out by the first communication unit
This voice communication data is given to the second communication unit, the voice communication data include all second voices letter of this call
Breath and speech answering information；
S304 second processings unit is then entered step by the history in this voice communication data and the second memory module
Role's voice messaging carries out statistical learning, generates new role's voice messaging, and new role's voice messaging is updated to second and deposits
The first communication unit is sent to by the second communication unit in storage module and by new role's voice messaging；
It then enters step S305 first processing units and new role's voice messaging is received by the first communication unit, and will
It is stored in corresponding first memory module.
The above method disaggregatedly deposits the voice data of each role by setting different storage units
Storage, while after each voice, by the historic role voice messaging in this voice communication data and the second memory module
Statistical learning is carried out, generates new role's voice messaging, and new role's voice messaging is updated to the first memory module and second
In memory module so that intelligent sound replys the demand for being more in line with user, effectively improves the sensory experience of user.
In certain embodiments, the speech answering information includes voice content information and voice configuration information, the side
Method includes：Voice playing unit plays the voice content information according to the voice configuration information.The voice configuration information
Including tone information, timbre information etc..User can set the role and say together in the key word information for setting certain role
Tone color, the tone during words etc. so that speech answering information is played out with preset voice configuration information, promote user's body
It tests.Certainly, user can also input the tone of voice to user and intonation is analyzed, and the corresponding tone and intonation is selected
Broadcasting voice content information, such as the second voice messaging tone input by user are higher, then can select tone more tenderness
The tone plays voice content information, so as to effectively promote user experience.
In certain embodiments, default role is also stored in first memory module and wakes up information, the method packet
It includes：First processing units receive role and wake up information, and information is waken up with default role when the role that judgement receives wakes up information
During matching, role's voice messaging is obtained from the first memory module.Other than role ID, user can set according to actual needs
Determine role and wake up information, interactive voice is carried out so as to quickly and conveniently wake up role.Default role, which wakes up information, can include text
Word information, voice messaging, face information etc..
In certain embodiments, the method includes：First processing units receive role's operational order, are operated according to role
It instructs and corresponding operating is performed to role's voice messaging in the first memory module, role's operational order includes to be operated
Role ID.Role's operational order includes the instructions such as increasing role, delete, changing, looking into, consequently facilitating being carried out to multiple roles
Management.
It should be noted that although the various embodiments described above have been described herein, not thereby limit
The scope of patent protection of the present invention.Therefore, based on the present invention innovative idea, to embodiment described herein carry out change and repair
The equivalent structure or equivalent flow shift for changing or being made using description of the invention and accompanying drawing content, directly or indirectly will be with
Upper technical solution is used in other related technical areas, is included within the scope of patent protection of the present invention.
Claims (8)
1. a kind of device for supporting different virtual role interactions, which is characterized in that described device includes terminal and server；It is described
Terminal is connect with server；The terminal includes the first communication unit, first processing units, the first storage unit, speech play
Unit；First storage unit includes multiple first memory modules；First processing units respectively with the first communication unit, first
Storage unit, voice playing unit connection；It is single that the server includes second processing unit, the second communication unit and the second storage
Member, the second processing unit are connect respectively with the second communication unit, the second storage unit, first communication unit and second
Communication unit connects, and second storage unit includes multiple second memory modules；
The first processing units are used to obtain the key word information of multiple roles successively, according to key word information and Role Information
Correspondence, by the first communication unit obtain the second communication unit send Role Information, and successively Role Information store
Into the first memory module, every one first memory module corresponds to one Role Information of storage；The key word information includes role ID,
The Role Information includes role's voice messaging；
The first processing units are additionally operable to obtain the first voice messaging input by user, and voice knowledge is carried out to the first voice messaging
Not, it determines the corresponding role ID of the voice messaging, obtains role's voice messaging of the role ID；And for receiving user's input
The second voice messaging, and to the second voice messaging carry out semantic analysis, according to role's voice messaging generation take the second voice believe
Corresponding speech answering information is ceased, and passes through voice playing unit and sends out the speech answering information；
The first processing units are additionally operable to when receiving voice END instruction, this voice is sent by the first communication unit
Communicating data to the second communication unit, the voice communication data include all second voice messagings and voice of this call
Return information；
The second processing unit is used for the historic role voice messaging in this voice communication data and the second memory module
Carry out statistical learning, generate new role's voice messaging, and new role's voice messaging be updated in the second memory module and
For new role's voice messaging to be sent to the first communication unit by the second communication unit；
The first processing units are used to receive new role's voice messaging by the first communication unit, and be stored in corresponding to
The first memory module in.
2. the device of different virtual role interactions is supported as described in claim 1, which is characterized in that the speech answering information
Including voice content information and voice configuration information, the voice playing unit is used to play institute according to the voice configuration information
Predicate sound content information.
3. the device of different virtual role interactions is supported as described in claim 1, which is characterized in that first memory module
In be also stored with default role and wake up information, the first processing units wake up information for receiving role, when judgement receives
Role when waking up information and default role and waking up information matches, role's voice messaging is obtained from the first memory module.
4. the device of different virtual role interactions is supported as described in claim 1, which is characterized in that the first processing units
It is additionally operable to receive role's operational order, role's voice messaging in the first memory module be performed according to role's operational order corresponding
Operation, role's operational order include role ID to be operated.
A kind of 5. method for supporting different virtual role interactions, which is characterized in that the method is applied to support different virtual angles
The device of color interaction, described device include terminal and server；The terminal is connect with server；It is logical that the terminal includes first
Believe unit, first processing units, the first storage unit, voice playing unit；First storage unit is deposited including multiple first
Store up module；First processing units are connect respectively with the first communication unit, the first storage unit, voice playing unit；The service
Device includes second processing unit, the second communication unit and the second storage unit, and the second processing unit communicates respectively with second
Unit, the connection of the second storage unit, first communication unit and the connection of the second communication unit, second storage unit include
Multiple second memory modules；It the described method comprises the following steps：
First processing units obtain the key word information of multiple roles successively, are closed according to key word information is corresponding with Role Information
System obtains the Role Information that the second communication unit sends, and Role Information is stored to first and deposited successively by the first communication unit
It stores up in module, every one first memory module corresponds to one Role Information of storage；The key word information includes role ID, the role
Information includes role's voice messaging；
First processing units obtain the first voice messaging input by user, and speech recognition is carried out to the first voice messaging, and determining should
The corresponding role ID of voice messaging obtains role's voice messaging of the role ID；And receive the second voice letter input by user
Breath, and semantic analysis is carried out to the second voice messaging, the corresponding voice of the second voice messaging is taken according to the generation of role's voice messaging
Return information, and pass through voice playing unit and send out the speech answering information；
First processing units send this voice communication data extremely when receiving voice END instruction, by the first communication unit
Second communication unit, the voice communication data include all second voice messagings of this call and speech answering information；
Second processing unit counts the historic role voice messaging in this voice communication data and the second memory module
Study generates new role's voice messaging, and new role's voice messaging is updated in the second memory module and for inciting somebody to action newly
Role's voice messaging the first communication unit is sent to by the second communication unit；
First processing units receive new role's voice messaging by the first communication unit, and are stored in corresponding first and deposit
It stores up in module.
6. the method for different virtual role interactions is supported as claimed in claim 5, which is characterized in that the speech answering information
Including voice content information and voice configuration information, the method includes：
Voice playing unit plays the voice content information according to the voice configuration information.
7. the method for different virtual role interactions is supported as claimed in claim 5, which is characterized in that first memory module
In be also stored with default role and wake up information, the method includes：
First processing units receive role and wake up information, and information is waken up with default role when the role that judgement receives wakes up information
During matching, role's voice messaging is obtained from the first memory module.
8. the method for different virtual role interactions is supported as claimed in claim 5, which is characterized in that the method includes：
First processing units receive role's operational order, and role's voice in the first memory module is believed according to role's operational order
Breath performs corresponding operating, and role's operational order includes role ID to be operated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711464018.7A CN108182942B (en) | 2017-12-28 | 2017-12-28 | Method and device for supporting interaction of different virtual roles |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711464018.7A CN108182942B (en) | 2017-12-28 | 2017-12-28 | Method and device for supporting interaction of different virtual roles |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108182942A true CN108182942A (en) | 2018-06-19 |
| CN108182942B CN108182942B (en) | 2021-11-26 |
Family
ID=62548650
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711464018.7A Active CN108182942B (en) | 2017-12-28 | 2017-12-28 | Method and device for supporting interaction of different virtual roles |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108182942B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109360558A (en) * | 2018-10-15 | 2019-02-19 | 珠海格力电器股份有限公司 | Voice response method and device |
| CN110399474A (en) * | 2019-07-18 | 2019-11-01 | 腾讯科技（深圳）有限公司 | A kind of Intelligent dialogue method, apparatus, equipment and storage medium |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1573923A (en) * | 2003-05-27 | 2005-02-02 | 微软公司 | System and method for user modeling to enhance named entity recognition |
| CN101266793A (en) * | 2007-03-14 | 2008-09-17 | 财团法人工业技术研究院 | Device and method for reducing recognition error through context relationship between dialog rounds |
| JP2013190473A (en) * | 2012-03-12 | 2013-09-26 | Brother Ind Ltd | Voice identification device and program |
| CN104423552A (en) * | 2013-09-03 | 2015-03-18 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| US20150081303A1 (en) * | 1999-06-10 | 2015-03-19 | West View Research, Llc | Personnel transport device with computerized information and display apparatus |
| CN104732968A (en) * | 2013-12-20 | 2015-06-24 | 携程计算机技术（上海）有限公司 | Voice control system evaluation system and method |
| CN104932862A (en) * | 2014-03-21 | 2015-09-23 | 朱江 | Multi-role interactive method based on voice recognition |
| US9189742B2 (en) * | 2013-11-20 | 2015-11-17 | Justin London | Adaptive virtual intelligent agent |
| CN105830058A (en) * | 2013-12-16 | 2016-08-03 | 三菱电机株式会社 | Dialog manager |
| CN105868360A (en) * | 2016-03-29 | 2016-08-17 | 乐视控股（北京）有限公司 | Content recommendation method and device based on voice recognition |
| US20160247502A1 (en) * | 2015-02-23 | 2016-08-25 | Electronics And Telecommunications Research Institute | Audio signal processing apparatus and method robust against noise |
| CN105975622A (en) * | 2016-05-28 | 2016-09-28 | 蔡宏铭 | Multi-role intelligent chatting method and system |
| CN106448670A (en) * | 2016-10-21 | 2017-02-22 | 竹间智能科技（上海）有限公司 | Dialogue automatic reply system based on deep learning and reinforcement learning |
| CN106844627A (en) * | 2017-01-20 | 2017-06-13 | 竹间智能科技（上海）有限公司 | On-line study method and device based on conversational system |
| US20170274535A1 (en) * | 2016-03-23 | 2017-09-28 | Electronics And Telecommunications Research Institute | Interaction device and interaction method thereof |
| CN107277672A (en) * | 2017-06-07 | 2017-10-20 | 福州瑞芯微电子股份有限公司 | A kind of method and apparatus for supporting awakening mode to automatically switch |
| CN107340991A (en) * | 2017-07-18 | 2017-11-10 | 百度在线网络技术（北京）有限公司 | Switching method, device, equipment and the storage medium of speech roles |
-
2017
- 2017-12-28 CN CN201711464018.7A patent/CN108182942B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150081303A1 (en) * | 1999-06-10 | 2015-03-19 | West View Research, Llc | Personnel transport device with computerized information and display apparatus |
| CN1573923A (en) * | 2003-05-27 | 2005-02-02 | 微软公司 | System and method for user modeling to enhance named entity recognition |
| CN101266793A (en) * | 2007-03-14 | 2008-09-17 | 财团法人工业技术研究院 | Device and method for reducing recognition error through context relationship between dialog rounds |
| JP2013190473A (en) * | 2012-03-12 | 2013-09-26 | Brother Ind Ltd | Voice identification device and program |
| CN104423552A (en) * | 2013-09-03 | 2015-03-18 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| US9189742B2 (en) * | 2013-11-20 | 2015-11-17 | Justin London | Adaptive virtual intelligent agent |
| CN105830058A (en) * | 2013-12-16 | 2016-08-03 | 三菱电机株式会社 | Dialog manager |
| CN104732968A (en) * | 2013-12-20 | 2015-06-24 | 携程计算机技术（上海）有限公司 | Voice control system evaluation system and method |
