CN109840009A - A kind of intelligence true man's advertisement screen interactive system and implementation method - Google Patents

A kind of intelligence true man's advertisement screen interactive system and implementation method Download PDF

Info

Publication number
CN109840009A
CN109840009A CN201711212325.6A CN201711212325A CN109840009A CN 109840009 A CN109840009 A CN 109840009A CN 201711212325 A CN201711212325 A CN 201711212325A CN 109840009 A CN109840009 A CN 109840009A
Authority
CN
China
Prior art keywords
server
true man
voice
intelligent
dialogue
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.)
Pending
Application number
CN201711212325.6A
Other languages
Chinese (zh)
Inventor
吴杰
谢伟平
张锐
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang Thinker Technology Co Ltd
Original Assignee
Zhejiang Thinker Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang Thinker Technology Co Ltd filed Critical Zhejiang Thinker Technology Co Ltd
Priority to CN201711212325.6A priority Critical patent/CN109840009A/en
Publication of CN109840009A publication Critical patent/CN109840009A/en
Pending legal-status Critical Current

Links

Landscapes

  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

The present invention relates to artificial intelligence and advertisement media field, a kind of intelligent true man's advertisement screen interactive system and implementation method are disclosed.System includes: intelligent true man's server, advertisement video playing module, sound collection and playing module, image capture module, internet and Website server, speech-to-text server, face Emotion identification server;Method is that client issues dialogic voice;Voice collecting playing module acquisition input customer voice and facial image are sent to associated server, judge dialog logic parallel according to returned content by it;The true man's dialogic voice or intermediate language real voice or conclusion sound bite for transferring respective logic, realize Effective Dialogue or transition to client playing from voice collecting playing module or terminate dialogue.Artificial intelligence technology of the present invention can be realized advertisement screen based on artificial intelligence technology to exchange with personalizing for true man, and targetedly realize product marketing according to customer anger and dialogue.

Description

A kind of intelligence true man's advertisement screen interactive system and implementation method
Technical field
The present invention relates to artificial intelligence technology and advertisement media field more particularly to a kind of intelligent true man's advertisement screen interaction systems System and implementation method.
Background technique
The rapid development of artificial intelligence of new generation will change human society life deeply, change the world.Really to realize people The intelligence of work intelligence personalizes, and realizes that artificial intelligence is exchanged with personalizing for true man, in addition to enabling artificial intelligence system to divide It distinguishes study real voice, personalize outside speech exchange with true man, it is necessary to enable artificial intelligence to analyze and judge true man Mood, true man occur glad, indignation, it is sad, detest, the frightened and moods such as surprised when, could accordingly make correctly Personalize response.This mood analysis and judgement are also required to that face recognition technology is relied on to realize to a great extent, first people After work intelligently needs to capture, acquire and recognize real's expressive features, could make on this basis further analysis with Judgement, to obtain real's emotion judgment result.Based on these technologies, we are try to for artificial intelligence technology being applied to Advertisement media field.It can be realized the advertisement screen under new artificial intelligence advertisement technology to exchange with personalizing for true man.But Also lacking one kind at present can intelligent true man's advertisement screen of practical application and the system and implementation method of human-computer interaction.
Summary of the invention
The present invention provides a kind of intelligent true man's advertisement screen interactive system and implementation methods, can make quasi- true with artificial intelligence The advertisement screen of specific true man realizes that the quasi- true dialogue with client realizes that mood interactive perception and voice dialogue exchange.The present invention provides A kind of intelligent true man's advertisement screen interactive system and seven kinds of intelligent true man's advertisement screens interaction implementation methods, skill of the present invention Art scheme is respectively:
A kind of intelligence true man's advertisement screen interactive system, comprising: intelligent true man's server 1, advertisement video playing module 2, language Sound acquires playing module 3, image capture module 4, internet and Website server 5, speech-to-text server 6, face mood Identify server 7;The advertisement video playing module 2, voice collecting playing module 3 and image capture module 4 are electrically connected intelligence True man's server 1;Intelligent true man's server 1 is by internet and the network connection of Website server 5 in speech-to-text server 6, face Emotion identification server 7;
The voice collecting playing module 3 acquires customer voice and send to intelligent true man's server 1;Intelligent true man's server 1 By customer voice through internet and Website server 5 send to speech-to-text server 6 by its convert speech into text after obtain Keyword or keyword are taken, then passes back to intelligent true man's server 1 through internet and Website server 5;
Described image acquisition module 4 acquires client's Facial Expression Image and send to intelligent true man's server 1;Intelligent true man's service Device 1 send client's Facial Expression Image to face Emotion identification server 7 by it to face table through internet and Website server 5 End of love is analyzed to obtain face mood referring back to value, then through internet and Website server 5 and pass back to intelligent true man Server 1;
The intelligence true man server 1 is transferred from storage unit corresponding according to keyword or keyword and referring back to value The voice and video clip of the specific true man of logic passes through advertisement video playing module 2 and voice collecting playing module 3 to client Play, to be reference with customer anger and logic of language, realize human customer and the specific true man's dummy model of artificial intelligence it Between intend true exchanges and dialogues.
Further, a kind of intelligent true man's advertisement screen interactive system, further includes: limb action identification server 8, Voice mood identifies server 9;The intelligence true man server 1 is by internet and the network connection of Website server 5 in limbs Action recognition server 8, voice mood identification server 9 are by internet and the network connection of Website server 5 in intelligent true man Server 1;
Described image acquisition module 4 acquires limb action image and send to intelligent true man's server 1;Intelligent true man's server 1 Limb action image is sent through internet and Website server 5 to limb action and identifies that server 8 carries out limb action by it Analysis passes back to intelligent true man's server 1 to obtain limbs mood referring back to value, then through internet and Website server 5;
The voice collecting playing module 3 acquires customer voice and send to intelligent true man's server 1;Intelligent true man's server 1 Customer voice is sent through internet and Website server 5 to voice mood identification server 9, customer voice is analyzed by it To obtain voice mood referring back to value, then through internet and Website server 5 pass back to intelligent true man's server 1;
The intelligence true man server 1 transfers phase referring back to value according to limbs mood or voice mood from storage unit The voice of the specific true man of logic and video clip is answered to pass through advertisement video playing module 2 and voice collecting playing module 3 to visitor Family plays, to be further reference with customer anger, realizes between human customer and the specific true man's dummy model of artificial intelligence Intend true exchanges and dialogues.
Further, a kind of intelligent true man's advertisement screen interactive system, advertisement video playing module 2 include: that specific true man are empty Analog model viewing area 21, advertisement background and product viewing area 22, client viewing area 23;Specific true man's dummy model viewing area 21 displays play the iconic model image of specific true man or the video clip of specific true man;Advertisement background and product viewing area 22 For showing advertisement background or product;Client viewing area 23 is for showing or playing the client that image capture module 4 collects Facial image or half body image or whole body images.
The first intelligent true man's advertisement screen interaction implementation method, includes the following steps:
S1, client issue dialogic voice;
S2, the acquisition input customer voice of voice collecting playing module 3;
S3, intelligent true man's server 1 send customer voice to speech-to-text server 6, and speech-to-text server 6 will Customer voice Content Transformation is text, and is sent to intelligent true man's server 1;
S4, intelligent true man's server 1 are judged following literal logic by conversation content parallel:
S401, intelligent true man's server 1 judge whether dialogue can identify or continue;If it is judged that dialogue can not It identifies or continues, then follow the steps S7;If it is judged that dialogue can identify and continue, S5 is thened follow the steps;
Whether S402, intelligent true man's server 1 judge dialogue beyond setting time;If it is judged that when dialogue is beyond setting Between, then follow the steps S7;If it is judged that dialogue thens follow the steps S5 without departing from setting time;
S403, intelligent true man's server 1 judge in text with the presence or absence of keyword or keyword;If it is judged that in text There is no keyword or keywords, then follow the steps S7;If it is judged that then executing step there are keyword or keyword in text Rapid S5;
S5, intelligent true man's server 1 transfer true man's language of respective logic according to keyword or keyword from storage unit Tablet section;
Selected real voice segment is sent to voice collecting playing module 3 and played by S6, intelligent true man's server 1;
S7, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid S9;If it is determined that needing not continue to talk with, S8 is thened follow the steps;
S8, intelligent true man's server 1 control voice collecting playing module 3 and play logical sessions conclusion voice sheet to user Section is to terminate to talk with;Step S10 is executed in next step;
S9, intelligent true man's server 1 transfer intermediate language real voice segment from storage unit and play mould by voice collecting Block 3 is talked with to client playing with transition;Step S2 is executed in next step;
S10, terminate service.
Second of intelligent true man's advertisement screen interaction implementation method, includes the following steps:
S0, intelligent true man's server 1 are triggered through voice collecting playing module 3 to client playing opening remarks to open pair Words;Step S1 is executed in next step.
S1, client issue dialogic voice;
S2, the acquisition input customer voice of voice collecting playing module 3;
S3, intelligent true man's server 1 send customer voice to speech-to-text server 6, and speech-to-text server 6 will Customer voice Content Transformation is text, and is sent to intelligent true man's server 1;
S4, intelligent true man's server 1 are judged following literal logic by conversation content parallel:
S401, intelligent true man's server 1 judge whether dialogue can identify or continue;If it is judged that dialogue can not It identifies or continues, then follow the steps S7;If it is judged that dialogue can identify and continue, S5 is thened follow the steps;
Whether S402, intelligent true man's server 1 judge dialogue beyond setting time;If it is judged that when dialogue is beyond setting Between, then follow the steps S7;If it is judged that dialogue thens follow the steps S5 without departing from setting time;
S403, intelligent true man's server 1 judge in text with the presence or absence of keyword or keyword;If it is judged that in text There is no keyword or keywords, then follow the steps S7;If it is judged that then executing step there are keyword or keyword in text Rapid S5;
S5, intelligent true man's server 1 transfer true man's language of respective logic according to keyword or keyword from storage unit Tablet section;
Selected real voice segment is sent to voice collecting playing module 3 and played by S6, intelligent true man's server 1;
S7, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid S9;If it is determined that needing not continue to talk with, S8 is thened follow the steps;
S8, intelligent true man's server 1 control voice collecting playing module 3 and play logical sessions conclusion voice sheet to user Section is to terminate to talk with;Step S10 is executed in next step;
S9, intelligent true man's server 1 transfer intermediate language real voice segment from storage unit and play mould by voice collecting Block 3 is talked with to client playing with transition;Step S2 is executed in next step;
S10, terminate service.
The third intelligent true man's advertisement screen interaction implementation method, includes the following steps:
SA1, intelligent true man's server 1 are triggered, and intelligent true man's server 1 controls image capture module 4 and starts to acquire;
SA2, the acquisition input client's facial image of image capture module 4;
Client's facial image is sent to face Emotion identification server 7 by SA3, intelligent true man's server 1, and face mood is known Other server 7 carries out expression shape change identification to client's facial image, obtains face mood by it and send to intelligence very referring back to value People's server 1;
SA4, intelligent true man's server 1 judge whether dialogue can continue to carry out referring back to value by face mood;If Judgement dialogue cannot continue, and then follow the steps SA7;If it is determined that dialogue can continue to carry out, SA5 is thened follow the steps;
SA5, intelligent true man's server 1 transfer respective logic referring back to value according to different faces mood from storage unit Voice and video clip;
SA6, intelligent true man's server 1 by selected voice and video clip be sent to voice collecting playing module 3 and Advertisement video playing module 2 is from it to client playing;
SA7, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid SA2;If it is determined that needing not continue to talk with, SA8 is thened follow the steps;
SA8, intelligent true man's server 1 control voice collecting playing module 3 and advertisement video playing module 2 to client playing Voice and video end fragment are to terminate to talk with;
SA9, terminate service.
4th kind of intelligent true man's advertisement screen interaction implementation method, includes the following steps:
SB1, intelligent true man's server 1 image capture module 4 that is triggered start to acquire;
SB2, the acquisition input client's limb action of image capture module 4;
Client's limb action image is sent to the identification service of action recognition server action by SB3, intelligent true man's server 1 Device carries out action recognition to client's limb action image and is sent referring back to value to intelligent true man's server by its acquisition limb action 1;
SB4, intelligent true man's server 1 judge whether dialogue can continue to carry out referring back to value by limb action;If Judgement dialogue cannot continue, and then follow the steps SB7;If it is determined that dialogue can continue to carry out, SB5 is thened follow the steps;
SB5, intelligent true man's server 1 transfer respective logic referring back to value according to different limb actions from storage unit Voice and video clip;
SB6, intelligent true man's server 1 by selected voice and video clip be sent to voice collecting playing module 3 and Advertisement video playing module 2 is from it to client playing;
SB7, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid SB2;If it is determined that needing not continue to talk with, SB8 is thened follow the steps;
SB8, intelligent true man's server 1 control voice collecting playing module 3 and advertisement video playing module 2 to client playing Voice and video end fragment are to terminate to talk with;
SB9, terminate service.
5th kind of intelligent true man's advertisement screen interaction implementation method, includes the following steps:
SC1, intelligent true man's server 1 are triggered through voice collecting playing module 3 to client playing opening remarks to open Dialogue;
SC2, client issue dialogic voice;
SC3, the acquisition input customer voice of voice collecting playing module 3;
Customer voice is sent to voice mood identification server voice mood identification service by SC4, intelligent true man's server 1 Customer voice is carried out ripple identification and obtains voice mood referring back to being worth and send to intelligent true man's server 1 by it by device 9;
SC5, intelligent true man's server 1 judge whether dialogue can continue referring back to value by voice mood;If Judgement dialogue cannot continue, and then follow the steps SC8;If it is determined that dialogue can continue to carry out, SC6 is thened follow the steps;
SC6, intelligent true man's server 1 transfer respective logic referring back to value according to different phonetic mood from storage unit Voice and video clip;
SC7, intelligent true man's server 1 by selected voice and video clip be sent to voice collecting playing module 3 and Advertisement video playing module 2 is from it to client playing;
SC8, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid SC10;If it is determined that needing not continue to talk with, SC9 is thened follow the steps;
SC9, intelligent true man's server 1 control voice collecting playing module 3 and play logical sessions conclusion voice to user Segment is to terminate to talk with;Step SC11 is executed in next step;
SC10, intelligent true man's server 1 are transferred intermediate language real voice segment from storage unit and are played by voice collecting Module 3 is talked with to client playing with transition;Step SC3 is executed in next step;
SC11, terminate service.
6th kind of intelligent true man's advertisement screen interaction implementation method, includes the following steps:
SS1, intelligent true man's server 1 image capture module 4 that is triggered start to acquire true man's image and are broadcast by voice collecting Amplification module 3 opens dialogue to client playing opening remarks;
SS2, client issue dialogic voice;
SS3, the acquisition of voice collecting playing module 3 input customer voice and/or the acquisition input client's language of image capture module 4 Sound and/or client's facial image;
SS4, intelligent true man's server 1 by customer voice and/or facial image, be sent to speech-to-text server 6 and/ Or face Emotion identification server 7, each server respectively identify customer voice content and/or human face expression variation, and Keyword or keyword and/or face mood is obtained respectively by it to send referring back to value to intelligent true man's server 1;
SS5, intelligent true man's server 1 carry out parallel logic judgement to the following contents:
SS501, intelligent true man's server 1 judge whether dialogue can identify or continue;If it is judged that dialogue nothing Method identifies or continues, and thens follow the steps SS8;If it is judged that dialogue can identify and continue, then follow the steps SS6;
Whether SS502, intelligent true man's server 1 judge dialogue beyond setting time;If it is judged that dialogue is beyond setting Time thens follow the steps SS8;If it is judged that dialogue thens follow the steps SS6 without departing from setting time;
SS503, intelligent true man's server 1 judge in text with the presence or absence of keyword or keyword;If it is judged that text In be not present keyword or keyword, then follow the steps SS8;If it is judged that then being held in text there are keyword or keyword Row step SS6;
SS504, intelligent true man's server 1 judge whether dialogue can continue referring back to value by face mood;Such as Fruit judges that dialogue can not identify or continue, and thens follow the steps SS8;If it is judged that dialogue can identify and continue into Row, thens follow the steps SS6;
SS6, intelligent true man's server 1 are transferred from storage unit according to different keywords or keyword and referring back to value The voice and video clip of respective logic;
SS7, intelligent true man's server 1 by selected voice and video clip be sent to voice collecting playing module 3 and Advertisement video playing module 2 is from it to client playing;
SS8, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid SS10;If it is determined that needing not continue to talk with, SS9 is thened follow the steps;
SS9, intelligent true man's server 1 control voice collecting playing module 3 and advertisement video playing module 2 to client playing Voice and video end fragment are to terminate to talk with;Step SS11 is executed in next step;
SS10, intelligent true man's server 1 are transferred intermediate language real voice/expression segment from storage unit and are adopted by voice Collection playing module 3 and advertisement video playing module 2 are talked with to client playing with transition;Step SS3 is executed in next step;
SS11, terminate service.
7th kind of intelligent true man's advertisement screen interaction implementation method, includes the following steps:
ST1, intelligent true man's server 1 image capture module 4 that is triggered start to acquire true man's image and are broadcast by voice collecting Amplification module 3 opens dialogue to client playing opening remarks;
ST2, client issue dialogic voice;
ST3, the acquisition of voice collecting playing module 3 input customer voice and/or the acquisition input client's language of image capture module 4 Sound and/or client's facial image and/or body movement;
Customer voice and/or facial image and/or body movement are sent to voice and turned by ST4, intelligent true man's server 1 Text service device 6 and/or face Emotion identification server 7 and/or limb action identification server 8 and/or voice mood identification Server 9, each server is respectively to customer voice content and/or human face expression variation and/or voice ripple and/or limb action It is identified, and obtains keyword or keyword and/or face mood and/or voice mood and/or limbs mood respectively by it It send referring back to value to intelligent true man's server 1;
ST5, intelligent true man's server 1 carry out parallel logic judgement to the following contents:
ST501, intelligent true man's server 1 judge whether dialogue can identify or continue;If it is judged that dialogue nothing Method identifies or continues, and thens follow the steps ST8;If it is judged that dialogue can identify and continue, then follow the steps ST6;
Whether ST502, intelligent true man's server 1 judge dialogue beyond setting time;If it is judged that dialogue is beyond setting Time thens follow the steps ST8;If it is judged that dialogue thens follow the steps ST6 without departing from setting time;
ST503, intelligent true man's server 1 judge in text with the presence or absence of keyword or keyword;If it is judged that text In be not present keyword or keyword, then follow the steps ST8;If it is judged that then being held in text there are keyword or keyword Row step ST6;
ST504, intelligent true man's server 1 judge whether dialogue can continue referring back to value by face mood;Such as Fruit judges that dialogue can not identify or continue, and thens follow the steps ST8;If it is judged that dialogue can identify and continue into Row, thens follow the steps ST6;
ST505, intelligent true man's server 1 judge whether dialogue can continue referring back to value by limb action;Such as Fruit judges that dialogue beyond setting time, thens follow the steps ST8;If it is judged that dialogue is then executed without departing from setting time Step ST6;
ST506, intelligent true man's server 1 judge whether dialogue can continue referring back to value by voice mood;Such as Fruit judges that there is no keyword or keywords in text, then follow the steps ST8;If it is judged that in text there are keyword or Keyword thens follow the steps ST6;
ST6, intelligent true man's server 1 are transferred from storage unit according to different keywords or keyword and referring back to value The voice and video clip of respective logic;
ST7, intelligent true man's server 1 by selected voice and video clip be sent to voice collecting playing module 3 and Advertisement video playing module 2 is from it to client playing;
ST8, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid ST10;If it is determined that needing not continue to talk with, ST9 is thened follow the steps;
ST9, intelligent true man's server 1 control voice collecting playing module 3 and advertisement video playing module 2 to client playing Voice and video end fragment are to terminate to talk with;Step ST11 is executed in next step;
ST10, intelligent true man's server 1 transferred from storage unit intermediate language real voice/expression/movement segment and by Voice collecting playing module 3 and advertisement video playing module 2 are talked with to client playing with transition;Step ST3 is executed in next step;
ST11, terminate service.
Compared with prior art, the beneficial effects of the present invention are:
1, the present invention provides a kind of intelligent true man's advertisement screen interactive system, by the way that artificial intelligence technology is used for advertisement screen, Advertisement screen is set to pass through the voice content of acquisition client, face's mood, voice mood, the client's feelings of limbs mood acquisition instantly Thread wish, so as to make targeted product distribution to client;
2, the present invention provides a kind of intelligent true man's advertisement screen interactive system, and the dummy model image of specific true man is showed in On advertisement screen, voice, movement, expression are carried out by for example certain star image models of the dummy model image of specific true man and client Exchange, it is more standby to intend true effect;
3, the present invention provides seven kinds of intelligent true man's advertisement screen interaction implementation methods, and advertisement screen is enable to pass through acquisition client's Voice content, face's mood, voice mood, limbs mood also make specific true man virtual to obtain customer anger wish instantly Specific true man's model in model viewing area can be done according to the voice content of client, face's mood, voice mood, limbs mood Out correspondingly, logical voice or expression or movement are to respond;
4, the intelligent true man's advertisement screen interaction implementation method of the first of the invention, only acquires the voice content of client, specifically Be suitable for client actively make a sound unlatching dialogue the case where;Under this method, advertisement screen can only show static star image, Only by voice and customer communication, this kind of scheme cost of implementation is low, and technical difficulty is small;
5, of the invention second intelligent true man's advertisement screen interaction implementation method, also only acquires the voice content of client, Based on first method, specific this system is suitable for client and reaches acquisition zone, and this system is triggered after activation, actively to client Issue the case where voice opening remarks are to open dialogue;This kind of scheme cost of implementation is low, and technical difficulty is small;
6, the intelligent true man's advertisement screen interaction implementation method of the third of the invention, only acquires the human face expression of client, specifically This system be suitable for customer and reach acquisition zone, this system is triggered after activation, and the human face expression of automatic collection client becomes Change to obtain expression mood;This kind of scheme cost of implementation is compared with technical difficulty is smaller;
7, of the invention the 4th kind intelligent true man's advertisement screen interaction implementation method acquires the limb action of client only to obtain Limb action mood, specific this system are suitable for customer and reach acquisition zone, and this system is triggered after activation, automatic collection The limb action of client changes to obtain limbs mood;This kind of scheme cost of implementation is smaller, and technical difficulty is smaller;
8,5th kind of the invention intelligent true man's advertisement screen interaction implementation method, to acquire simultaneously client voice content and Voice mood, specifically when intelligent true man exchange with customer voice, voice mood plays the role of mood reference;This kind of scheme Cost of implementation is low, and technical difficulty is small;
9,6th kind of the invention intelligent true man's advertisement screen interaction implementation method, to acquire simultaneously client voice content and Face mood, specifically when intelligent true man exchange with customer voice, face mood plays the role of mood reference;This kind of scheme Practicability and technical difficulty are moderate;
10,6th kind of the invention intelligent true man's advertisement screen interaction implementation method, to acquire simultaneously client voice content, Face mood, limb action, voice mood, specifically when intelligent true man exchange with customer voice, face mood, limbs are dynamic Make, voice mood all plays the role of mood reference;This kind of solution technique is complex;But in intelligent true man and customer communication When, most have and intends true effect.
Detailed description of the invention
Fig. 1 is a kind of intelligent true man's advertisement screen interactive system embodiment 1 of the present invention;
Fig. 2 is a kind of intelligent true man's advertisement screen interactive system embodiment 2 of the present invention;
Fig. 3 is a kind of implementation method steps flow chart Fig. 1 of intelligent true man's advertisement screen interactive system of the present invention;
Fig. 4 is a kind of implementation method steps flow chart Fig. 2 of intelligent true man's advertisement screen interactive system of the present invention;
Fig. 5 is a kind of implementation method steps flow chart Fig. 3 of intelligent true man's advertisement screen interactive system of the present invention;
Fig. 6 is a kind of implementation method steps flow chart Fig. 4 of intelligent true man's advertisement screen interactive system of the present invention;
Fig. 7 is a kind of implementation method steps flow chart Fig. 5 of intelligent true man's advertisement screen interactive system of the present invention;
Fig. 8 is a kind of implementation method steps flow chart Fig. 6 of intelligent true man's advertisement screen interactive system of the present invention;
Fig. 9 is a kind of implementation method steps flow chart Fig. 7 of intelligent true man's advertisement screen interactive system of the present invention.
Label declaration
Intelligent true man's server 1;Advertisement video playing module 2;Specific true man's dummy model viewing area 21;Advertisement background and Product viewing area 22;Client viewing area 23;Integral type image capture module 24;Voice collecting playing module 3;Microphone array 30;Image capture module 4;Internet and Website server 5;Speech-to-text server 6;Face Emotion identification server 7;Limb Body action recognition server 8;Voice mood identifies server 9;Client 100;Sound and video acquisition area 200.
Specific embodiment
The following further describes the present invention with reference to the drawings.
As shown in FIG. 1, FIG. 1 is a kind of intelligent true man's advertisement screen interactive system embodiments 1 of the present invention.Comprising: which intelligence is true People's server 1, advertisement video playing module 2, voice collecting playing module 3, image capture module 4, internet and website service Device 5, speech-to-text server 6, face Emotion identification server 7;The advertisement video playing module 2, voice collecting play Module 3 and image capture module 4 are electrically connected intelligent true man's server 1;Intelligent true man's server 1 passes through internet and website service Device 5 is connected to the network in speech-to-text server 6, face Emotion identification server 7;
The voice collecting playing module 3 acquires customer voice and send to intelligent true man's server 1;Intelligent true man's server 1 By customer voice through internet and Website server 5 send to speech-to-text server 6 by its convert speech into text after obtain Keyword or keyword are taken, then passes back to intelligent true man's server 1 through internet and Website server 5;
Described image acquisition module 4 acquires client's Facial Expression Image and send to intelligent true man's server 1;Intelligent true man's service Device 1 send client's Facial Expression Image to face Emotion identification server 7 by it to face table through internet and Website server 5 End of love is analyzed to obtain face mood referring back to value, then through internet and Website server 5 and pass back to intelligent true man Server 1;
The intelligence true man server 1 is transferred from storage unit corresponding according to keyword or keyword and referring back to value The voice and video clip of the specific true man of logic passes through advertisement video playing module 2 and voice collecting playing module 3 to client 100 play, to be reference with customer anger and logic of language, realize human customer and the specific true man's dummy model of artificial intelligence Between intend true exchanges and dialogues.
Wherein, advertisement video playing module 2 includes: that specific true man's dummy model viewing area 21, advertisement background and product are aobvious Show area 22, client viewing area 23;The display of specific true man's dummy model viewing area 21 or the iconic model for playing specific true man The video clip of image or specific true man;Advertisement background and product viewing area 22 are for showing advertisement background or product;Client is aobvious Show area 23 for showing or playing the client's facial image or half body image or whole body images that image capture module 4 collects.
In addition, also illustrating client 100, sound and video acquisition area 200 in figure;When 100 station of client to sound and video 200 position of acquisition zone.Specifically, intelligent true man's server 1 is triggered, broadcast by voice collecting playing module 3 to client 100 Opening remarks are put to open dialogue;The input client's facial image of the acquisition of image capture module 4 at the same time or client's limb action are sent To intelligent true man's server 1;Then client 100 issues dialogic voice;The acquisition input customer voice of voice collecting playing module 3 is sent To intelligent true man's server 1.Intelligent true man's server 1 got after sending related data to different external servers keyword or Keyword and/or face mood and/or voice mood and/or limbs mood are referring back to value;Logic is completed according to these information Judgement and organizational logical framework send selected voice and video clip according to these information intelligents true man server 1 It is played to voice collecting playing module 3 and advertisement video playing module 2 from it to client 100;To realize human customer and people Work intelligently intends true exchanges and dialogues between specific true man's dummy model.
As shown in Fig. 2, Fig. 2 is a kind of intelligent true man's advertisement screen interactive system embodiment 2 of the present invention.Implement with shown in Fig. 1 Unlike example, one of Fig. 2 intelligence true man's advertisement screen interactive system, further includes: limb action identifies server 8, voice Emotion identification server 9;The intelligence true man server 1 is by internet and the network connection of Website server 5 in limb action Identify that server 8, voice mood identification server 9 are serviced by internet and the network connection of Website server 5 in intelligent true man Device 1.
Described image acquisition module 4 acquires limb action image and send to intelligent true man's server 1;Intelligent true man's server 1 Limb action image is sent through internet and Website server 5 to limb action and identifies that server 8 carries out limb action by it Analysis passes back to intelligent true man's server 1 to obtain limbs mood referring back to value, then through internet and Website server 5;
The voice collecting playing module 3 acquires customer voice and send to intelligent true man's server 1;Intelligent true man's server 1 Customer voice is sent through internet and Website server 5 to voice mood identification server 9, customer voice is analyzed by it To obtain voice mood referring back to value, then through internet and Website server 5 pass back to intelligent true man's server 1;
The intelligence true man server 1 transfers phase referring back to value according to limbs mood or voice mood from storage unit The voice of the specific true man of logic and video clip is answered to pass through advertisement video playing module 2 and voice collecting playing module 3 to visitor Family plays, to be further reference with customer anger, realizes between human customer and the specific true man's dummy model of artificial intelligence Intend true exchanges and dialogues.
In addition, Fig. 2 embodiment does not use independent image capture module 4, but use and advertisement video playing module 2 The integral type image capture module 24 being wholely set;And microphone array 30 is utilized specifically to be used as voice collecting and play Module 3, because microphone array 30 compares general voice collecting or playing device, microphone array 30 can detect sound source automatically Position tracks speaker, and spatially auditory localization is more accurate;And there is spatial processing a little, to multi signal sky time-frequency three The processing of dimension makes up mono signal in noise suppressed, and echo inhibits, and advantage is had more in terms of Reverberation Rejection.Secondly microphone array 30 can Intelligent true man's clothes are indirectly coupled to through it to be directly connected to intelligent true man's server 1 or be connected to advertisement video playing module 2 Business device 1.
As needed, the intelligent true man's server 1 or internet of a kind of intelligent true man's advertisement screen interactive system of the present invention and Website server 5, speech-to-text server 6, face Emotion identification server 7, limb action identify server 8, voice feelings Thread identification server 9 can replace with terminal computer according to the hardware configuration of computer.
As shown in figure 3, Fig. 3 is a kind of implementation method steps flow chart Fig. 1 of intelligent true man's advertisement screen interactive system of the present invention. Its steps flow chart is as follows:
S1, client 100 issue dialogic voice;
S2, the acquisition input customer voice of voice collecting playing module 3;
S3, intelligent true man's server 1 send customer voice to speech-to-text server 6, and speech-to-text server 6 will Customer voice Content Transformation is text, and is sent to intelligent true man's server 1;
S4, intelligent true man's server 1 are judged following literal logic by conversation content parallel:
S401, intelligent true man's server 1 judge whether dialogue can identify or continue;If it is judged that dialogue can not It identifies or continues, then follow the steps S7;If it is judged that dialogue can identify and continue, S5 is thened follow the steps;
Whether S402, intelligent true man's server 1 judge dialogue beyond setting time;If it is judged that when dialogue is beyond setting Between, then follow the steps S7;If it is judged that dialogue thens follow the steps S5 without departing from setting time;
S403, intelligent true man's server 1 judge in text with the presence or absence of keyword or keyword;If it is judged that in text There is no keyword or keywords, then follow the steps S7;If it is judged that then executing step there are keyword or keyword in text Rapid S5;
S5, intelligent true man's server 1 transfer true man's language of respective logic according to keyword or keyword from storage unit Tablet section;
Selected real voice segment is sent to voice collecting playing module 3 and played by S6, intelligent true man's server 1;
S7, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid S9;If it is determined that needing not continue to talk with, S8 is thened follow the steps;
S8, intelligent true man's server 1 control voice collecting playing module 3 and play logical sessions conclusion voice sheet to user Section is to terminate to talk with;Step S10 is executed in next step;
S9, intelligent true man's server 1 transfer intermediate language real voice segment from storage unit and play mould by voice collecting Block 3 is played to client 100 is talked with transition;Step S2 is executed in next step;
S10, terminate service.
As shown in figure 4, Fig. 4 is a kind of implementation method steps flow chart Fig. 2 of intelligent true man's advertisement screen interactive system of the present invention. Its steps flow chart is as follows:
S0, intelligent true man's server 1, which are triggered, plays opening remarks to client 100 by voice collecting playing module 3 to open Open dialogue;Step S1 is executed in next step.
S1, client 100 issue dialogic voice;
S2, the acquisition input customer voice of voice collecting playing module 3;
S3, intelligent true man's server 1 send customer voice to speech-to-text server 6, and speech-to-text server 6 will Customer voice Content Transformation is text, and is sent to intelligent true man's server 1;
S4, intelligent true man's server 1 are judged following literal logic by conversation content parallel:
S401, intelligent true man's server 1 judge whether dialogue can identify or continue;If it is judged that dialogue can not It identifies or continues, then follow the steps S7;If it is judged that dialogue can identify and continue, S5 is thened follow the steps;
Whether S402, intelligent true man's server 1 judge dialogue beyond setting time;If it is judged that when dialogue is beyond setting Between, then follow the steps S7;If it is judged that dialogue thens follow the steps S5 without departing from setting time;
S403, intelligent true man's server 1 judge in text with the presence or absence of keyword or keyword;If it is judged that in text There is no keyword or keywords, then follow the steps S7;If it is judged that then executing step there are keyword or keyword in text Rapid S5;
S5, intelligent true man's server 1 transfer true man's language of respective logic according to keyword or keyword from storage unit Tablet section;
Selected real voice segment is sent to voice collecting playing module 3 and played by S6, intelligent true man's server 1;
S7, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid S9;If it is determined that needing not continue to talk with, S8 is thened follow the steps;
S8, intelligent true man's server 1 control voice collecting playing module 3 and play logical sessions conclusion voice sheet to user Section is to terminate to talk with;Step S10 is executed in next step;
S9, intelligent true man's server 1 transfer intermediate language real voice segment from storage unit and play mould by voice collecting Block 3 is played to client 100 is talked with transition;Step S2 is executed in next step;
S10, terminate service.
As shown in figure 5, Fig. 5 is a kind of implementation method steps flow chart Fig. 3 of intelligent true man's advertisement screen interactive system of the present invention. Its steps flow chart is as follows:
SA1, intelligent true man's server 1 are triggered, and intelligent true man's server 1 controls image capture module 4 and starts to acquire;
SA2, the acquisition input client's facial image of image capture module 4;
Client's facial image is sent to face Emotion identification server 7 by SA3, intelligent true man's server 1, and face mood is known Other server 7 carries out expression shape change identification to client's facial image, obtains face mood by it and send to intelligence very referring back to value People's server 1;
SA4, intelligent true man's server 1 judge whether dialogue can continue to carry out referring back to value by face mood;If Judgement dialogue cannot continue, and then follow the steps SA7;If it is determined that dialogue can continue to carry out, SA5 is thened follow the steps;
SA5, intelligent true man's server 1 transfer respective logic referring back to value according to different faces mood from storage unit Voice and video clip;
SA6, intelligent true man's server 1 by selected voice and video clip be sent to voice collecting playing module 3 and Advertisement video playing module 2 is played from it to client 100;
SA7, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid SA2;If it is determined that needing not continue to talk with, SA8 is thened follow the steps;
SA8, intelligent true man's server 1 control voice collecting playing module 3 and advertisement video playing module 2 to client 100 Voice and video end fragment are played to terminate to talk with;
SA9, terminate service.
As shown in fig. 6, Fig. 6 is a kind of implementation method steps flow chart Fig. 4 of intelligent true man's advertisement screen interactive system of the present invention. Its steps flow chart is as follows:
SB1, intelligent true man's server 1 image capture module 4 that is triggered start to acquire;
SB2, the acquisition input client's limb action of image capture module 4;
Client's limb action image is sent to the identification service of action recognition server action by SB3, intelligent true man's server 1 Device carries out action recognition to client's limb action image and is sent referring back to value to intelligent true man's server by its acquisition limb action 1;
SB4, intelligent true man's server 1 judge whether dialogue can continue to carry out referring back to value by limb action;If Judgement dialogue cannot continue, and then follow the steps SB7;If it is determined that dialogue can continue to carry out, SB5 is thened follow the steps;
SB5, intelligent true man's server 1 transfer respective logic referring back to value according to different limb actions from storage unit Voice and video clip;
SB6, intelligent true man's server 1 by selected voice and video clip be sent to voice collecting playing module 3 and Advertisement video playing module 2 is played from it to client 100;
SB7, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid SB2;If it is determined that needing not continue to talk with, SB8 is thened follow the steps;
SB8, intelligent true man's server 1 control voice collecting playing module 3 and advertisement video playing module 2 to client 100 Voice and video end fragment are played to terminate to talk with;
SB9, terminate service.
As shown in fig. 7, Fig. 7 is a kind of implementation method steps flow chart Fig. 5 of intelligent true man's advertisement screen interactive system of the present invention. Its steps flow chart is as follows:
SC1, intelligent true man's server 1, which are triggered, plays opening remarks to client 100 by voice collecting playing module 3 to open Open dialogue;
SC2, client 100 issue dialogic voice;
SC3, the acquisition input customer voice of voice collecting playing module 3;
Customer voice is sent to voice mood identification server voice mood identification service by SC4, intelligent true man's server 1 Customer voice is carried out ripple identification and obtains voice mood referring back to being worth and send to intelligent true man's server 1 by it by device 9;
SC5, intelligent true man's server 1 judge whether dialogue can continue referring back to value by voice mood;If Judgement dialogue cannot continue, and then follow the steps SC8;If it is determined that dialogue can continue to carry out, SC6 is thened follow the steps;
SC6, intelligent true man's server 1 transfer respective logic referring back to value according to different phonetic mood from storage unit Voice and video clip;
SC7, intelligent true man's server 1 by selected voice and video clip be sent to voice collecting playing module 3 and Advertisement video playing module 2 is played from it to client 100;
SC8, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid SC10;If it is determined that needing not continue to talk with, SC9 is thened follow the steps;
SC9, intelligent true man's server 1 control voice collecting playing module 3 and play logical sessions conclusion voice to user Segment is to terminate to talk with;Step SC11 is executed in next step;
SC10, intelligent true man's server 1 are transferred intermediate language real voice segment from storage unit and are played by voice collecting Module 3 is played to client 100 is talked with transition;Step SC3 is executed in next step;
SC11, terminate service.
As shown in figure 8, Fig. 8 is a kind of implementation method steps flow chart Fig. 6 of intelligent true man's advertisement screen interactive system of the present invention. Its steps flow chart is as follows:
SS1, intelligent true man's server 1 image capture module 4 that is triggered start to acquire true man's image and are broadcast by voice collecting Amplification module 3 plays opening remarks to client 100 to open dialogue;
SS2, client 100 issue dialogic voice;
SS3, the acquisition of voice collecting playing module 3 input customer voice and/or the acquisition input client's language of image capture module 4 Sound and/or client's facial image;
SS4, intelligent true man's server 1 by customer voice and/or facial image, be sent to speech-to-text server 6 and/ Or face Emotion identification server 7, each server respectively identify customer voice content and/or human face expression variation, and Keyword or keyword and/or face mood is obtained respectively by it to send referring back to value to intelligent true man's server 1;
SS5, intelligent true man's server 1 carry out parallel logic judgement to the following contents:
SS501, intelligent true man's server 1 judge whether dialogue can identify or continue;If it is judged that dialogue nothing Method identifies or continues, and thens follow the steps SS8;If it is judged that dialogue can identify and continue, then follow the steps SS6;
Whether SS502, intelligent true man's server 1 judge dialogue beyond setting time;If it is judged that dialogue is beyond setting Time thens follow the steps SS8;If it is judged that dialogue thens follow the steps SS6 without departing from setting time;
SS503, intelligent true man's server 1 judge in text with the presence or absence of keyword or keyword;If it is judged that text In be not present keyword or keyword, then follow the steps SS8;If it is judged that then being held in text there are keyword or keyword Row step SS6;
SS504, intelligent true man's server 1 judge whether dialogue can continue referring back to value by face mood;Such as Fruit judges that dialogue can not identify or continue, and thens follow the steps SS8;If it is judged that dialogue can identify and continue into Row, thens follow the steps SS6;
SS6, intelligent true man's server 1 are transferred from storage unit according to different keywords or keyword and referring back to value The voice and video clip of respective logic;
SS7, intelligent true man's server 1 by selected voice and video clip be sent to voice collecting playing module 3 and Advertisement video playing module 2 is played from it to client 100;
SS8, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid SS10;If it is determined that needing not continue to talk with, SS9 is thened follow the steps;
SS9, intelligent true man's server 1 control voice collecting playing module 3 and advertisement video playing module 2 to client 100 Voice and video end fragment are played to terminate to talk with;Step SS11 is executed in next step;
SS10, intelligent true man's server 1 are transferred intermediate language real voice/expression segment from storage unit and are adopted by voice Collection playing module 3 and advertisement video playing module 2 are played to client 100 is talked with transition;Step SS3 is executed in next step;
SS11, terminate service.
As shown in figure 9, Fig. 9 is a kind of implementation method steps flow chart Fig. 7 of intelligent true man's advertisement screen interactive system of the present invention. Its steps flow chart is as follows:
ST1, intelligent true man's server 1 image capture module 4 that is triggered start to acquire true man's image and are broadcast by voice collecting Amplification module 3 plays opening remarks to client 100 to open dialogue;
ST2, client 100 issue dialogic voice;
ST3, the acquisition of voice collecting playing module 3 input customer voice and/or the acquisition input client's language of image capture module 4 Sound and/or client's facial image and/or body movement;
Customer voice and/or facial image and/or body movement are sent to voice and turned by ST4, intelligent true man's server 1 Text service device 6 and/or face Emotion identification server 7 and/or limb action identification server 8 and/or voice mood identification Server 9, each server is respectively to customer voice content and/or human face expression variation and/or voice ripple and/or limb action It is identified, and obtains keyword or keyword and/or face mood and/or voice mood and/or limbs mood respectively by it It send referring back to value to intelligent true man's server 1;
ST5, intelligent true man's server 1 carry out parallel logic judgement to the following contents:
ST501, intelligent true man's server 1 judge whether dialogue can identify or continue;If it is judged that dialogue nothing Method identifies or continues, and thens follow the steps ST8;If it is judged that dialogue can identify and continue, then follow the steps ST6;
Whether ST502, intelligent true man's server 1 judge dialogue beyond setting time;If it is judged that dialogue is beyond setting Time thens follow the steps ST8;If it is judged that dialogue thens follow the steps ST6 without departing from setting time;
ST503, intelligent true man's server 1 judge in text with the presence or absence of keyword or keyword;If it is judged that text In be not present keyword or keyword, then follow the steps ST8;If it is judged that then being held in text there are keyword or keyword Row step ST6;
ST504, intelligent true man's server 1 judge whether dialogue can continue referring back to value by face mood;Such as Fruit judges that dialogue can not identify or continue, and thens follow the steps ST8;If it is judged that dialogue can identify and continue into Row, thens follow the steps ST6;
ST505, intelligent true man's server 1 judge whether dialogue can continue referring back to value by limb action;Such as Fruit judges that dialogue beyond setting time, thens follow the steps ST8;If it is judged that dialogue is then executed without departing from setting time Step ST6;
ST506, intelligent true man's server 1 judge whether dialogue can continue referring back to value by voice mood;Such as Fruit judges that there is no keyword or keywords in text, then follow the steps ST8;If it is judged that in text there are keyword or Keyword thens follow the steps ST6;
ST6, intelligent true man's server 1 are transferred from storage unit according to different keywords or keyword and referring back to value The voice and video clip of respective logic;
ST7, intelligent true man's server 1 by selected voice and video clip be sent to voice collecting playing module 3 and Advertisement video playing module 2 is played from it to client 100;
ST8, intelligent true man's server 1 judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then step is executed Rapid ST10;If it is determined that needing not continue to talk with, ST9 is thened follow the steps;
ST9, intelligent true man's server 1 control voice collecting playing module 3 and advertisement video playing module 2 to client 100 Voice and video end fragment are played to terminate to talk with;Step ST11 is executed in next step;
ST10, intelligent true man's server 1 transferred from storage unit intermediate language real voice/expression/movement segment and by Voice collecting playing module 3 and advertisement video playing module 2 are played to client 100 is talked with transition;Step is executed in next step ST3;
ST11, terminate service.
Embodiment described above only describe the preferred embodiments of the invention, not to model of the invention It encloses and is defined, without departing from the spirit of the design of the present invention, those of ordinary skill in the art are to technical side of the invention The various changes and improvements that case is made should all be fallen into the protection scope that claims of the present invention determines.

Claims (10)

1. a kind of intelligence true man's advertisement screen interactive system characterized by comprising intelligent true man's server (1), advertisement video are broadcast Amplification module (2), voice collecting playing module (3), image capture module (4), internet and Website server (5), voice turn text Character server (6), face Emotion identification server (7);The advertisement video playing module (2), voice collecting playing module (3) Intelligent true man's server (1) is electrically connected with image capture module (4);Intelligent true man's server (1) is taken by internet and website Device (5) network connection be engaged in speech-to-text server (6), face Emotion identification server (7);
Voice collecting playing module (3) the acquisition customer voice is sent to intelligent true man's server (1);Intelligent true man's server (1) customer voice is sent through internet and Website server (5) to speech-to-text server (6) and text is converted speech by it Keyword or keyword are obtained after word, then pass back to intelligent true man's server (1) through internet and Website server (5);
Described image acquisition module (4) acquisition client's Facial Expression Image is sent to intelligent true man's server (1);Intelligent true man's service It is right by its to face Emotion identification server (7) that device (1) send client's Facial Expression Image through internet and Website server (5) Human face expression variation is analyzed to obtain face mood referring back to value, then is passed back to through internet and Website server (5) Intelligent true man's server (1);
Intelligence true man's server (1) is accordingly patrolled according to keyword or keyword and transferring from storage unit referring back to value The voice and video clip of the specific true man collected passes through advertisement video playing module (2) and voice collecting playing module (3) to visitor Family plays, to be reference with customer anger and logic of language, realizes human customer and the specific true man's dummy model of artificial intelligence Between intend true exchanges and dialogues.
2. a kind of intelligent true man's advertisement screen interactive system according to claim 1, which is characterized in that further include: limbs are dynamic Make identification server (8), voice mood identification server (9);Intelligence true man's server (1) is taken by internet and website Business device (5) network connection identifies that server (8), voice mood identification server (9) pass through internet and website in limb action Server (5) is connected to the network in intelligent true man's server (1);
Described image acquisition module (4) acquisition limb action image is sent to intelligent true man's server (1);Intelligent true man's server (1) limb action image is sent through internet and Website server (5) to limb action identification server (8) and limbs is moved by it It is analyzed to obtain limbs mood referring back to value, then passes back to intelligent true man through internet and Website server (5) and take It is engaged in device (1);
Voice collecting playing module (3) the acquisition customer voice is sent to intelligent true man's server (1);Intelligent true man's server (1) by customer voice through internet and Website server (5) send to voice mood identification server (9) by its to customer voice into Row analysis passes back to intelligent true man's server to obtain voice mood referring back to value, then through internet and Website server (5) (1);
Intelligence true man's server (1) is transferred from storage unit according to limbs mood or voice mood referring back to value accordingly The voice and video clip of the specific true man of logic by advertisement video playing module (2) and voice collecting playing module (3) to Client playing, to be further reference with customer anger, realize human customer and the specific true man's dummy model of artificial intelligence it Between intend true exchanges and dialogues.
3. a kind of intelligent true man's advertisement screen interactive system according to claim 1 or 2, which is characterized in that advertisement video is broadcast Amplification module (2) includes: specific true man's dummy model viewing area (21), advertisement background and product viewing area (22), client viewing area (23);Specific true man's dummy model viewing area (21) display or play specific true man iconic model image or specific true man Video clip;Advertisement background and product viewing area (22) are for showing advertisement background or product;Client viewing area (23) is used for Display plays client's facial image or half body image or whole body images that image capture module (4) collect.
4. a kind of intelligence true man's advertisement screen interaction implementation method, which comprises the steps of:
S1, client issue dialogic voice;
S2, voice collecting playing module (3) acquisition input customer voice;
S3, intelligent true man's server (1) send customer voice to speech-to-text server (6), speech-to-text server (6) It is text by customer voice Content Transformation, and send to intelligent true man's server (1);
S4, intelligent true man's server (1) are judged following literal logic by conversation content parallel:
S401, intelligent true man's server (1) judge whether dialogue can identify or continue;If it is judged that dialogue can not be known Not or continue, thens follow the steps S7;If it is judged that dialogue can identify and continue, S5 is thened follow the steps;
Whether S402, intelligent true man's server (1) judge dialogue beyond setting time;If it is judged that when dialogue is beyond setting Between, then follow the steps S7;If it is judged that dialogue thens follow the steps S5 without departing from setting time;
S403, intelligent true man's server (1) judge in text with the presence or absence of keyword or keyword;If it is judged that in text not There are keyword or keywords, then follow the steps S7;If it is judged that being thened follow the steps in text there are keyword or keyword S5;
S5, intelligent true man's server (1) transfer the real voice of respective logic according to keyword or keyword from storage unit Segment;
Selected real voice segment is sent to voice collecting playing module (3) and played by S6, intelligent true man's server (1);
S7, intelligent true man's server (1) judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then follow the steps S9;If it is determined that needing not continue to talk with, S8 is thened follow the steps;
S8, intelligent true man's server (1) control voice collecting playing module (3) and play logical sessions conclusion voice sheet to user Section is to terminate to talk with;Step S10 is executed in next step;
S9, intelligent true man's server (1) transfer intermediate language real voice segment by voice collecting playing module from storage unit (3) talked with to client playing with transition;Step S2 is executed in next step;
S10, terminate service.
5. a kind of intelligent true man's advertisement screen interaction implementation method according to claim 4, which is characterized in that further include initial Step:
S0, intelligent true man's server (1) are triggered through voice collecting playing module (3) to client playing opening remarks to open pair Words;Step S1 is executed in next step.
6. a kind of intelligence true man's advertisement screen interaction implementation method, which comprises the steps of:
SA1, intelligent true man's server (1) are triggered, and intelligent true man's server (1) control image capture module (4) starts to acquire;
SA2, image capture module (4) acquisition input client's facial image;
Client's facial image is sent to face Emotion identification server (7) by SA3, intelligent true man's server (1), and face mood is known Other server (7) carries out expression shape change identification to client's facial image, obtains face mood by it and send referring back to value to intelligence True man's server (1);
SA4, intelligent true man's server (1) judge whether dialogue can continue to carry out referring back to value by face mood;If sentenced Disconnected dialogue cannot continue, and then follow the steps SA7;If it is determined that dialogue can continue to carry out, then follow the steps
SA5;
SA5, intelligent true man's server (1) transfer respective logic referring back to value according to different faces mood from storage unit Voice and video clip;
SA6, intelligent true man's server (1) by selected voice and video clip be sent to voice collecting playing module (3) and Advertisement video playing module (2) You Qixiang client playing;
SA7, intelligent true man's server (1) judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then follow the steps SA2;If it is determined that needing not continue to talk with, SA8 is thened follow the steps;
SA8, intelligent true man's server (1) controls voice collecting playing module (3) and advertisement video playing module (2) is broadcast to client Voice and video end fragment are put to terminate to talk with;
SA9, terminate service.
7. a kind of intelligence true man's advertisement screen interaction implementation method, which comprises the steps of:
SB1, intelligent true man's server (1) image capture module (4) that is triggered start to acquire;
SB2, image capture module (4) acquisition input client's limb action;
Client's limb action image is sent to action recognition server action identification server by SB3, intelligent true man's server (1) Action recognition is carried out to client's limb action image to be sent referring back to value to intelligent true man's server by its acquisition limb action (1);
SB4, intelligent true man's server (1) judge whether dialogue can continue to carry out referring back to value by limb action;If sentenced Disconnected dialogue cannot continue, and then follow the steps SB7;If it is determined that dialogue can continue to carry out, then follow the steps
SB5;
SB5, intelligent true man's server (1) transfer respective logic referring back to value according to different limb actions from storage unit Voice and video clip;
SB6, intelligent true man's server (1) by selected voice and video clip be sent to voice collecting playing module (3) and Advertisement video playing module (2) You Qixiang client playing;
SB7, intelligent true man's server (1) judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then follow the steps SB2;If it is determined that needing not continue to talk with, SB8 is thened follow the steps;
SB8, intelligent true man's server (1) controls voice collecting playing module (3) and advertisement video playing module (2) is broadcast to client Voice and video end fragment are put to terminate to talk with;
SB9, terminate service.
8. a kind of intelligence true man's advertisement screen interaction implementation method, which comprises the steps of:
SC1, intelligent true man's server (1) are triggered through voice collecting playing module (3) to client playing opening remarks to open Dialogue;
SC2, client issue dialogic voice;
SC3, voice collecting playing module (3) acquisition input customer voice;
Customer voice is sent to voice mood identification server voice mood identification server by SC4, intelligent true man's server (1) (9) customer voice is subjected to ripple identification and voice mood is obtained referring back to being worth and send to intelligent true man's server (1) by it;
SC5, intelligent true man's server (1) judge whether dialogue can continue referring back to value by voice mood;If sentenced Disconnected dialogue cannot continue, and then follow the steps SC8;If it is determined that dialogue can continue to carry out, SC6 is thened follow the steps;
SC6, intelligent true man's server (1) transfer respective logic referring back to value according to different phonetic mood from storage unit Voice and video clip;
SC7, intelligent true man's server (1) by selected voice and video clip be sent to voice collecting playing module (3) and Advertisement video playing module (2) You Qixiang client playing;
SC8, intelligent true man's server (1) judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then follow the steps SC10;If it is determined that needing not continue to talk with, SC9 is thened follow the steps;
SC9, intelligent true man's server (1) control voice collecting playing module (3) and play logical sessions conclusion voice to user Segment is to terminate to talk with;Step SC11 is executed in next step;
SC10, intelligent true man's server (1) transfer intermediate language real voice segment from storage unit and play mould by voice collecting Block (3) is talked with to client playing with transition;Step SC3 is executed in next step;
SC11, terminate service.
9. a kind of intelligence true man's advertisement screen interaction implementation method, which comprises the steps of:
SS1, intelligent true man's server (1) image capture module (4) that is triggered start to acquire true man's image and are broadcast by voice collecting Amplification module (3) opens dialogue to client playing opening remarks;
SS2, client issue dialogic voice;
SS3, voice collecting playing module (3) acquisition input customer voice and/or image capture module (4) acquisition input client's language Sound and/or client's facial image;
SS4, intelligent true man's server (1) by customer voice and/or facial image, be sent to speech-to-text server (6) and/ Or face Emotion identification server (7), each server respectively identify customer voice content and/or human face expression variation, And it obtains keyword or keyword and/or face mood respectively by it and send referring back to value to intelligent true man's server (1);
SS5, intelligent true man's server (1) carry out parallel logic judgement to the following contents:
SS501, intelligent true man's server (1) judge whether dialogue can identify or continue;If it is judged that dialogue can not It identifies or continues, then follow the steps SS8;If it is judged that dialogue can identify and continue, SS6 is thened follow the steps;
Whether SS502, intelligent true man's server (1) judge dialogue beyond setting time;If it is judged that when dialogue is beyond setting Between, then follow the steps SS8;If it is judged that dialogue thens follow the steps SS6 without departing from setting time;
SS503, intelligent true man's server (1) judge in text with the presence or absence of keyword or keyword;If it is judged that in text There is no keyword or keywords, then follow the steps SS8;If it is judged that then being executed in text there are keyword or keyword Step SS6;
SS504, intelligent true man's server (1) judge whether dialogue can continue referring back to value by face mood;If Judge that dialogue can not identify or continue, thens follow the steps SS8;If it is judged that dialogue can identify and continue, Then follow the steps SS6;
SS6, intelligent true man's server (1) transfer phase according to different keywords or keyword and referring back to value from storage unit Answer the voice and video clip of logic;
SS7, intelligent true man's server (1) by selected voice and video clip be sent to voice collecting playing module (3) and Advertisement video playing module (2) You Qixiang client playing;
SS8, intelligent true man's server (1) judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then follow the steps SS10;If it is determined that needing not continue to talk with, SS9 is thened follow the steps;
SS9, intelligent true man's server (1) controls voice collecting playing module (3) and advertisement video playing module (2) is broadcast to client Voice and video end fragment are put to terminate to talk with;Step SS11 is executed in next step;
SS10, intelligent true man's server (1) transfer intermediate language real voice/expression segment and by voice collecting from storage unit Playing module (3) and advertisement video playing module (2) are talked with to client playing with transition;Step SS3 is executed in next step;
SS11, terminate service.
10. a kind of intelligence true man's advertisement screen interaction implementation method, which comprises the steps of:
ST1, intelligent true man's server (1) image capture module (4) that is triggered start to acquire true man's image and are broadcast by voice collecting Amplification module (3) opens dialogue to client playing opening remarks;
ST2, client issue dialogic voice;
ST3, voice collecting playing module (3) acquisition input customer voice and/or image capture module (4) acquisition input client's language Sound and/or client's facial image and/or body movement;
Customer voice and/or facial image and/or body movement are sent to voice and turn text by ST4, intelligent true man's server (1) Character server (6) and/or face Emotion identification server (7) and/or limb action identification server (8) and/or voice mood It identifies server (9), each server is respectively to customer voice content and/or human face expression variation and/or voice ripple and/or limb Body movement is identified, and obtains keyword or keyword and/or face mood and/or voice mood and/or limb respectively by it Body mood is sent referring back to value to intelligent true man's server (1);
ST5, intelligent true man's server (1) carry out parallel logic judgement to the following contents:
ST501, intelligent true man's server (1) judge whether dialogue can identify or continue;If it is judged that dialogue can not It identifies or continues, then follow the steps ST8;If it is judged that dialogue can identify and continue, ST6 is thened follow the steps;
Whether ST502, intelligent true man's server (1) judge dialogue beyond setting time;If it is judged that when dialogue is beyond setting Between, then follow the steps ST8;If it is judged that dialogue thens follow the steps ST6 without departing from setting time;
ST503, intelligent true man's server (1) judge in text with the presence or absence of keyword or keyword;If it is judged that in text There is no keyword or keywords, then follow the steps ST8;If it is judged that then being executed in text there are keyword or keyword Step ST6;
ST504, intelligent true man's server (1) judge whether dialogue can continue referring back to value by face mood;If Judge that dialogue can not identify or continue, thens follow the steps ST8;If it is judged that dialogue can identify and continue, Then follow the steps ST6;
ST505, intelligent true man's server (1) judge whether dialogue can continue referring back to value by limb action;If Judge that dialogue beyond setting time, thens follow the steps ST8;If it is judged that dialogue then executes step without departing from setting time Rapid ST6;
ST506, intelligent true man's server (1) judge whether dialogue can continue referring back to value by voice mood;If Judge that there is no keyword or keywords in text, then follow the steps ST8;If it is judged that there are keyword or passes in text Key word, thens follow the steps ST6;
ST6, intelligent true man's server (1) transfer phase according to different keywords or keyword and referring back to value from storage unit Answer the voice and video clip of logic;
ST7, intelligent true man's server (1) by selected voice and video clip be sent to voice collecting playing module (3) and Advertisement video playing module (2) You Qixiang client playing;
ST8, intelligent true man's server (1) judge whether to need to continue to talk with;If it is determined that needing to continue to talk with, then follow the steps ST10;If it is determined that needing not continue to talk with, ST9 is thened follow the steps;
ST9, intelligent true man's server (1) controls voice collecting playing module (3) and advertisement video playing module (2) is broadcast to client Voice and video end fragment are put to terminate to talk with;Step ST11 is executed in next step;
ST10, intelligent true man's server (1) transfer intermediate language real voice/expression/movement segment and by language from storage unit Sound acquisition playing module (3) and advertisement video playing module (2) are talked with to client playing with transition;Step ST3 is executed in next step;
ST11, terminate service.
CN201711212325.6A 2017-11-28 2017-11-28 A kind of intelligence true man's advertisement screen interactive system and implementation method Pending CN109840009A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711212325.6A CN109840009A (en) 2017-11-28 2017-11-28 A kind of intelligence true man's advertisement screen interactive system and implementation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711212325.6A CN109840009A (en) 2017-11-28 2017-11-28 A kind of intelligence true man's advertisement screen interactive system and implementation method

Publications (1)

Publication Number Publication Date
CN109840009A true CN109840009A (en) 2019-06-04

Family

ID=66879409

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711212325.6A Pending CN109840009A (en) 2017-11-28 2017-11-28 A kind of intelligence true man's advertisement screen interactive system and implementation method

Country Status (1)

Country Link
CN (1) CN109840009A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111047358A (en) * 2019-12-10 2020-04-21 秒针信息技术有限公司 Member information query method and system based on face recognition

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110283190A1 (en) * 2010-05-13 2011-11-17 Alexander Poltorak Electronic personal interactive device
WO2013027893A1 (en) * 2011-08-22 2013-02-28 Kang Jun-Kyu Apparatus and method for emotional content services on telecommunication devices, apparatus and method for emotion recognition therefor, and apparatus and method for generating and matching the emotional content using same
CN104574005A (en) * 2015-02-15 2015-04-29 蔡耿新 Advertisement display management system integrating augmented reality, somatic sense and green screening technology and advertisement display management method integrating augmented reality, somatic sense and green screening technology
CN105681830A (en) * 2016-02-06 2016-06-15 优住(北京)科技股份公司 Advertisement playing system and playing platform based on interest points of user
CN206133569U (en) * 2016-08-19 2017-04-26 乔中力 Intelligence LED install alternately
CN106710590A (en) * 2017-02-24 2017-05-24 广州幻境科技有限公司 Voice interaction system with emotional function based on virtual reality environment and method
CN107278374A (en) * 2016-12-01 2017-10-20 深圳前海达闼云端智能科技有限公司 Interactive advertisement display method, terminal and smart city interactive system
CN107301168A (en) * 2017-06-01 2017-10-27 深圳市朗空亿科科技有限公司 Intelligent robot and its mood exchange method, system

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110283190A1 (en) * 2010-05-13 2011-11-17 Alexander Poltorak Electronic personal interactive device
WO2013027893A1 (en) * 2011-08-22 2013-02-28 Kang Jun-Kyu Apparatus and method for emotional content services on telecommunication devices, apparatus and method for emotion recognition therefor, and apparatus and method for generating and matching the emotional content using same
CN104574005A (en) * 2015-02-15 2015-04-29 蔡耿新 Advertisement display management system integrating augmented reality, somatic sense and green screening technology and advertisement display management method integrating augmented reality, somatic sense and green screening technology
CN105681830A (en) * 2016-02-06 2016-06-15 优住(北京)科技股份公司 Advertisement playing system and playing platform based on interest points of user
CN206133569U (en) * 2016-08-19 2017-04-26 乔中力 Intelligence LED install alternately
CN107278374A (en) * 2016-12-01 2017-10-20 深圳前海达闼云端智能科技有限公司 Interactive advertisement display method, terminal and smart city interactive system
CN106710590A (en) * 2017-02-24 2017-05-24 广州幻境科技有限公司 Voice interaction system with emotional function based on virtual reality environment and method
CN107301168A (en) * 2017-06-01 2017-10-27 深圳市朗空亿科科技有限公司 Intelligent robot and its mood exchange method, system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111047358A (en) * 2019-12-10 2020-04-21 秒针信息技术有限公司 Member information query method and system based on face recognition

Similar Documents

Publication Publication Date Title
CN108000526B (en) Dialogue interaction method and system for intelligent robot
CN112162628A (en) Multi-mode interaction method, device and system based on virtual role, storage medium and terminal
Bousmalis et al. Spotting agreement and disagreement: A survey of nonverbal audiovisual cues and tools
CN110413841A (en) Polymorphic exchange method, device, system, electronic equipment and storage medium
CN107300970A (en) Virtual reality exchange method and device
JP2017016566A (en) Information processing device, information processing method and program
Hasegawa et al. Active agent oriented multimodal interface system
CN109101663A (en) A kind of robot conversational system Internet-based
Kontogiorgos et al. A multimodal corpus for mutual gaze and joint attention in multiparty situated interaction
CN115167656A (en) Interactive service method and device based on artificial intelligence virtual image
Furui Speech recognition technology in the ubiquitous/wearable computing environment
CN111063346A (en) Cross-media star emotion accompany interaction system based on machine learning
CN109840009A (en) A kind of intelligence true man's advertisement screen interactive system and implementation method
Koshtura et al. Development of a Speech-to-Text Program for People with Haring Impairments.
Nakano et al. Implementation and evaluation of a multimodal addressee identification mechanism for multiparty conversation systems
JP7400364B2 (en) Speech recognition system and information processing method
CN101013571A (en) Interaction method and system for using voice order
Furui et al. Ubiquitous speech processing
Strauß et al. Wizard-of-Oz Data Collection for Perception and Interaction in Multi-User Environments.
CN209591484U (en) A kind of Multi-media interactive device based on intelligent true man modeling
Chiba et al. Estimation of user’s willingness to talk about the topic: Analysis of interviews between humans
CN113961680A (en) Human-computer interaction based session processing method and device, medium and electronic equipment
CN113852849A (en) Intelligent hotel room management method
CN209056016U (en) A kind of information kiosk and its system for human-computer interaction
CN113157241A (en) Interaction equipment, interaction device and interaction system

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
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20190604

WD01 Invention patent application deemed withdrawn after publication