CN108198559A - A kind of voice control robot system for learning action - Google Patents
A kind of voice control robot system for learning action Download PDFInfo
- Publication number
- CN108198559A CN108198559A CN201810075101.3A CN201810075101A CN108198559A CN 108198559 A CN108198559 A CN 108198559A CN 201810075101 A CN201810075101 A CN 201810075101A CN 108198559 A CN108198559 A CN 108198559A
- Authority
- CN
- China
- Prior art keywords
- training
- action
- voice
- text
- control
- 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
Links
- 238000012549 training Methods 0.000 claims abstract description 318
- 238000000034 method Methods 0.000 claims abstract description 42
- 238000004891 communication Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims 1
- 238000005452 bending Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000006698 induction Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
- B25J11/0005—Manipulators having means for high-level communication with users, e.g. speech generator, face recognition means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J13/00—Controls for manipulators
- B25J13/003—Controls for manipulators by means of an audio-responsive input
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; 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 OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/06—Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids
Abstract
Present specification, which provides one kind, can learn the voice control robot system of action, in one embodiment, first speech recognition equipment receives the training voice that trainer sends out, training voice is identified, generation training speech text, the training action corresponding with training voice of action recognition device intake trainer's displaying, generates training action program, and described information fusing device will training speech text and training action program pairing typing speech text table corresponding with operation program.Second speech recognition equipment receives the control voice that controller sends out, control voice is identified, generate control voice text, action selection device match selection in speech text table corresponding with action sequence goes out trained speech text, and then training action program is obtained, action executive device performs the action of training action procedure stipulation.
Description
Technical field
The application belongs to robotic technology field, specifically, being related to a kind of system of voice control robot for learning action
System and method.
Background technology
In home-services robot field, there is the robot system of many voice control types at present, it, can by voice control
Calling Internet resources, obtaining feedback output or calling the preset program of robot interior, the fortune of robot is controlled
It is dynamic.
Above-mentioned robot system has the defects of certain in terms of handicapped old man is assisted, for complicated auxiliary movement
It cannot complete glibly.
The contextual model that the Intelligent robot system of another voice control can be set according to preprogramming, makes one
Serial complicated action, but because practical application scene is multifarious, many preset contextual models are excessively coarse, it is impossible to realize
Accurately, handicapped old man, children etc. are helped customizablely.
Invention content
For realize it is smooth, accurate, customizable assist old man, children etc., the present disclosure provides a kind of technical solution,
Voice is associated with the method called with action with a variety of including a variety of voice control robot systems for learning action.
【One】The present disclosure provides a kind of voice control robot system for learning action, system includes training eventually
End, communication link and robot body, training terminal include the first speech recognition equipment, action recognition device and information fusion dress
It puts, robot body includes the second speech recognition equipment, action selection device and action executive device, training terminal and robot
Ontology carries out information exchange by communication link.
First speech recognition equipment receives the training voice that trainer sends out, and training voice is identified, is generated
Training speech text, the training action corresponding with training voice of action recognition device intake trainer's displaying, generation instruction
Practice operation program, described information fusing device will training speech text and training action program pairing typing speech text and action
Program corresponds to table.
The action recognition device includes sensing component, identification software and processor, sensing component intake trainer's displaying
With the corresponding training action of training voice, it is dynamic to generate training the information of identification software and processor processing sensing component intake
Make program.
Second speech recognition equipment receives the control voice that controller sends out, and control voice is identified, and generates
Control voice text, action selection device match selection in speech text table corresponding with action sequence goes out and control voice
The identical and similar trained speech text of text, and then obtain training the corresponding training action program of speech text, the action
Executive device performs the action of training action procedure stipulation.
The speech text table corresponding with operation program that training obtains is sent to the robot body by the trained terminal,
The variation of the trained terminal monitoring speech text table corresponding with operation program, newer speech text is corresponding with operation program
Table is sent to the robot body.
The action executive device performs the mechanical structure unit of training action program and the trainer shows that training is dynamic
The extremity body structures unit of work has the relationship mapped one by one.
The action include four limbs, finger, neck, eyes, lip with respect to or the change of absolute spatial position, shape
Variation, further includes the change of the spatial position of action assisting device, the variation of shape.
【Two】The present disclosure provides a kind of methods that voice is associated with action and is called, and include the following steps：
S101：Training terminal receives the training voice that trainer sends out, and training voice is identified, and generates training voice text
This；
S102：The training action corresponding with training voice of training terminal intake trainer's displaying, generates training action program；
S103：Training terminal will train speech text and training action program pairing typing speech text corresponding with operation program
Table；
S104：Robot body receives the control voice that controller sends out, and control voice is identified, generation control voice text
This；
S105：Robot body speech text and match selection in the corresponding table of action sequence go out it is identical with control voice text and
Similar trained speech text, and then obtain training the corresponding training action program of speech text；
S106：Robot body performs the action of training action procedure stipulation.
【Three】The present disclosure provides a kind of voice control robot system for learning action, system includes training and controls
Terminal processed, communication link and robot body, Training Control terminal are melted comprising speech recognition equipment, action recognition device and information
It attaches together and puts, robot body includes action selection device and action executive device, and Training Control terminal passes through with robot body
Communication link carries out information exchange, and speech recognition equipment receives the training voice that trainer sends out, and training voice is identified,
Generation training speech text, the training action corresponding with training voice of action recognition device intake trainer's displaying, generation instruction
Practice operation program, information fuse device will training speech text and training action program pairing typing speech text and operation program
The speech text table corresponding with operation program that training obtains is sent to the robot body by corresponding table, training terminal, training
The speech recognition equipment of control terminal receives the control voice that controller sends out, and control voice is identified, generation control language
Control voice text is sent to robot body by sound text, Training Control terminal, and action selection device is in speech text with moving
Make sequence and correspond to match selection in table to go out identical with control voice text and similar trained speech text, and then obtain training language
The corresponding training action program of sound text, the action executive device perform the action of training action procedure stipulation.
【Four】The present disclosure provides a kind of methods that voice is associated with action and is called, and include the following steps：
S201：Training Control terminal receives the training voice that trainer sends out, and training voice is identified, and generates training voice
Text；
S202：The training action corresponding with training voice of Training Control terminal intake trainer's displaying, generates training action journey
Sequence；
S203：Training Control terminal will training speech text and training action program pairing typing speech text and operation program pair
Answer table；
S204：The speech recognition equipment of Training Control terminal receives the control voice that controller sends out, and control voice is known
Not, control voice text is generated；
S205：Control voice text is sent to robot body by Training Control terminal；
S206：The action selection device of robot body match selection in speech text table corresponding with action sequence goes out and controls
The identical and similar trained speech text of speech text, and then obtain training the corresponding training action program of speech text；
S207：Robot body performs the action of training action procedure stipulation.
【Five】The present disclosure provides a kind of voice control robot system for learning action, system includes training and controls
Terminal processed, communication link and robot body, Training Control terminal include the first speech recognition equipment, action recognition device and letter
Fusing device is ceased, robot body includes third speech recognition equipment, action selection device and action executive device, Training Control
Terminal carries out information exchange with robot body by communication link, and speech recognition equipment receives the training language that trainer sends out
Sound, to training voice be identified, generate training speech text, action recognition device intake trainer displaying with training voice
Corresponding training action, generates training action program, and information fuse device will training speech text and the pairing of training action program
Typing speech text table corresponding with operation program, the speech text table corresponding with operation program that training terminal obtains training are sent
To robot body, Training Control terminal receives the control voice that controller sends out, control voice is sent to robot body,
Control voice is identified in the third speech recognition equipment of robot body, generates control voice text, acts selection device
Go out identical with control voice text and similar trained speech text with match selection in the corresponding table of action sequence in speech text,
And then obtain training the corresponding training action program of speech text, the action executive device performs training action procedure stipulation
Action.
【Six】The present disclosure provides a kind of methods that voice is associated with action and is called, and include the following steps：
S301：Training Control terminal receives the training voice that trainer sends out, and training voice is identified, and generates training voice
Text；
S302：The training action corresponding with training voice of Training Control terminal intake trainer's displaying, generates training action journey
Sequence；
S303：Training Control terminal will training speech text and training action program pairing typing speech text and operation program pair
Answer table；
S304：The control voice that the reception controller of Training Control terminal sends out, robot body is sent to by control voice；
S305：Control voice is identified in the third speech recognition equipment of robot body, generates control voice text,
S306：The action selection device of robot body match selection in speech text table corresponding with action sequence goes out and controls
The identical and similar trained speech text of speech text, and then obtain training the corresponding training action program of speech text；
S307：Robot body performs the action of training action procedure stipulation.
Description of the drawings
Attached drawing 1 forms figure for the system logic of embodiment 1 and embodiment 2；
Attached drawing 2 forms figure for the system logic of embodiment 2 and embodiment 3.
Specific embodiment
In order to enable the technical characteristic of the present invention is definitely, intuitively, embodiment is described below in conjunction with the accompanying drawings, this
The technical staff in field is it should be understood that these embodiments are exemplary in nature, not to the limitation of technical solution of the present invention, and
And section Example can be combined with each other or be combined with other known solutions.
【Embodiment 1】
Figure is formed referring to the system logic of attached the present embodiment shown in FIG. 1.System includes training terminal, communication link and robot
Ontology.In the present embodiment, training terminal is a smart mobile phone.The first speech recognition equipment, action are included in training terminal
Identification device and information fuse device.Robot body includes the second speech recognition equipment, action selection device and action and performs
Device.Training terminal is interacted with robot body by communication link information.
First speech recognition equipment, the second speech recognition equipment include frequency induction device, audio processing chip, language
Sound identification chip, speech recognition software and speech recognition software are rely the hardware resources such as processor of operation.
First speech recognition equipment receives the training voice that trainer sends out, and training voice is identified, is generated
Training speech text, the training action corresponding with training voice of action recognition device intake trainer's displaying, generation instruction
Practice operation program, described information fusing device will training speech text and training action program pairing typing speech text and action
Program corresponds to table.
The action recognition device senses component, image recognition software and image recognition software including camera module etc. and relies
The hardware resources such as the processor with operation.It can be optical device, ultrasonic radar etc. to sense component, intake trainer's displaying with
The corresponding training action of training voice, the action include the opposite or absolute space position of four limbs, finger, neck, eyes, lip
Change, the variation of shape put, further include the change of the spatial position of action assisting device, the variation of shape, identification software and
The information of processor processing sensing component intake, generates training action program.
When the first speech recognition equipment and action recognition device are run, trainer effectively knows in the first speech recognition equipment
Trained voice is sent out in other range, such as says " going ahead ", " three steps of going ahead ", " turning right ", " turning right 90 degree ", " bow
Above-mentioned voice is converted to electric current letter by the voices such as body ", " bending one's legs ", " semi-girder ", the voice sensing device of the first speech recognition equipment
Number, the audio processing chip of the first speech recognition equipment handles current signal, and obtained result is transferred to speech recognition
Chip and speech recognition software processing, match corresponding text, as training using the sound bank of the first speech recognition equipment
Speech text preserves in memory.
Trainer is while trained voice is sent out（Or the later timeslot set by training software）Show with
Training voice " going ahead ", " three steps of going ahead ", " turning right ", " turning right 90 degree ", " bending ", " bending one's legs ", " semi-girder " etc.
The corresponding training action of voice, the image that the sensing component of action recognition device will absorb above-mentioned training action, by image file
In memory, the identification software of action recognition device analyzes corresponding image file using its recognizer, obtains shadow for storage
As the information of action process sequence included in file, this information includes four limbs, finger, neck, eyes, the lip of trainer
The change of the spatial position at positions, the variation of shape are waited, further includes the spatial position of the action assisting device of trainer's carrying
Change and the variation of shape, the action assisting device can be the stage properties such as wooden stick, crutch, gloves, clothes, can be used to imitate limbs
Structure make action.Action recognition algorithm can be according to the difference of sequence image frame corresponding with linearly variation in image file
Divide variation, analyze the main body of generation action, distance, angle, amplitude and the speed of action, such as：Trainer's right arm is from parallel
Become in backbone perpendicular to backbone, then action recognition algorithm software can analyze right arm from the image of sensing component capture and generate
90 degree of rotation；Trainer moves ahead three steps, then action recognition algorithm software can be analyzed from the image of sensing component capture
Go out step number and travel distance.The information of the action process sequence forms a program file, referred to as training action program.
Preferably, action process sequence information is inserted each mechanical structure list with acting executive device by action recognition device
Member has the process template of format match relationship, calibrate order, period, direction, angle that each mechanical structure unit should run,
The information such as speed and amplitude generate training action program.The characteristic information of the mechanical structure unit of the action executive device is pre-
Terminal is trained in first typing, and the extremity body structures unit for showing training action with trainer has the relationship mapped one by one.
Described information fusing device, the processor and memory space of operation of relying including information fusion software and software.Letter
It ceases fusing device and handles the training speech text of the first speech recognition equipment generation and the training action of action recognition device generation
Training speech text and training action program are matched typing speech text table corresponding with operation program by program.
The speech text table corresponding with operation program that training obtains is sent to the robot body by the trained terminal,
The variation of the trained terminal monitoring speech text table corresponding with operation program, newer speech text is corresponding with operation program
Table is sent to the robot body.
Second speech recognition equipment receives the control voice that controller sends out, and control voice is identified, and generates
Control voice text, action selection device match selection in speech text table corresponding with action sequence goes out and control voice
The identical and similar trained speech text of text, and then obtain training the corresponding training action program of speech text, the action
Executive device performs the action of training action procedure stipulation.
Controller, for example, the old man in robot body at one's side, when needing to control the robot body act, second
Send out control voice in the effectively perceive and identification range of speech recognition equipment, for example, " going ahead ", " three steps of going ahead ", " to
Right-hand rotation ", " turning right 90 degree ", " bending ", " bending one's legs ", " semi-girder " etc., the second speech recognition equipment receive what controller sent out
Control voice is identified in control voice, generates control voice text.
Action selection device goes out in speech text with match selection in the corresponding table of action sequence identical with control voice text
With similar trained speech text, and then obtain training speech text corresponding training action program.
The action that executive device performs training action procedure stipulation is acted, i.e., each mechanical structure unit is with time of procedure stipulation
Sequence, sequential, direction, angle, speed and amplitude operating.
Preferably, action executive device performs the mechanical structure unit of training action program and the trainer shows training
The extremity body structures unit of action has the relationship mapped one by one.
【Embodiment 2】
A kind of method that voice with action is associated with and is called is present embodiments provided, can be applied in the system of above-described embodiment 1,
Include the following steps：
S11：Training terminal receives the training voice that trainer sends out, and training voice is identified, and generates training speech text；
S12：The training action corresponding with training voice of training terminal intake trainer's displaying, generates training action program；
S13：Training terminal will training speech text and training action program pairing typing speech text table corresponding with operation program；
S21：Robot body receives the control voice that controller sends out, and control voice is identified, generation control voice text
This；
S22：Robot body speech text and match selection in the corresponding table of action sequence go out it is identical with control voice text and
Similar trained speech text, and then obtain training the corresponding training action program of speech text；
S23：Robot body performs the action of training action procedure stipulation.
It is apparent to those skilled in the art that for convenience and simplicity of description, the method for foregoing description
In, the concrete property and the course of work of module and unit can refer to the correspondence situation in aforementioned system embodiment 1, herein not
It repeats again.
【Embodiment 3】
Figure is formed referring to the system logic of attached the present embodiment shown in Fig. 2.Present embodiments provide a kind of language for learning action
Sound control robot system on the basis of embodiment 1, increases the control ability of trained terminal, reduces robot body
The second speech recognition equipment.In the present embodiment, controller can control robot using Training Control terminal with remote speech
Ontology.
In the present embodiment, system includes Training Control terminal, communication link and robot body, Training Control terminal packet
Containing speech recognition equipment, action recognition device and information fuse device, robot body includes action selection device and action is held
Luggage is put, and Training Control terminal carries out information exchange with robot body by communication link.
Speech recognition equipment receives the training voice that trainer sends out, and training voice is identified, and generates training voice
Text, the training action corresponding with training voice of action recognition device intake trainer's displaying, generates training action program, letter
Speech text and training action program pairing typing speech text table corresponding with operation program will be trained by ceasing fusing device, and training is eventually
The speech text table corresponding with operation program that training obtains is sent to the robot body by end.
Controller can be same people with trainer, be sent out in the speech recognition equipment sphere of action of Training Control terminal
Control voice, speech recognition equipment receive the control voice that controller sends out, control voice are identified, and generate control voice
Control voice text is sent to robot body by text, Training Control terminal.
The action selection device of robot body match selection in speech text table corresponding with action sequence goes out and controls
The identical and similar trained speech text of speech text, and then obtain training the corresponding training action program of speech text, it is described
Act the action that executive device performs training action procedure stipulation.
【Embodiment 4】
A kind of method that voice with action is associated with and is called is present embodiments provided, can be applied in the system of above-described embodiment 3,
Include the following steps：
S201：Training Control terminal receives the training voice that trainer sends out, and training voice is identified, and generates training voice
Text；
S202：The training action corresponding with training voice of Training Control terminal intake trainer's displaying, generates training action journey
Sequence；
S203：Training Control terminal will training speech text and training action program pairing typing speech text and operation program pair
Answer table.
S204：The speech recognition equipment of Training Control terminal receives the control voice that controller sends out, to control voice into
Row identification, generates control voice text；
S205：Control voice text is sent to robot body by Training Control terminal；
S206：Robot body speech text and match selection in the corresponding table of action sequence go out it is identical with control voice text and
Similar trained speech text, and then obtain training the corresponding training action program of speech text；
S207：Robot body performs the action of training action procedure stipulation.
It is apparent to those skilled in the art that for convenience and simplicity of description, the method for foregoing description
In, the concrete property and the course of work of module and unit can refer to the correspondence situation in aforementioned system embodiment 3, herein not
It repeats again.
【Embodiment 5】
A kind of voice control robot system for learning action is present embodiments provided, on the basis of embodiment 1, is increased
The control ability of training terminal,
In the present embodiment, system includes Training Control terminal, communication link and robot body, and Training Control terminal includes the
One speech recognition equipment, action recognition device and information fuse device, robot body include third speech recognition equipment, action
Selection device and action executive device, Training Control terminal carry out information exchange, voice with robot body by communication link
Identification device receives the training voice that trainer sends out, and training voice is identified, and generates training speech text, action recognition
The training action corresponding with training voice of device intake trainer's displaying, generates training action program, and information fuse device will
Training speech text and training action program pairing typing speech text table corresponding with operation program, training terminal obtain training
Speech text table corresponding with operation program be sent to robot body.
Training Control terminal receives the control voice that controller sends out, and control voice is sent to robot body, machine
Control voice is identified in the third speech recognition equipment of human body, generates control voice text, action selection device is in language
Sound text goes out identical with control voice text and similar trained speech text with match selection in the corresponding table of action sequence, and then
It obtains training the corresponding training action program of speech text, the action executive device performs the dynamic of training action procedure stipulation
Make.
Preferably, the second speech recognition software in the third speech recognition software and embodiment 1 can be multiplexed.
Preferably, the third speech recognition software only includes speech recognition software and speech recognition software is rely operation
The hardware resources such as processor, without including frequency induction device, audio processing chip.
【Embodiment 6】
A kind of method that voice with action is associated with and is called is present embodiments provided, can be applied in the system of above-described embodiment 5,
Include the following steps：
S301：Training Control terminal receives the training voice that trainer sends out, and training voice is identified, and generates training voice
Text；
S302：The training action corresponding with training voice of Training Control terminal intake trainer's displaying, generates training action journey
Sequence；
S303：Training Control terminal will training speech text and training action program pairing typing speech text and operation program pair
Answer table.
S304：The control voice that the reception controller of Training Control terminal sends out, robot sheet is sent to by control voice
Body；
S305：Control voice is identified in the third speech recognition equipment of robot body, generates control voice text,
S306：The action selection device of robot body match selection in speech text table corresponding with action sequence goes out and controls
The identical and similar trained speech text of speech text, and then obtain training the corresponding training action program of speech text；
S307：Robot body performs the action of training action procedure stipulation.
It is apparent to those skilled in the art that for convenience and simplicity of description, the method for foregoing description
In, the concrete property and the course of work of module and unit can refer to the correspondence situation in aforementioned system embodiment 5, herein not
It repeats again.
In the described system and method for above-described embodiment, since the intelligent terminals such as smart mobile phone are expected stronger and stronger
Voice and image processing function so that robot body can still pass through intelligence in the case of no recognition capability
The sensing capability of terminal-pair external image realizes the diversification of the Motor ability of robot body indirectly.
Those skilled in the art are described herein it will be appreciated that in said one or multiple embodiments
Function can be realized with hardware, software, firmware or their arbitrary combination.It when implemented in software, can be by these work(
It can store in computer-readable medium or be passed as one or more instructions on computer-readable medium or code
It is defeated.
It can be realized in a manner of hardware with reference to the step of described method of present disclosure or algorithm, also may be used
It is realized in a manner of being to perform software instruction by processor.Software instruction can be made of corresponding software module, software mould
Block can be stored on RAM, flash memory, ROM, EPROM, EEPROM, register or any other form well known in the art
In storage medium.A kind of illustrative storage medium is coupled to processor, so as to enable a processor to read from the storage medium
Information, and information can be written to the storage medium.Certainly, storage medium can also be the component part of processor.Processor and
Storage medium can be located in ASIC.
In embodiment provided herein, it should be understood that disclosed system and method can pass through others
Mode is realized.The division of the module, only a kind of division of logic function, can there is other division side in actual implementation
Formula.Multiple modules or unit may be combined or can be integrated into another system or some features can be ignored or does not hold
Row.
Above example is only to illustrate the technical solution of the application, rather than its limitations, although with reference to the foregoing embodiments
The application is described in further detail, it will be understood by those of ordinary skill in the art that：It still can be to aforementioned each
Technical solution recorded in embodiment modifies or carries out equivalent replacement to which part technical characteristic；And these are changed
Or it replaces, the spirit and scope of each embodiment technical solution of the application that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of voice control robot system for learning action, which is characterized in that system includes training terminal, communication link
And robot body, training terminal include the first speech recognition equipment, action recognition device and information fuse device, robot sheet
Body includes the second speech recognition equipment, action selection device and action executive device, and training terminal passes through logical with robot body
Believe that link carries out information exchange.
2. system according to claim 1, which is characterized in that first speech recognition equipment receives what trainer sent out
Training voice is identified in training voice, generates training speech text, action recognition device intake trainer's displaying
Training action corresponding with training voice, generates training action program, and described information fusing device will training speech text and instruction
Practice operation program pairing typing speech text table corresponding with operation program.
3. system according to claim 2, which is characterized in that the action recognition device is soft including sensing component, identification
Part and processor, the training action corresponding with training voice of sensing component intake trainer's displaying, identification software and processor
The information of processing sensing component intake, generates training action program.
4. system according to claim 2, which is characterized in that the trained terminal is by the speech text that training obtains with moving
Make program and correspond to table to be sent to the robot body, the change of the trained terminal monitoring speech text table corresponding with operation program
Change, newer speech text table corresponding with operation program is sent to the robot body.
5. system according to claim 2, which is characterized in that second speech recognition equipment receives what controller sent out
Control voice is identified in control voice, generates control voice text, the action selection device is in speech text and action
Sequence pair answers match selection in table to go out identical with control voice text and similar trained speech text, and then obtains training voice
The corresponding training action program of text, the action executive device perform the action of training action procedure stipulation, and action performs dress
The extremity body structures unit for putting the mechanical structure unit and trainer displaying training action that perform training action program has one by one
The relationship of mapping.
6. a kind of be associated with voice with action the method called, which is characterized in that includes the following steps：
S101：Training terminal receives the training voice that trainer sends out, and training voice is identified, and generates training voice text
This；
S102：The training action corresponding with training voice of training terminal intake trainer's displaying, generates training action program；
S103：Training terminal will train speech text and training action program pairing typing speech text corresponding with operation program
Table；
S104：Robot body receives the control voice that controller sends out, and control voice is identified, generation control voice text
This；
S105：Robot body speech text and match selection in the corresponding table of action sequence go out it is identical with control voice text and
Similar trained speech text, and then obtain training the corresponding training action program of speech text；
S106：Robot body performs the action of training action procedure stipulation.
7. a kind of voice control robot system for learning action, which is characterized in that system includes Training Control terminal, communication
Link and robot body, Training Control terminal include speech recognition equipment, action recognition device and information fuse device, machine
Human body includes action selection device and action executive device, and Training Control terminal is carried out with robot body by communication link
Information exchange, speech recognition equipment receive the training voice that trainer sends out, and training voice is identified, and generate training voice
Text, the training action corresponding with training voice of action recognition device intake trainer's displaying, generates training action program, letter
Speech text and training action program pairing typing speech text table corresponding with operation program will be trained by ceasing fusing device, and training is eventually
The speech text table corresponding with operation program that training obtains is sent to the robot body, the voice of Training Control terminal by end
Identification device receives the control voice that controller sends out, and control voice is identified, and generates control voice text, Training Control
Control voice text is sent to robot body by terminal, and action selection device is in speech text table corresponding with action sequence
With selecting identical with control voice text and similar trained speech text, and then obtain training the corresponding training of speech text
Operation program, the action executive device perform the action of training action procedure stipulation.
8. a kind of be associated with voice with action the method called, which is characterized in that includes the following steps：
S201：Training Control terminal receives the training voice that trainer sends out, and training voice is identified, and generates training voice
Text；
S202：The training action corresponding with training voice of Training Control terminal intake trainer's displaying, generates training action journey
Sequence；
S203：Training Control terminal will training speech text and training action program pairing typing speech text and operation program pair
Answer table；
S204：The speech recognition equipment of Training Control terminal receives the control voice that controller sends out, and control voice is known
Not, control voice text is generated；
S205：Control voice text is sent to robot body by Training Control terminal；
S206：The action selection device of robot body match selection in speech text table corresponding with action sequence goes out and controls
The identical and similar trained speech text of speech text, and then obtain training the corresponding training action program of speech text；
S207：Robot body performs the action of training action procedure stipulation.
9. a kind of voice control robot system for learning action, which is characterized in that system includes Training Control terminal, communication
Link and robot body, Training Control terminal include the first speech recognition equipment, action recognition device and information fuse device,
Robot body includes third speech recognition equipment, action selection device and action executive device, Training Control terminal and machine
Human body carries out information exchange by communication link, and speech recognition equipment receives the training voice that trainer sends out, to training language
Sound is identified, and generates training speech text, the training corresponding with training voice of action recognition device intake trainer's displaying
Action, generates training action program, and information fuse device will training speech text and training action program pairing typing voice text
The speech text table corresponding with operation program that training obtains is sent to robot sheet by this table corresponding with operation program, training terminal
Body, Training Control terminal receive the control voice that controller sends out, control voice are sent to robot body, robot body
Third speech recognition equipment control voice is identified, generate control voice text, action selection device in speech text
Go out identical with control voice text and similar trained speech text, and then instructed with match selection in the corresponding table of action sequence
Practice the corresponding training action program of speech text, the action executive device performs the action of training action procedure stipulation.
10. a kind of be associated with voice with action the method called, which is characterized in that includes the following steps：
S301：Training Control terminal receives the training voice that trainer sends out, and training voice is identified, and generates training voice
Text；
S302：The training action corresponding with training voice of Training Control terminal intake trainer's displaying, generates training action journey
Sequence；
S303：Training Control terminal will training speech text and training action program pairing typing speech text and operation program pair
Answer table；
S304：The control voice that the reception controller of Training Control terminal sends out, robot body is sent to by control voice；
S305：Control voice is identified in the third speech recognition equipment of robot body, generates control voice text,
S306：The action selection device of robot body match selection in speech text table corresponding with action sequence goes out and controls
The identical and similar trained speech text of speech text, and then obtain training the corresponding training action program of speech text；
S307：Robot body performs the action of training action procedure stipulation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810075101.3A CN108198559A (en) | 2018-01-26 | 2018-01-26 | A kind of voice control robot system for learning action |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810075101.3A CN108198559A (en) | 2018-01-26 | 2018-01-26 | A kind of voice control robot system for learning action |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108198559A true CN108198559A (en) | 2018-06-22 |
Family
ID=62590823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810075101.3A Pending CN108198559A (en) | 2018-01-26 | 2018-01-26 | A kind of voice control robot system for learning action |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108198559A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109693243A (en) * | 2018-12-05 | 2019-04-30 | 汕尾职业技术学院 | A kind of robot |
| CN110599823A (en) * | 2019-09-05 | 2019-12-20 | 北京科技大学 | Service robot teaching method based on fusion of teaching video and spoken voice |
| CN112880701A (en) * | 2021-01-12 | 2021-06-01 | 东莞市伟创动力科技有限公司 | Intersection broadcasting method, device, robot, system and medium for vehicle navigation |
| CN113449641A (en) * | 2021-06-29 | 2021-09-28 | 张家港江苏科技大学产业技术研究院 | Object recognition system based on cloud edge-end cooperation |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1380846A (en) * | 2000-03-31 | 2002-11-20 | 索尼公司 | Robot device, robot device action control method, external force detecting device and method |
| JP2015047262A (en) * | 2013-08-30 | 2015-03-16 | ブラザー工業株式会社 | Information processor, information processing system and program |
| CN104793542A (en) * | 2015-04-17 | 2015-07-22 | 深圳市优必选科技有限公司 | Robot system based on cellphone control |
| CN104965426A (en) * | 2015-06-24 | 2015-10-07 | 百度在线网络技术（北京）有限公司 | Intelligent robot control system, method and device based on artificial intelligence |
| CN205097196U (en) * | 2015-10-27 | 2016-03-23 | 众德迪克科技（北京）有限公司 | Robot with interactive function |
| CN105488044A (en) * | 2014-09-16 | 2016-04-13 | 华为技术有限公司 | Data processing method and device |
| CN106292424A (en) * | 2016-08-09 | 2017-01-04 | 北京光年无限科技有限公司 | Music data processing method and device for anthropomorphic robot |
| CN106547884A (en) * | 2016-11-03 | 2017-03-29 | 深圳量旌科技有限公司 | A kind of behavior pattern learning system of augmentor |
| CN106557165A (en) * | 2016-11-14 | 2017-04-05 | 北京智能管家科技有限公司 | The action simulation exchange method of smart machine and device and smart machine |
| CN106845628A (en) * | 2017-02-07 | 2017-06-13 | 珠海金萝卜智动科技有限公司 | The method and apparatus that robot generates new command by internet autonomous learning |
| CN107122040A (en) * | 2017-03-16 | 2017-09-01 | 杭州德宝威智能科技有限公司 | A kind of interactive system between artificial intelligence robot and intelligent display terminal |
| CN107291654A (en) * | 2016-03-31 | 2017-10-24 | 深圳光启合众科技有限公司 | The intelligent decision system and method for robot |
| CN107443396A (en) * | 2017-08-25 | 2017-12-08 | 魔咖智能科技（常州）有限公司 | A kind of intelligence for imitating human action in real time accompanies robot |
| CN107450367A (en) * | 2017-08-11 | 2017-12-08 | 上海思依暄机器人科技股份有限公司 | A kind of voice transparent transmission method, apparatus and robot |
-
2018
- 2018-01-26 CN CN201810075101.3A patent/CN108198559A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1380846A (en) * | 2000-03-31 | 2002-11-20 | 索尼公司 | Robot device, robot device action control method, external force detecting device and method |
| JP2015047262A (en) * | 2013-08-30 | 2015-03-16 | ブラザー工業株式会社 | Information processor, information processing system and program |
| CN105488044A (en) * | 2014-09-16 | 2016-04-13 | 华为技术有限公司 | Data processing method and device |
| CN104793542A (en) * | 2015-04-17 | 2015-07-22 | 深圳市优必选科技有限公司 | Robot system based on cellphone control |
| CN104965426A (en) * | 2015-06-24 | 2015-10-07 | 百度在线网络技术（北京）有限公司 | Intelligent robot control system, method and device based on artificial intelligence |
| CN205097196U (en) * | 2015-10-27 | 2016-03-23 | 众德迪克科技（北京）有限公司 | Robot with interactive function |
| CN107291654A (en) * | 2016-03-31 | 2017-10-24 | 深圳光启合众科技有限公司 | The intelligent decision system and method for robot |
| CN106292424A (en) * | 2016-08-09 | 2017-01-04 | 北京光年无限科技有限公司 | Music data processing method and device for anthropomorphic robot |
| CN106547884A (en) * | 2016-11-03 | 2017-03-29 | 深圳量旌科技有限公司 | A kind of behavior pattern learning system of augmentor |
| CN106557165A (en) * | 2016-11-14 | 2017-04-05 | 北京智能管家科技有限公司 | The action simulation exchange method of smart machine and device and smart machine |
| CN106845628A (en) * | 2017-02-07 | 2017-06-13 | 珠海金萝卜智动科技有限公司 | The method and apparatus that robot generates new command by internet autonomous learning |
| CN107122040A (en) * | 2017-03-16 | 2017-09-01 | 杭州德宝威智能科技有限公司 | A kind of interactive system between artificial intelligence robot and intelligent display terminal |
| CN107450367A (en) * | 2017-08-11 | 2017-12-08 | 上海思依暄机器人科技股份有限公司 | A kind of voice transparent transmission method, apparatus and robot |
| CN107443396A (en) * | 2017-08-25 | 2017-12-08 | 魔咖智能科技（常州）有限公司 | A kind of intelligence for imitating human action in real time accompanies robot |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109693243A (en) * | 2018-12-05 | 2019-04-30 | 汕尾职业技术学院 | A kind of robot |
| CN110599823A (en) * | 2019-09-05 | 2019-12-20 | 北京科技大学 | Service robot teaching method based on fusion of teaching video and spoken voice |
| CN110599823B (en) * | 2019-09-05 | 2021-08-13 | 北京科技大学 | Service robot teaching method based on fusion of teaching video and spoken voice |
| CN112880701A (en) * | 2021-01-12 | 2021-06-01 | 东莞市伟创动力科技有限公司 | Intersection broadcasting method, device, robot, system and medium for vehicle navigation |
| CN113449641A (en) * | 2021-06-29 | 2021-09-28 | 张家港江苏科技大学产业技术研究院 | Object recognition system based on cloud edge-end cooperation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108198559A (en) | A kind of voice control robot system for learning action | |
| CN106826838B (en) | Interaction bionic mechanical arm control method based on Kinect visual depth sensor | |
| JP6505748B2 (en) | Method for performing multi-mode conversation between humanoid robot and user, computer program implementing said method and humanoid robot | |
| CN105825268B (en) | The data processing method and system of object manipulator action learning | |
| Ding et al. | Kinect microphone array-based speech and speaker recognition for the exhibition control of humanoid robots | |
| US20050246063A1 (en) | Robot for participating in a joint performance with a human partner | |
| JP2011000681A (en) | Communication robot | |
| JP2006154926A (en) | Electronic equipment operation system using character display and electronic apparatuses | |
| CN108009490A (en) | A kind of determination methods of chat robots system based on identification mood and the system | |
| JP2004090109A (en) | Robot device and interactive method for robot device | |
| CN107443396A (en) | A kind of intelligence for imitating human action in real time accompanies robot | |
| JP7205148B2 (en) | ROBOT, CONTROL METHOD AND PROGRAM | |
| CN106446141B (en) | Interactive data processing method for intelligent robot system and robot system | |
| Salem et al. | Towards an integrated model of speech and gesture production for multi-modal robot behavior | |
| CN105489088A (en) | Voice bionic sign language bidirectional translation system | |
| JP2017204067A (en) | Sign language conversation support system | |
| CN109214347A (en) | A kind of sign language interpretation method across languages, device and mobile device | |
| CN110125932B (en) | Dialogue interaction method for robot, robot and readable storage medium | |
| CN109741744B (en) | AI robot conversation control method and system based on big data search | |
| CN113221681B (en) | Model training method, action gesture generating method, device, equipment and medium | |
| KR101933150B1 (en) | Command/information input system and method | |
| JPWO2019181144A1 (en) | Information processing equipment and information processing methods, and robot equipment | |
| CN108459707A (en) | It is a kind of using intelligent terminal identification maneuver and the system that controls robot | |
| JP4062591B2 (en) | Dialog processing apparatus and method, and robot apparatus | |
| JP2005202076A (en) | Uttering control device and method and robot apparatus |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180622 |