| CN104932862A (en) * | 2014-03-21 | 2015-09-23 | 朱江 | Multi-role interactive method based on voice recognition |
| US20160247502A1 (en) * | 2015-02-23 | 2016-08-25 | Electronics And Telecommunications Research Institute | Audio signal processing apparatus and method robust against noise |
| US20170274535A1 (en) * | 2016-03-23 | 2017-09-28 | Electronics And Telecommunications Research Institute | Interaction device and interaction method thereof |
| CN105868360A (en) * | 2016-03-29 | 2016-08-17 | 乐视控股（北京）有限公司 | Content recommendation method and device based on voice recognition |
| CN105975622A (en) * | 2016-05-28 | 2016-09-28 | 蔡宏铭 | Multi-role intelligent chatting method and system |
| CN106448670A (en) * | 2016-10-21 | 2017-02-22 | 竹间智能科技（上海）有限公司 | Dialogue automatic reply system based on deep learning and reinforcement learning |
| CN106844627A (en) * | 2017-01-20 | 2017-06-13 | 竹间智能科技（上海）有限公司 | On-line study method and device based on conversational system |
| CN107277672A (en) * | 2017-06-07 | 2017-10-20 | 福州瑞芯微电子股份有限公司 | A kind of method and apparatus for supporting awakening mode to automatically switch |
| CN107340991A (en) * | 2017-07-18 | 2017-11-10 | 百度在线网络技术（北京）有限公司 | Switching method, device, equipment and the storage medium of speech roles |
Non-Patent Citations (2)
| Title |
|---|
| I. POTAMITIS，等: "AN INTEGRATED SYSTEM FOR SMART-HOME CONTROL OF APPLIANCES BASED ON REMOTE SPEECH INTERACTION", 《EUROSPEECH》 * |
| 李真，等: "具有情感和语音交互能力的虚拟人系统", 《计算机工程》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109360558A (en) * | 2018-10-15 | 2019-02-19 | 珠海格力电器股份有限公司 | Voice response method and device |
| CN109360558B (en) * | 2018-10-15 | 2021-02-09 | 珠海格力电器股份有限公司 | Voice response method and device |
| CN110399474A (en) * | 2019-07-18 | 2019-11-01 | 腾讯科技（深圳）有限公司 | A kind of Intelligent dialogue method, apparatus, equipment and storage medium |
| CN110399474B (en) * | 2019-07-18 | 2023-06-09 | 腾讯科技（深圳）有限公司 | Intelligent dialogue method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108182942B (en) | 2021-11-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022095380A1 (en) | Ai-based virtual interaction model generation method and apparatus, computer device and storage medium | |
| CN106951468B (en) | Talk with generation method and device | |
| CN107797984B (en) | Intelligent interaction method, equipment and storage medium | |
| WO2015090137A1 (en) | A voice message search method, device, and system | |
| CN107506166A (en) | Information cuing method and device, computer installation and readable storage medium storing program for executing | |
| CN105704013A (en) | Context-based topic updating data processing method and apparatus | |
| CN109981910A (en) | Business recommended method and apparatus | |
| CN111694940A (en) | User report generation method and terminal equipment | |
| US10659605B1 (en) | Automatically unsubscribing from automated calls based on call audio patterns | |
| CN110825863B (en) | Text pair fusion method and device | |
| CN110766442A (en) | Customer information verification method, device, computer equipment and storage medium | |
| CN108920649A (en) | A kind of information recommendation method, device, equipment and medium | |
| US20220059094A1 (en) | Transcription of audio | |
| US11716394B2 (en) | Arrangements for detecting bi-directional artificial intelligence (AI) voice communications and negotiating direct digital communications | |
| US12086727B2 (en) | Modified media detection | |
| CN112242135A (en) | A method for processing voice data and an intelligent customer service device | |
| CN117253478A (en) | Voice interaction method and related device | |
| CN108182942A (en) | A kind of method and apparatus for supporting different virtual role interactions | |
| CN110111795B (en) | Voice processing method and terminal equipment | |
| CN109165286A (en) | Automatic question-answering method, device and computer readable storage medium | |
| CN114242047B (en) | Voice processing method and device, electronic equipment and storage medium | |
| CN108257600A (en) | Method of speech processing and device | |
| CN106910075A (en) | Intelligent processing system and method that client mobile communication is complained | |
| CN118787935A (en) | A voice interactive control system based on online games | |
| WO2024174714A1 (en) | Authenticity verification method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 350003 building 18, No.89, software Avenue, Gulou District, Fuzhou City, Fujian Province Applicant after: Ruixin Microelectronics Co., Ltd Address before: 350003 building 18, No.89, software Avenue, Gulou District, Fuzhou City, Fujian Province Applicant before: FUZHOU ROCKCHIP ELECTRONICS Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |