CN106294854A - A kind of man-machine interaction method for intelligent robot and device - Google Patents
A kind of man-machine interaction method for intelligent robot and device Download PDFInfo
- Publication number
- CN106294854A CN106294854A CN201610703758.0A CN201610703758A CN106294854A CN 106294854 A CN106294854 A CN 106294854A CN 201610703758 A CN201610703758 A CN 201610703758A CN 106294854 A CN106294854 A CN 106294854A
- Authority
- CN
- China
- Prior art keywords
- topic
- information
- output
- input
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 80
- 230000003993 interaction Effects 0.000 title claims abstract description 48
- 230000002452 interceptive effect Effects 0.000 claims abstract description 64
- 230000007704 transition Effects 0.000 claims abstract description 22
- 230000008569 process Effects 0.000 claims abstract description 21
- 238000012546 transfer Methods 0.000 claims description 10
- 239000011159 matrix material Substances 0.000 claims description 8
- 230000007613 environmental effect Effects 0.000 claims description 7
- 230000002123 temporal effect Effects 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000003739 neck Anatomy 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3346—Query execution using probabilistic model
-
- 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
- G10L15/26—Speech to text systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/02—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Human Computer Interaction (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Acoustics & Sound (AREA)
- Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Probability & Statistics with Applications (AREA)
- Manipulator (AREA)
- Machine Translation (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
A kind of man-machine interaction method for intelligent robot and device, wherein, the method includes: dialogue interactive information obtaining step, obtains the dialogue interactive information of user's input;Input topic determines step, and dialogue interactive information is carried out semantic analysis, obtains the input topic corresponding to talking with interactive information;Output topic determines step, according to input topic, utilizes input topic and each candidate to export the topic transition probability between topic, exports from candidate and determine output topic topic.The method is when determining output topic, it is no longer to use fixing dialogue storehouse or language model to realize as existing man-machine interaction method, but determine output topic by the topic transition probability relevant with current interaction scenarios, so enable to that whole interactive process is more lively and hommization.
Description
Technical field
The present invention relates to robotics, specifically, relate to a kind of man-machine interaction side for intelligent robot
Method and device.
Background technology
Along with the development of science and technology, the introducing of information technology, computer technology and artificial intelligence technology, machine
Industrial circle is the most progressively walked out in the research of people, gradually extend to the necks such as medical treatment, health care, family, amusement and service occupation
Territory.And people for the requirement of robot also conform to the principle of simplicity the multiple mechanical action of substance be promoted to have anthropomorphic question and answer, autonomy and with
The intelligent robot that other robot interacts, man-machine interaction the most just becomes the key factor determining intelligent robot development.
Chat system, as the nucleus module in man-machine interactive system, plays very important in man-machine interactive system
Role.In daily interactive process, the chat between user and robot occupies most of ratio.Therefore,
The quality of chat system is directly connected to the Consumer's Experience of intelligent robot product, and then affects the going or staying of user, promotes intelligence
The language interaction capabilities of robot, is problem demanding prompt solution.
Summary of the invention
For solving the problems referred to above, the invention provides a kind of man-machine interaction method for intelligent robot, comprising:
Dialogue interactive information obtaining step, obtains the dialogue interactive information of user's input;
Input topic determines step, and described dialogue interactive information is carried out semantic analysis, obtains handing over corresponding to described dialogue
The input topic of mutual information;
Output topic determines step, according to described input topic, utilizes described input topic and each candidate to export topic
Between topic transition probability, export from candidate and topic determine output topic.
According to one embodiment of present invention, described method also includes:
Feedback information generation step, according to scene information and the described output topic of the current interaction process got, raw
Become feedback information.
According to one embodiment of present invention, in described feedback information generation step, judge that output topic is wrapped respectively
Whether each candidate's feedback information contained mates with described scene information, and determines required feedback letter according to matching result
Breath.
According to one embodiment of present invention, any one during described scene information includes item set forth below or several:
Temporal information, location information, environmental information and people information.
According to one embodiment of present invention, determine in step at described output topic, according to the ID of described input topic,
Utilize multidimensional topic transfer matrix to determine the ID of output topic, and determine output topic according to the ID of described output topic.
Present invention also offers a kind of human-computer interaction device for intelligent robot, comprising:
Dialogue interactive information acquisition module, for obtaining the dialogue interactive information of user's input;
Input topic determines module, for described dialogue interactive information is carried out semantic analysis, obtains corresponding to described right
The input topic of words interactive information;
Output topic determines module, for according to described input topic, utilizes described input topic to export with each candidate
Topic transition probability between topic, exports from candidate and determines output topic topic.
According to one embodiment of present invention, described device also includes:
Feedback information generation module, for the scene information according to the current interaction process got and described output words
Topic, generates feedback information.
According to one embodiment of present invention, described feedback information generation module is configured to judge that exporting topic is wrapped respectively
Whether each candidate's feedback information contained mates with described scene information, and determines required according to successful match matching result
Feedback information.
According to one embodiment of present invention, any one during described scene information includes item set forth below or several:
Temporal information, location information, environmental information and people information.
According to one embodiment of present invention, described output topic determines that module is configured to according to described input topic
ID, utilizes multidimensional topic transfer matrix to determine the ID of output topic, and determines output words according to the ID of described output topic
Topic.
The man-machine interaction method for intelligent robot that the present invention can be provided can come in conjunction with actual interaction scenarios
Determining final feedback information, compared to existing man-machine interaction method, the method enables to the feedback information of output more
Meet current interaction scenarios, thus user and experience in intelligent robot interaction are greatly improved.
Other features and advantages of the present invention will illustrate in the following description, and, partly become from description
Obtain it is clear that or understand by implementing the present invention.The purpose of the present invention and other advantages can be by description, rights
Structure specifically noted in claim and accompanying drawing realizes and obtains.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
In having technology to describe, required accompanying drawing does and simply introduces:
Fig. 1 is according to an embodiment of the invention for the flowchart of man-machine interaction method of intelligent robot;
Fig. 2 be in accordance with another embodiment of the present invention for intelligent robot man-machine interaction method realize flow process
Figure;
Fig. 3 is according to an embodiment of the invention for the structural representation of human-computer interaction device of intelligent robot.
Detailed description of the invention
Describe embodiments of the present invention in detail below with reference to drawings and Examples, whereby how the present invention is applied
Technological means solves technical problem, and the process that realizes reaching technique effect can fully understand and implement according to this.Need explanation
As long as not constituting conflict, each embodiment in the present invention and each feature in each embodiment can be combined with each other,
The technical scheme formed is all within protection scope of the present invention.
Meanwhile, in the following description, many details are elaborated for illustrative purposes, to provide real to the present invention
Execute the thorough understanding of example.It will be apparent, however, to one skilled in the art, that the present invention can tool here
Body details or described ad hoc fashion are implemented.
It addition, can be in the department of computer science of such as one group of computer executable instructions in the step shown in the flow chart of accompanying drawing
System performs, and, although show logical order in flow charts, but in some cases, can be to be different from herein
Order perform shown or described by step.
Existing intelligent robot is substantially based on dialogue storehouse or language model in fact with being used for the process of engaging in the dialogue
Existing, accuracy and the high-quality of the answer that intelligent robot fed back can be ensured in this way most possibly.
But, this conversational mode based on dialogue storehouse or language model equally exists many drawbacks, and one of them shows
Writing drawback is if the problem of user's input is identical, then identical answer feedback can be given every time and use by intelligent robot
Family, or from several limited answers, randomly select one feed back to user.And this has also resulted in existing for intelligence
The man-machine interaction method of robot can make intelligent robot with the answer of output in the interaction that engages in the dialogue more
Fixing, so that whole interaction seems the most stiff, this most just have impact on the Consumer's Experience of intelligent robot.
For the problems referred to above in the presence of prior art, the invention provides a kind of new people for intelligent robot
Machine interaction method.First the method determines the input topic corresponding to the interactive information that user is currently entered, subsequently by defeated
The topic transition probability entered between topic and output topic determines output topic in many topics of comforming.
In order to clearly illustrate the man-machine interaction method for intelligent robot provided by the present invention realize former
Manage, realize process and advantage, below in conjunction with different embodiments, man-machine interaction method provided by the present invention is made
Illustrate further.
Embodiment one:
Fig. 1 shows the flowchart of the man-machine interaction method for intelligent robot that the present embodiment provided.
As it is shown in figure 1, first the man-machine interaction method that the present embodiment is provided obtains user's input in step S101
Dialogue interactive information.In the present embodiment, the dialogue interactive information of user's input that the method is accessed in step S101 was both
It can be user's text envelope of passing through that the keyboard (including dummy keyboard and/or physical keyboard) provisioned in intelligent robot is inputted
Breath, it is also possible to be user's voice of passing through that the voice messaging collecting device (such as mike) provisioned in intelligent robot is inputted
Information, can also be the text message that obtains through natural language processing of the voice messaging inputted by user.
It is pointed out that in other embodiments of the invention, the method can also pass through other in step S101
Rational method obtains the dialogue interactive information of corresponding form, the invention is not restricted to this.
After obtaining talking with interactive information, the method can be mutual to dialogue obtained in step S101 in step s 102
Information carries out semantic analysis, thus obtains the input topic corresponding to this dialogue interactive information.Specifically, in the present embodiment, should
Method is processed by the dialogue interactive information inputting user, can go out the semantic letter expressed desired by user with brief analysis
Breath, also can be obtained by the input topic corresponding to this dialogue interactive information according to obtained semantic information.
Such as, user inputs such as the dialogue interactive information of " sunlight outside today is true " to intelligent robot, that
The method is by resolving this dialogue interactive information, it can be deduced that corresponding to the input topic of above-mentioned dialogue interactive information
For " chat gas ".
It is pointed out that in other embodiments of the invention, it is also possible to determined by other reasonable manners
Dialogue input topic corresponding to interactive information, the invention is not restricted to this.Such as, in one embodiment of the invention, the party
Method can determine input topic by the way of the keyword in the dialogue interactive information that counting user inputs or key word.
As it is shown in figure 1, after obtaining the input topic corresponding to dialogue interactive information, the method can root in step s 103
According to this input topic, utilize input topic and each candidate to export the topic transition probability between topic, export words from candidate
Topic is determined final output topic.
In the present embodiment, all can there is topic transition probability in topic set between any two topic, wherein, topic turns
Carry out the height of the probability shifted between shifting probability two topics of sign, i.e. obtained the probability of another topic by a topic
Just.The numerical value of the topic transition probability between two topics is the highest, then then entered into what another said by one of them topic
The probability (i.e. the probability of topic transfer) of topic is the biggest;Otherwise, the numerical value of the topic transition probability between two topics is the least,
The most then entered into the probability of another topic by one of them topic the least.In the present embodiment, between each topic
Topic transition probability is relevant with current interaction scenarios.
So, according to the topic transition probability between each topic, also it is assured that out corresponding to inputting each of topic
Individual candidate exports the probit that topic is occurred, thus just can export from candidate according to this probit and determine topic finally
Output topic.
In the present embodiment, it is preferable that the method in step s 103 from the value of topic transition probability at certain probability model
Enclose in interior output topic and randomly select out final output topic.Certainly, in other embodiments of the invention, it is also possible to logical
Cross other reasonable manners to export from candidate topic is chosen final output topic, the invention is not restricted to this.
It can be seen that the man-machine interaction method for intelligent robot that provided of the present embodiment is really from foregoing description
Surely, during output topic, it is no longer to use fixing dialogue storehouse or language model to realize as existing man-machine interaction method, and
It is to determine output topic by the topic transition probability relevant with current interaction scenarios, the output topic energy so determined
Enough make whole interactive process more lively and hommization.
Embodiment two:
Fig. 2 shows the flowchart of the man-machine interaction method for intelligent robot that the present embodiment provided.
As in figure 2 it is shown, the man-machine interaction method that the present embodiment is provided obtains user's input the most in step s 201
Dialogue interactive information, carries out semantic analysis to dialogue interactive information accessed in step S201 subsequently in step S202,
Thus obtain the input topic corresponding to this dialogue interactive information.
It is pointed out that step S201 and step S202 in the present embodiment realizes principle and the process that realizes with above-mentioned
In embodiment one, step S101 is similar with the content involved by step S102, therefore at this no longer to step S201 and step S202
Identical content repeats.
In the present embodiment, the method in addition to getting the dialogue interactive information of user's input, also can in step s 201
Obtain the scene information relevant to current interaction process.Specifically, in the present embodiment, the method is accessed in step s 201
Scene information preferably include: temporal information, location information, environmental information and people information.Wherein, temporal information can table
Levying the concrete time that current session process is occurred, location information can characterize the particular location that current session process is occurred,
Environmental information then can characterize user during current session and the correlate around intelligent robot, and people information is then
Personage involved in current session and relevant device/device can be characterized.Wherein, temporal information and positional information can divide
Not time service function and positioning function by GPS obtains, and environmental information and/or people information can be come by robot vision
Capture.
It is pointed out that in other embodiments of the invention, the scene that the method is accessed in step s 201
Information both can be only any one in item listed above or several, it is also possible to is other unlisted reasonable items, it is also possible to be
After any one in item listed above several with the combination in any of other unlisted reasonable items, the invention is not restricted to this.Permissible
Cognitive ground, in case of need, may be used for any information that scene judges, can serve as scene information and obtain
Take.Certainly, the scene information got is the most, then the scene reconstructed by these scene informations and real scene
The most accurate, the feedback information of so final output the most just more conforms to the actual expectation of user.
After the input topic of the dialogue interactive information obtained by obtaining corresponding to step S201, this method can be in step
S203 obtains the ID of input topic, and utilizes multidimensional topic transfer matrix to determine the ID of output topic.Obtaining output words
After the ID of topic, the method also can be obtained by the topic corresponding to this ID, i.e. obtains exporting topic.
In the present embodiment, different semantemes can be classified by the method, and marks each classification with unique ID
Know.The such as ID corresponding to " boring " can be 1001, and the ID corresponding to " merely cuisines " can be 2001, and " chat gas " is corresponding
ID can be 2002, the ID corresponding to " work of earning a livelihood " can be 2003 etc..
Such as, if user inputs such as the dialogue interactive information of " good boring ", then can be obtained by semantic analysis
Semanteme to dialogue interactive information is " boring " or " being fed up with things ", and this semanteme then can be corresponding to a unique numeral
ID (such as 1001), has the most also determined that out the ID of the input topic corresponding to dialogue interactive information.
All can there is a topic transition probability between the ID of each topic, also be the formation of more than one between the most all topics
Dimension topic transfer matrix, the topic transfer that in multidimensional topic transfer matrix, each element i.e. represents between corresponding two topics is general
Rate.It is pointed out that in different embodiments of the invention, the value of the topic transition probability between topic both can be logical
Cross statistics experience and carry out artificial default settings, it is also possible to obtain by being processed to analyze to dialogue interactive information, this
Invention is not limited to this.
In step S203, according to the ID of input topic, known multidimensional topic transfer matrix is utilized also to be assured that
Go out to export the ID of topic.So in step S204 the method the most just can obtain according to the ID determined in step S203 right
The topic answered, i.e. exports topic.
Such as, if the dialogue interactive information of user's input is " good boring ", the topic ID of this dialogue interactive information is
Topic transition probability between topic " merely cuisines " and input topic that 1001, ID is " 2001 " is 0.2, if ID is " 2002 "
Topic transition probability between topic " chat gas " and input topic is 0.5, and ID is topic " work of earning a livelihood " and the input words of " 2003 "
Topic transition probability between topic is 0.3.The value of the topic transition probability exporting topic due to these three candidate is all higher than
Certain probit, therefore the method just can export from above three candidate and randomly select one topic as final defeated
Go out topic, such as by " chat gas " as final output topic.
Again as in figure 2 it is shown, in the present embodiment, after determining output topic, the method will be distinguished in step S205
Whether mating of each candidate's feedback information that calculating scene information is comprised with output topic, and according to step in step S206
Matching result obtained by rapid S205 determines final required feedback information from candidate's feedback information, and this is anti-
Feedforward information exports to user.
Scene dialogue is noteworthy characterized by scene dialogue and has many restrictive conditions for distinguishing general dialogue and exist.With
General chat conversations is compared, and much for general chat conversations, answer the most accurately is not appropriate under current scene defeated
Go out to user.And other sub-fraction answer, owing to it using higher dialogue constraint condition, therefore these answers are often
Than the answer of general dialogue chat, there is more preferable Consumer's Experience.
Various scenes in real world are expressed as computer and can process by the method that the present embodiment is provided
Data mode, specifically, the method is by according to accessed scene information, by the scene table corresponding to this scene information
It is shown as the form corresponding to the field under different dimensions.
Such as, for " in Beijing, this morning is fine day " such a scene, the method can arrange initial time and be
" point on the same day eight ", end time be " point on the same day 12 ", location information be " Beijing ", Weather information be the information of " fine ".These are four years old
The information of individual field just defines a kind of discernible well-determined scene of computer, and i.e. " in Beijing, this morning is fine
My god ".
Therefore to make the feedback information finally given can more conform to current session scene, the method is in step
In S205, whether meeting each candidate's feedback information that calculating scene information is comprised with output topic respectively mates.Such as, when
User is to carry out man-machine interaction, at this moment the method meeting with intelligent robot under " at home, outside is sunny " this scene
All candidate's feedback informations in the output topic " chat gas " determined are analyzed, judge that these candidates feed back with this
Whether information meets " environment: fine;Position: family " whether this scene information, i.e. candidate's feedback information match with current scene.
Such as, for " at home, outside is sunny " this session operational scenarios, for two in output topic " chat gas "
Individual candidate's feedback information " weather is the best, wants that consideration is gone out for a walk the when of boring " and " China's ALONG COASTAL FUJIAN meets with again
Meet typhoon ", Article 1 candidate's feedback information matches with scene information, and Article 2 candidate's feedback information is with scene information not
Coupling, therefore Article 1 candidate's feedback information the most just can be exported to use by the method as final required feedback information
Family.
It can be seen that the man-machine interaction method for intelligent robot that the present embodiment is provided can from foregoing description
Determining final feedback information in conjunction with actual interaction scenarios, compared to existing man-machine interaction method, the method enables to
The feedback information of output more conforms to current interaction scenarios, thus user and body in intelligent robot interaction are greatly improved
Test.
Present invention also offers a kind of human-computer interaction device for intelligent robot, Fig. 3 shows should in the present embodiment
The structural representation of device.
As it is shown on figure 3, the human-computer interaction device that the present embodiment is provided preferably includes: dialogue interactive information acquisition module
301, input topic determines that module 302, output topic determine module 303 and feedback information generation module 304.Wherein, dialogue
Interactive information acquisition module 301 is for obtaining the dialogue interactive information of user's input.
It is pointed out that in different embodiments of the invention, dialogue interactive information acquisition module 301 can be different
Device or equipment realize obtaining the dialogue interactive information of multi-form.Such as, in an embodiment of invention, dialogue is handed over
Mutual information acquisition module 301 can use the keyboard provisioned in intelligent robot to realize, and so dialogue interactive information obtains mould
Block 301 the most just can get the text message of user's input;In another embodiment of the present invention, dialogue interactive information obtains
Delivery block 301 can also use the voice messaging collecting device (such as mike) provisioned in intelligent robot to realize, so
Dialog information acquisition module 301 is the most just to get the voice messaging of user's input, further, dialog information acquisition module
301 can also carry out natural language processing to this voice messaging got thus obtain corresponding text message;And at this
In another bright embodiment, dialog information acquisition module 301 can also use corresponding device or equipment to obtain language simultaneously
Message breath and text message.
After getting dialogue interactive information, dialog information acquisition module 301 can be by the transmission of this dialogue interactive information to defeated
Enter topic and determine module 302, to be determined that module 302 carries out semantic analysis to dialogue interactive information by input topic, thus obtain
To the input topic corresponding to talking with interactive information.
Specifically, in the present embodiment, input topic determines that module 302 is carried out by the dialogue interactive information inputting user
Process, can go out, with brief analysis, the semantic information expressed desired by user, also can be obtained by according to obtained semantic information
Input topic corresponding to this dialogue interactive information.
It is pointed out that in other embodiments of the invention, input topic determines that module 302 can also pass through other
Reasonable manner determines dialogue input topic corresponding to interactive information, the invention is not restricted to this.Such as, the present invention's
In one embodiment, input topic determines the keyword in the dialogue interactive information that module 302 can be inputted by counting user
Or the mode of key word determines input topic.
After obtaining the input topic corresponding to dialogue interactive information, input topic determines that module 302 can will input topic
Be transferred to export topic and determine module 303, with by output topic determine module 303 according to this input topic, according to input topic
And each candidate exports the dialogue transition probability between topic, export from candidate and topic is determined output topic.
It is pointed out that in the present embodiment, output topic determine module 303 determine output topic concrete principle and
Realize process and step S203 in step S103 in above-described embodiment one or embodiment two and the content class involved by step S204
Seemingly, therefore no longer output topic is determined that module 303 repeats at this.
In the present embodiment, dialogue interactive information acquisition module 301 can also be as described by step S201 in embodiment two
Obtain the scene information relevant to current interaction process like that.Feedback information generation module 304 the most just can be according to dialog information
Scene information accessed by acquisition module 301 and output topic determine that the output topic that module 303 is determined generates instead
Feedforward information.
It is pointed out that in the present embodiment, feedback information generation module 304 generates according to scene information and output topic
In the implementing principle and realize process and above-described embodiment two of feedback information involved by step S205 and step S206
Content is similar to, therefore no longer repeats feedback information generation module 304 at this.
It should be understood that disclosed embodiment of this invention is not limited to ad hoc structure disclosed herein or processes step
Suddenly, the equivalent that should extend to these features that those of ordinary skill in the related art are understood substitutes.It should also be understood that
It is that term as used herein is only used for describing the purpose of specific embodiment, and is not intended to limit.
" embodiment " mentioned in description or " embodiment " mean special characteristic, the structure in conjunction with the embodiments described
Or characteristic is included at least one embodiment of the present invention.Therefore, the phrase " reality that description various places throughout occurs
Execute example " or " embodiment " same embodiment might not be referred both to.
Although above-mentioned example is for illustrating present invention principle in one or more application, but for the technology of this area
For personnel, in the case of without departing substantially from the principle of the present invention and thought, hence it is evident that can in form, usage and the details of enforcement
Above various modifications may be made and need not pay creative work.Therefore, the present invention is defined by the appended claims.
Claims (10)
1. the man-machine interaction method for intelligent robot, it is characterised in that including:
Dialogue interactive information obtaining step, obtains the dialogue interactive information of user's input;
Input topic determines step, described dialogue interactive information is carried out semantic analysis, obtains believing alternately corresponding to described dialogue
The input topic of breath;
Output topic determines step, according to described input topic, utilizes described input topic and each candidate to export between topic
Topic transition probability, export from candidate and topic determine output topic.
2. the method for claim 1, it is characterised in that described method also includes:
Feedback information generation step, according to scene information and the described output topic of the current interaction process got, generates anti-
Feedforward information.
3. method as claimed in claim 2, it is characterised in that in described feedback information generation step, judge output respectively
Whether each candidate's feedback information that topic is comprised mates with described scene information, and determines required according to matching result
Feedback information.
4. method as claimed in claim 3, it is characterised in that described scene information include any one in item set forth below or
Several:
Temporal information, location information, environmental information and people information.
5. the method as according to any one of Claims 1 to 4, it is characterised in that determine in step at described output topic, root
According to the ID of described input topic, multidimensional topic transfer matrix is utilized to determine the ID of output topic, and according to described output topic
ID determine output topic.
6. the human-computer interaction device for intelligent robot, it is characterised in that including:
Dialogue interactive information acquisition module, for obtaining the dialogue interactive information of user's input;
Input topic determines module, for described dialogue interactive information is carried out semantic analysis, obtains handing over corresponding to described dialogue
The input topic of mutual information;
Output topic determines module, for according to described input topic, utilizes described input topic and each candidate to export topic
Between topic transition probability, export from candidate and topic determine output topic.
7. device as claimed in claim 6, it is characterised in that described device also includes:
Feedback information generation module, for the scene information according to the current interaction process got and described output topic, raw
Become feedback information.
8. device as claimed in claim 7, it is characterised in that described feedback information generation module is configured to judge output respectively
Whether each candidate's feedback information that topic is comprised mates with described scene information, and determines required according to matching result
Feedback information.
9. device as claimed in claim 8, it is characterised in that described scene information include any one in item set forth below or
Several:
Temporal information, location information, environmental information and people information.
10. the device as according to any one of claim 6~9, it is characterised in that described output topic determines that module is configured to
According to the ID of described input topic, multidimensional topic transfer matrix is utilized to determine the ID of output topic, and according to described output words
The ID of topic determines output topic.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610703758.0A CN106294854B (en) | 2016-08-22 | 2016-08-22 | Man-machine interaction method and device for intelligent robot |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610703758.0A CN106294854B (en) | 2016-08-22 | 2016-08-22 | Man-machine interaction method and device for intelligent robot |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106294854A true CN106294854A (en) | 2017-01-04 |
CN106294854B CN106294854B (en) | 2019-12-24 |
Family
ID=57662122
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610703758.0A Active CN106294854B (en) | 2016-08-22 | 2016-08-22 | Man-machine interaction method and device for intelligent robot |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106294854B (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107016046A (en) * | 2017-02-20 | 2017-08-04 | 北京光年无限科技有限公司 | The intelligent robot dialogue method and system of view-based access control model displaying |
CN107146610A (en) * | 2017-04-10 | 2017-09-08 | 北京猎户星空科技有限公司 | A kind of determination method and device of user view |
CN108268443A (en) * | 2017-12-21 | 2018-07-10 | 北京百度网讯科技有限公司 | It determines the transfer of topic point and obtains the method, apparatus for replying text |
CN108319599A (en) * | 2017-01-17 | 2018-07-24 | 华为技术有限公司 | A kind of interactive method and apparatus |
CN108388674A (en) * | 2018-03-26 | 2018-08-10 | 百度在线网络技术(北京)有限公司 | Method and apparatus for pushed information |
CN109189363A (en) * | 2018-07-24 | 2019-01-11 | 上海常仁信息科技有限公司 | A kind of robot of human-computer interaction |
CN109979450A (en) * | 2019-03-11 | 2019-07-05 | 青岛海信电器股份有限公司 | Information processing method, device and electronic equipment |
CN110047479A (en) * | 2019-04-18 | 2019-07-23 | 深圳市三宝创新智能有限公司 | A kind of voice interaction device for intelligent robot |
CN110209792A (en) * | 2019-06-13 | 2019-09-06 | 苏州思必驰信息科技有限公司 | Talk with painted eggshell generation method and system |
CN110659355A (en) * | 2018-06-29 | 2020-01-07 | 南京芝兰人工智能技术研究院有限公司 | Conversation control method and system |
CN110706702A (en) * | 2019-10-15 | 2020-01-17 | 芜湖盟博科技有限公司 | Infinite stage multi-turn dialogue method for speech recognition |
CN111223477A (en) * | 2018-11-23 | 2020-06-02 | 珠海格力电器股份有限公司 | Method and device for topic upward movement during man-machine interaction |
CN111259128A (en) * | 2020-01-19 | 2020-06-09 | 出门问问信息科技有限公司 | Method and device for generating conversation target sequence and readable storage medium |
CN113392192A (en) * | 2020-03-12 | 2021-09-14 | 北大方正集团有限公司 | Language processing method, device, system and storage medium |
CN115809669A (en) * | 2022-12-30 | 2023-03-17 | 联通智网科技股份有限公司 | Conversation management method and electronic equipment |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130159291A1 (en) * | 2011-12-14 | 2013-06-20 | Microsoft Corporation | Ranking search results using weighted topologies |
US20150179168A1 (en) * | 2013-12-20 | 2015-06-25 | Microsoft Corporation | Multi-user, Multi-domain Dialog System |
CN105094315A (en) * | 2015-06-25 | 2015-11-25 | 百度在线网络技术(北京)有限公司 | Method and apparatus for smart man-machine chat based on artificial intelligence |
CN105740948A (en) * | 2016-02-04 | 2016-07-06 | 北京光年无限科技有限公司 | Intelligent robot-oriented interaction method and device |
CN105798918A (en) * | 2016-04-29 | 2016-07-27 | 北京光年无限科技有限公司 | Interactive method and device for intelligent robot |
-
2016
- 2016-08-22 CN CN201610703758.0A patent/CN106294854B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130159291A1 (en) * | 2011-12-14 | 2013-06-20 | Microsoft Corporation | Ranking search results using weighted topologies |
US20150179168A1 (en) * | 2013-12-20 | 2015-06-25 | Microsoft Corporation | Multi-user, Multi-domain Dialog System |
CN105094315A (en) * | 2015-06-25 | 2015-11-25 | 百度在线网络技术(北京)有限公司 | Method and apparatus for smart man-machine chat based on artificial intelligence |
CN105740948A (en) * | 2016-02-04 | 2016-07-06 | 北京光年无限科技有限公司 | Intelligent robot-oriented interaction method and device |
CN105798918A (en) * | 2016-04-29 | 2016-07-27 | 北京光年无限科技有限公司 | Interactive method and device for intelligent robot |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108319599A (en) * | 2017-01-17 | 2018-07-24 | 华为技术有限公司 | A kind of interactive method and apparatus |
US11308405B2 (en) | 2017-01-17 | 2022-04-19 | Huawei Technologies Co., Ltd. | Human-computer dialogue method and apparatus |
CN107016046A (en) * | 2017-02-20 | 2017-08-04 | 北京光年无限科技有限公司 | The intelligent robot dialogue method and system of view-based access control model displaying |
CN107146610A (en) * | 2017-04-10 | 2017-09-08 | 北京猎户星空科技有限公司 | A kind of determination method and device of user view |
CN108268443A (en) * | 2017-12-21 | 2018-07-10 | 北京百度网讯科技有限公司 | It determines the transfer of topic point and obtains the method, apparatus for replying text |
CN108388674B (en) * | 2018-03-26 | 2021-11-26 | 百度在线网络技术(北京)有限公司 | Method and device for pushing information |
CN108388674A (en) * | 2018-03-26 | 2018-08-10 | 百度在线网络技术(北京)有限公司 | Method and apparatus for pushed information |
CN110659355A (en) * | 2018-06-29 | 2020-01-07 | 南京芝兰人工智能技术研究院有限公司 | Conversation control method and system |
CN109189363A (en) * | 2018-07-24 | 2019-01-11 | 上海常仁信息科技有限公司 | A kind of robot of human-computer interaction |
CN111223477A (en) * | 2018-11-23 | 2020-06-02 | 珠海格力电器股份有限公司 | Method and device for topic upward movement during man-machine interaction |
CN109979450A (en) * | 2019-03-11 | 2019-07-05 | 青岛海信电器股份有限公司 | Information processing method, device and electronic equipment |
CN110047479A (en) * | 2019-04-18 | 2019-07-23 | 深圳市三宝创新智能有限公司 | A kind of voice interaction device for intelligent robot |
CN110209792A (en) * | 2019-06-13 | 2019-09-06 | 苏州思必驰信息科技有限公司 | Talk with painted eggshell generation method and system |
CN110706702B (en) * | 2019-10-15 | 2021-11-09 | 芜湖盟博科技有限公司 | Infinite stage multi-turn dialogue method for speech recognition |
CN110706702A (en) * | 2019-10-15 | 2020-01-17 | 芜湖盟博科技有限公司 | Infinite stage multi-turn dialogue method for speech recognition |
CN111259128A (en) * | 2020-01-19 | 2020-06-09 | 出门问问信息科技有限公司 | Method and device for generating conversation target sequence and readable storage medium |
CN113392192A (en) * | 2020-03-12 | 2021-09-14 | 北大方正集团有限公司 | Language processing method, device, system and storage medium |
CN115809669A (en) * | 2022-12-30 | 2023-03-17 | 联通智网科技股份有限公司 | Conversation management method and electronic equipment |
CN115809669B (en) * | 2022-12-30 | 2024-03-29 | 联通智网科技股份有限公司 | Dialogue management method and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN106294854B (en) | 2019-12-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106294854A (en) | A kind of man-machine interaction method for intelligent robot and device | |
CN108319599B (en) | Man-machine conversation method and device | |
WO2018036555A1 (en) | Session processing method and apparatus | |
CN105512228B (en) | A kind of two-way question and answer data processing method and system based on intelligent robot | |
CN106297789B (en) | Personalized interaction method and system for intelligent robot | |
CN106407178B (en) | A kind of session abstraction generating method, device, server apparatus and terminal device | |
CN106020488A (en) | Man-machine interaction method and device for conversation system | |
CN106934068A (en) | The method that robot is based on the semantic understanding of environmental context | |
CN106951468A (en) | Talk with generation method and device | |
CN106897263A (en) | Robot dialogue exchange method and device based on deep learning | |
CN111177310B (en) | Intelligent scene conversation method and device for power service robot | |
CN106357942A (en) | Intelligent response method and system based on context dialogue semantic recognition | |
CN105931638A (en) | Intelligent-robot-oriented dialog system data processing method and device | |
CN105704013A (en) | Context-based topic updating data processing method and apparatus | |
CN105488164A (en) | Question and answer (QA) data processing method and device, intelligent robot | |
CN110489513A (en) | A kind of intelligent robot social information processing method and the social intercourse system with people | |
CN106096717A (en) | Information processing method and system towards intelligent robot | |
CN103336788A (en) | Humanoid robot added Internet information acquisition method and system | |
CN109460459A (en) | A kind of conversational system automatic optimization method based on log study | |
CN105868183A (en) | Method and device for predicting staff demission | |
CN116483980A (en) | Man-machine interaction method, device and system | |
CN112256854A (en) | Intelligent AI conversation method and device based on AI natural language understanding | |
CN105988591A (en) | Intelligent robot-oriented motion control method and intelligent robot-oriented motion control device | |
CN106127526A (en) | Intelligent robot system and method for work thereof | |
CN108040111A (en) | A kind of apparatus and method for supporting natural language interaction |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |