CN105390136B - Vehicle arrangement control device and method for user's adaptive type service - Google Patents
Vehicle arrangement control device and method for user's adaptive type service Download PDFInfo
- Publication number
- CN105390136B CN105390136B CN201510514457.9A CN201510514457A CN105390136B CN 105390136 B CN105390136 B CN 105390136B CN 201510514457 A CN201510514457 A CN 201510514457A CN 105390136 B CN105390136 B CN 105390136B
- Authority
- CN
- China
- Prior art keywords
- user
- adaptive type
- service
- information
- vehicle arrangement
- 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.)
- Active
Links
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 80
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000008451 emotion Effects 0.000 claims description 5
- 239000000284 extract Substances 0.000 claims description 5
- 230000006870 function Effects 0.000 description 35
- 238000012545 processing Methods 0.000 description 15
- 239000013598 vector Substances 0.000 description 9
- 230000036651 mood Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000001914 filtration Methods 0.000 description 5
- 238000004378 air conditioning Methods 0.000 description 4
- 239000000470 constituent Substances 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012880 independent component analysis Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R16/00—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for
- B60R16/02—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements
- B60R16/037—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements for occupant comfort, e.g. for automatic adjustment of appliances according to personal settings, e.g. seats, mirrors, steering wheel
- B60R16/0373—Voice control
-
- 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
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- User Interface Of Digital Computer (AREA)
- Navigation (AREA)
Abstract
The present invention provides a kind of voice pattern for driver being distinguished by speech recognition and the driver is analyzed by much information, guides the vehicle arrangement control device and method for the service of user's adaptive type of best-of-breed functionality from vehicle to driver first.Vehicle arrangement control device for the service of user's adaptive type of the invention includes: characteristic information generating unit, and the characteristic information of user is generated according to the voice messaging of user；Voice messaging analysis unit obtains meaning information by parsing voice messaging；Adaptive type services determining section, and the adaptive type service to user is determined according to characteristic information and meaning information；And vehicle arrangement control unit, it controls the control object equipment including vehicle arrangement and to execute adaptive type service.The present invention can be realized natural speech recognition system, use common function convenient for driver.
Description
Technical field
The present invention relates to the device and methods of control vehicle arrangement more particularly to a kind of control vehicle arrangement to execute use
The vehicle arrangement control device and method for the service of user's adaptive type of family adaptive type service.
Background technique
Speech recognition is that the acoustic information from voice extracts phoneme i.e. language message and makes the one of machine recognition and reaction
Serial procedures.
Although generally believing that with voice dialogue be most natural, the simplest side in the information exchange medium of the mankind and machine
Method, but human speech must be converted to the code that machine is capable of handling can use voice and machine to talk with.Speech recognition is just
It is this process for being converted into code.
It is applicable in the speech recognition technology developed in recent years on vehicle at present, therefore only needs the voice command of driver
It can be driven simple convenient means, such as lifting window, starting and stop wiper, open air-conditioning, open or close preceding photograph
Lamp etc..
Illustrate current vehicle audio recognition methods below.
Current vehicle audio recognition methods includes that driver is connect when issuing equipment start command with voice by microphone
The step of receiving driver's voice, the step of being pre-processed analog signal for digital signal by filtering and analog-to-digital conversion, by mentioning
The step of taking eigen vector and classification voice pattern voice command recognition, and according to the voice command drive control object of identification
The step of device.
Current speech recognition can use speech engine to identify a small amount of vocabulary even large capacity vocabulary, only by pressing
I.e. logical (Push-to-Talk；PTT) speech identifying function activates when key.
But the currently used ability that is speech recognition system in the case where words person issues to speech recognition system and orders according to
The order constitutes the one way system of corresponding scene, can not two-way exchange.
KR published patent the 2014-0051630th discloses one kind and passes through the audio-visual navigation of speech recognition controlled vehicle
The method of system.But this method is also to provide speech identifying function by remote control speech recognition key, therefore can not solve above-mentioned ask
Topic.
Summary of the invention
Technical problem
To solve the above problems, distinguishing driver by speech recognition the purpose of the present invention is to provide one kind and passing through
Much information analyzes the voice pattern of the driver, is used for user's adaptive type from vehicle to driver's guidance best-of-breed functionality first
The vehicle arrangement control device and method of service.
The purpose of the present invention is not limited to above-mentioned purpose, and those skilled in the art can be defined by following record
Understand unmentioned other purposes.
Technical solution
To reach above-mentioned purpose, the present invention provides a kind of vehicle arrangement control device for user's adaptive type service, packet
Include: characteristic information generating unit generates the characteristic information of the user according to the voice messaging of user；Voice messaging analysis unit,
It obtains meaning information by parsing the voice messaging；Adaptive type service determining section, according to the characteristic information with it is described
Meaning information determines the adaptive type service to the user；And vehicle arrangement control unit, it controls including vehicle arrangement
Control object equipment to execute the adaptive type service.
Preferably, the characteristic information generating unit from the voice messaging extract formant (formant) value, frequency values,
At least one in speech energy value and linear predictive coding (linear prediction coding, hereinafter referred to as ' LPC ') value
A value, and the characteristic information is generated according at least one described value in real time.
Preferably, the characteristic information generating unit generates the age letter of the gender information of the user, the user in real time
At least one of the emotion information of breath and user information is as the characteristic information.
Preferably, the vehicle arrangement control device further include: voice messaging selector is receiving at least two languages
A voice messaging is selected when message ceases from multiple voice messagings.
Preferably, the voice messaging selector according to the size of voice messaging, input voice messaging be stored in advance
Voice messaging between comparison result, the user position and multilayer perceptron (multilayer perceptron) in
At least one select one voice messaging.
Preferably, the vehicle arrangement control device further include: voice messaging input unit is received from each seat of vehicle
The voice messaging, the vehicle arrangement control unit control vehicle arrangement to execute the adaptive type respectively by each seat
Service.
Preferably, the voice messaging input unit includes the directional microphone for being set to each seat
(directional microphone)。
Preferably, the vehicle arrangement control device further include: adaptive type service execution judging part, according to the user
The information of input judges whether to execute the adaptive type service；And alternative service determining section, it is not execute in judging result
When the adaptive type services, the alternative service for substituting the adaptive type service is determined according to the information that the user inputs.
Preferably, the vehicle arrangement of the vehicle arrangement control unit control is audio-visual navigation (Audio Video
Navigation；AVN) system.
Also, the present invention provides a kind of vehicle arrangement control method for user's adaptive type service, comprising: according to user
Voice messaging the step of generating the characteristic information of the user；The step of meaning information is obtained by parsing the voice messaging
Suddenly；The step of servicing the adaptive type of the user is determined according to the characteristic information and the meaning information；And control packet
It includes the control object equipment including vehicle arrangement and makes the step of executing adaptive type service.
Preferably, the step of generation extracts formant (formant) value, frequency values, voice from the voice messaging
At least one of energy value and linear predictive coding (linear prediction coding, hereinafter referred to as ' LPC ') value value,
And the characteristic information is generated according at least one described value in real time.
Preferably, the step of generation generate in real time the gender information of the user, the user age information and
At least one of the emotion information of user information is as the characteristic information.
Preferably, before the step of generation further include: when receiving at least two voice messagings from multiple described
A step of voice messaging is selected in voice messaging.
Preferably, the step of selection is according to the size of voice messaging, the voice messaging of input and pre-stored language
In the position of comparison result, the user between message breath and multilayer perceptron (multilayer perceptron) extremely
A kind of few one voice messaging of selection.
Preferably, before the step of selection further include: the step of receiving the voice messaging from each seat of vehicle,
The step of control is specifically to control vehicle arrangement to execute the adaptive type service respectively by each seat.
Preferably, the received step utilizes the directional microphone for being set to each seat.
Preferably, determination is also wrapped between the step of adaptive type service and the step of control of the user
It includes: the step of executing adaptive type service is judged whether according to the information that the user inputs；And in judging result for not
When executing adaptive type service, determine that the substitution for substituting the adaptive type service takes according to the information of user input
The step of business.
Preferably, the vehicle arrangement of the step control of control is image and sound guidance system (Audio Video
Navigation；AVN).
Technical effect
The present invention is distinguished driver by speech recognition and analyzes the voice pattern of the driver by much information, first
Best-of-breed functionality first is guided from vehicle to driver, to have the following beneficial effects:
First, it can comply with and gradually change into two-way communication from one way system to the trend that two-way exchange mode changes, from
And it can be realized natural speech recognition system.
Second, system uses common function according to driver's correspondingly recommendation function, therefore convenient for driver.
Detailed description of the invention
Fig. 1 is the concept map for showing vehicle Speaker identification system according to an embodiment of the invention；
Fig. 2 is the flow chart for showing the first embodiment of working method of vehicle Speaker identification system；
Fig. 3 is the flow chart for showing the second embodiment of working method of vehicle Speaker identification system；
Fig. 4 is to summarize the display vehicle arrangement control dress according to the preferred embodiment of the invention for the service of user's adaptive type
The block diagram set；
Fig. 5 is to summarize to show the vehicle arrangement controlling party according to the preferred embodiment of the invention for the service of user's adaptive type
The flow chart of method.
Specific embodiment
The preferred embodiments of the present invention are described in detail referring to the drawings.Firstly, it is necessary to it is to be noted that in the structure to each figure
In terms of at element addition appended drawing reference, added as far as possible being appeared in even if identical constituent element on different attached drawings identical
Appended drawing reference.And if think may be to this hair to illustrating for related known structure or function for judgement in explaining the present invention
Bright theme causes to obscure, then omits related detailed description.In addition, will be described below the preferred embodiment of the present invention, but this hair
Bright technical solution is simultaneously not restricted or limited to this, and person of ordinary skill in the field can do various deformation implementation.
The present invention is characterized in that according to the technology trends changed to two-way communication mode, it is this using speech recognition
Speaker identification function distinguishing driver and the voice pattern for analyzing the driver recommend most suitable function first for driver,
Artificial intelligence trend can be complied with.
Fig. 1 is the concept map for showing vehicle Speaker identification system according to an embodiment of the invention.
As shown in Figure 1, the present invention is the driver for the voice pattern for passing through speech recognition words person and analyzing each driver
Friendly vehicle interior system.
Private car is generally shared by more people, and the present invention stores the language of the characteristics of speech sounds of each driver and each driver of analysis
Sound pattern.The voice pattern of driver can be search ground, nearest call catalog, audio-frequency function etc. recently.After driver rides
Distinguish who driver words person is, confirmation is suitble to the function of driver's voice pattern on vehicle when speaking by microphone,
Guarantee faster to be easier access to the common function of driver.
The function of input unit 110 is the voice command for receiving driver.Input unit 110 can be microphone.
Identification part 120 is identified by the voice signal of the input of input unit 110.Identification part 120 turns text (Speech by sound
To Text, hereinafter referred to as ' STT ') receive is any voice for operation judgement.
Analysis portion 130 distinguishes words person, gender and age bracket is analyzed by the database (Database) of study, by finding out
Formant (formant) value identifies everyone characteristic.
Analysis portion 130 passes through formant (formant) value of words person's voice, basic frequency value, speech energy value, linear pre-
The gender of the users of real time discriminatings statistics such as survey coding (linear prediction coding, hereinafter referred to as ' LPC ') value/
Age/mood/state etc..
The characteristic for each driver that the storage of storage unit 160 is collected by analysis portion 130.At this point, the storage identification of storage unit 160
Portion 120 turns the result of text to the voice command sound that driver says.
Processing unit 140 goes to next scene using the DB plan stored by driver, and whether inquiry driver will go to
The corresponding scene that vehicle is recommended, so that vehicle guides driver's common function first.
For example, the purpose that the building of processing unit 140 recommends driver A often to go in special time period from vehicle to driver first
Ground, the radio broadcasting often listened, digital media broadcast (Digital Media Broadcasting；DMB) the scene of channel etc.,
Or it is prioritized to the music that driver often listens in the music for playing vehicle storage, or grasp the age of driver and passing through clothes
Business device plays age bracket people when listening to music likes the music listened.
Processing unit 140 searches for (searching) data, when ensuring the function needed for handling user, especially in user
The convenient function of adaptive type for the user personality that most suitable analysis portion 130 is analyzed in real time is capable of providing in the case where not specified precise information
It can information.
It is suitable that processing unit 140 provides selection music, recommendation radio broadcasting, recommending digital multimedia broadcasting, searching facility etc.
Distribution type facilitating functions information.
Inquire whether driver will make the system in vehicle by the specific letter according to driver by loudspeaker in output section 150
Cease the scene work obtained.
Processing unit 140 is transmitted to user and handles the result obtained in output section 150.
Fig. 2 is the flow chart for showing the first embodiment of working method of vehicle Speaker identification system.
When user says " broadcast " by microphone (MIC), in step S210, input unit 110 obtains the voice signal.
Then in step S220, identification part 120 executes speech identifying function, is converted to broadcasting command extensively by STT
Broadcast text.
Then in step S230, analysis portion 130 using study DB and resonance peak, basic frequency value, speech energy value,
LPC value etc. analyzes the voice command of driver and storage.
Then in step S240, processing unit 140 confirms current time, and according to the recognition result of identification part 120, analysis
The analysis result in portion 130, the information for being stored in storage unit 160 etc. generate waveform (wave) file, to guide time driver
The TBS that the frequency listened to is FM95.1 is broadcasted.
Then in step s 250, output section 150 with loudspeaker export " listening to FM95.1 TBS Traffic Announcement? ".
When receiving the information for agreeing to the content exported by loudspeaker from user, in step S260, vehicle it is audio-visual
(Audio Video Navigation, hereinafter referred to as ' the AVN ') system of navigation exports FM95.1 TBS Traffic Announcement.
On the contrary, if receiving the information for disagreeing the content exported by loudspeaker from user, in step S270, place
Reason portion 140 exports " please say channel " by the loudspeaker of output section 150.Then in step S270, driver passes through input
When portion 110 orders required frequency, processing unit 140 exports the broadcast of corresponding frequencies.
Speech recognition system differentiate after driver by driver characteristics recommend and in the case that driver refuses, as above to driving
The person of sailing inquires required function and by the function operation.
Speech recognition system has the case where erroneous judgement driver, therefore adds driver's identification function in AVN function and open
Open/close (ON/OFF) function, error when driver be set as close (OFF) make speech recognition system nonrecognition words person.
Speech recognition system and dependent of the invention carries out speech recognition, but by real with other module shared informations
Vehicular system now more friendly to driver.
Fig. 3 is the flow chart for showing the second embodiment of working method of vehicle Speaker identification system.
When saying broadcast or DMB with voice in step s310, analysis portion 130 is extracted by the step S320 of speech analysis
The characteristic of user.Then, processing unit 140 selects the broadcast for being most suitable for user personality.
With if, in the state connecting with server, analysis portion 130 is extracted by voice and is used when phonetic search music
The characteristic at family is simultaneously selected from the music list that different sexes/age/state crowd that server provides likes.
Even if not connecting with server, analysis portion 130 can also will be about the general characteristic of the music file (property liked
Not, age bracket, mood) it is stored in storage unit 160, most suitable music is thought according to the broadcasting of the voice status of user.
When passing through navigating search facility, if processing unit 140 searches for such as periphery dining room by the characteristics of speech sounds of user,
The dining room that each gender/age bracket is liked then is shown at first.In the case where using Xi Li (SIRI), when with phonetic search periphery dining room
Provide a user by YELP search as a result, the other informations such as gender/age bracket is also utilized to search for when searching for periphery dining room
The dining room that more approximate people like with the user.
If there is network connection, processing unit 140 carries out web search using voice command and user personality information and will letter
Breath is supplied to user.Congee shop is searched for when learning that health state of user is bad by speech analysis first, if when search periphery hospital
Think that user's illnesses for flu, then first look for internal medicine by analyzing user speech.
And when with phonetic search concert or concert, processing unit 140 passes through web search ring approximate with the user
The concert or concert that the people in border like.
Content described above is summarized below.
In step s310, input unit 110 obtains corresponding voice messaging when user gives orders or instructions.For example, user says " broadcast "
Or input unit 110 obtains the voice messaging when " DMB ".
Then, analysis portion 130 analyzes voice messaging and the judgement property in step S330~S350 in real time in step s 320
Not/age/state etc..
Analysis portion 130 judges that words person is male or women by voice messaging in step S330.Then, analysis portion
130 judge words person's age bracket (for example, two teens, three teens, four teens etc.) in step S340.Then, analysis portion
130 judge that mood/state of words person is good bad in step S350.
Then in step S360, processing unit 140 searches for the electricity for being most suitable for user according to the analysis result of analysis portion 130
Platform.For example, processing unit 140, which searches for two good teens males of mood/state, likes the radio station listened, or search mood/state is not
Three good teens males like the radio station listened, or search for the good teens women of mood/state and like the radio station listened, or search
Two teens women of mood/be not in good state like the radio station listened.
Then in step S370, output section 150 (is executed corresponding by the processing result of loudspeaker output processing part 140
Service).
The present invention is not limited to the drivers in vehicle, and when not only considering all speech recognitions when searching for information
Speech content, while also considering words person's characteristic.
Different from the prior art, the present invention is grasped the User Status that may change and is accordingly provided by analysis voice in real time
User's adaptive type information.
Illustrate that speech recognition of the invention is applicable in back-propagation algorithm (Back Propagation Algorithm) below
Method.
According to general noise filtering methods, opens speech recognition microphone and issued after the predetermined time and known for voice
The signal that speech recognition advances into microphone is judged as noise in vehicle and the noise in trap signal by other voice.
Although there is the directional microphone being arranged towards driver, due to the time of short duration before voice is given orders or instructions in vehicle
In the case that the signal of input is judged as noise, therefore speech recognition gives orders or instructions time point other seats are also given orders or instructions in addition to a driver
Voice is mutually mixed, therefore phonetic recognization rate declines.
Therefore, the present invention four seating area being set to point to property microphones in vehicle respectively, with driver region
On the basis of the input signal of microphone, by other regions, microphone signal is determined as noise and filters.During handling signal
The characteristic of the driver in real time discriminating driver region, to ensure that multimedia equipment provides the information of suitable driver.
This is described in further details below, driver's seat is defined as a-quadrant in explanation below, passenger seat is defined
For B area, the rear side seat of the back seat of driver's seat and passenger seat is respectively defined as the region C and the region D.
It the microphone in the region A, B, C, D while being opened when driver starts speech identifying function, passes through microphone and receive four
The voice signal in a region.When the vehicle noise for not being human speech is input to the microphone in four regions, its input value is almost
It is identical, therefore by a-quadrant filter vehicle noise figure.Also, analyze the voice in four regions.The expression in four regions is analyzed first
The speech vector value of gender, on the basis of a-quadrant, when from B, C, D extracted region to the vector value indicated with a-quadrant different sexes
When, a-quadrant filtering corresponds to the signal of the vector value.Age, mood/state are analyzed in the same way after having analyzed gender
Deng.
Although maximum voice signal is the voice signal of driver in a-quadrant, when there are also the voices in the region B, C, D
In the case where signal, the sound of a-quadrant driver can not be only extracted, therefore use this method.
Correlation (CORRELATION), independent component analysis (Independent Component can be passed through at this time
Analysis；ICA) other algorithms except technology, Wave beam forming (BEAM FORMING) technology differentiate that signal is independent or has
There is approximation.
It can be filtered by four microphones to grasp the individual characteristic of words person, can use grasp individual characteristic and obtain
The information filtering noise arrived, so as to improve discrimination.
Multilayer perceptron (multilayer perceptron) is illustrated below.
Existing theoretical voice (judgement when receiving voice for identification of perceptron (perceptron) relevant to voice
The content of voice) or differentiate people emotion.
Multilayer perceptron (multilayer perceptron) is that have among more than one between input layer and output layer
The neural network of layer.Network is that there is no the connection in each layer and output layers by input layer, concealment layer, the connection of output layer direction
To feedforward (Feedforward) network of input layer being directly connected to.
Vehicle generally has there are four seat, and the user of speech recognition system is usually driver in vehicle, makes in driver
During with speech recognition system, the voice overlapping of more people when the passenger when other seats gives orders or instructions, therefore speech recognition system
It can not identify the order of driver.The speech recognition system being commonly used is to be configured without voice before speech recognition section
Section and the section is identified as noise and crosses the structure of noise filtering in voice input interval.
The present invention is that the characteristic of voice is extracted using perceptron theory to identify words person's characteristic, according to the data in real time to words
Person provides the technology of suitable information.By perceptron, 1. adaptive type information can be provided by the characteristic of each words person, 2. can known
Not words person position and function needed for person if the position is provided.Below to 1. and being 2. described in further details.
1. providing adaptive type information according to words person's characteristic
In the case where using multilayer perceptron composition system, driver can be extracted the voice of more people is superimposed
Voice.This method is not limited to driver, can be also used for identifying other people.For example, the characteristics of speech sounds of a-quadrant is only extracted,
Ignore the voice signal in remaining region B, C, D.
The major premise of perceptron is to be formed with to utilize backpropagation (BACK PROPAGATION) skill previously according to a large amount of DB
The state of the algorithm of art training.
It models, such as is extracted by two teens of analysis and a large amount of voices of the good Soul women of state special about perceptron
Property (formant, basic frequency, energy value, LPC value etc.) and by input terminal input, output (OUTPUT) object be more than 20
Year and when the good Soul women of state, perceptron inside configuration determines suitable by backpropagation (BACK PROPAGATION) process
When weighting (WEIGHT) value.As above in the case where being trained to the people of multifrequency nature, no matter inputting any voice can
Characteristic is found in trained structure.LPC value is linear prediction symbolism value, is the voice symbol based on mankind's generation model
One of number change mode, the vectors with 26 dimensions.
When inputting formant, the basic frequency, 26 dimensional vector value of LPC model of a large amount of voices of special object, by anti-
The specified operation of weighted value appropriate (two teens and the good Soul women of state, three are repeated to multiple objects to expansion process
Teens and the road the Qing Shang area male ... being not in good state).
By above-mentioned training process, no matter receiving any voice can be by being input to the eigen vector to the voice
The perceptron structure of modeling learns the characteristic of words person.
Benchmark is selected as seat with putting call through immediately after connection (Push-to-Talk, hereinafter referred to as ' PTT ').If there are four PTT key,
It is the voice for needing to analyze by the phonetic decision that the microphone for receiving the position of corresponding PTT receives then according to its position, it will
Remaining is judged as noise and filters.Identify by filtered voice and provides optimal information, such as words person for words person
Command lookup periphery dining room is issued to media product, then searches out the periphery dining room of suitable words person's characteristic first.
Following characteristic can be exported by arranging content described above.
Firstly, differentiating the position PTT and extracting the vector for corresponding to each voice signal characteristic.
Then, the eigen vector of four kinds of signals is inputted to multilayer perceptron structure.
Then, the characteristic of each voice signal is extracted.
Then, when having other characteristics different from reference speech (A), judge other characteristics in A microphone signal
Value is judged as noise and filters.
Then, the data execution speech recognition obtained using a-quadrant voice is only extracted, judges the meaning of voice.
Then, the information of the order of most suitable a-quadrant words person is provided.
2. function needed for person if identifying words person position and providing the position
Benchmark is selected as seat with putting call through immediately after connection (Push-to-Talk, hereinafter referred to as ' PTT ').If there are four PTT key,
It is the voice for needing to analyze by the phonetic decision that the microphone for receiving the position of corresponding PTT receives then according to its position, it will
Remaining is judged as noise and filters.It, can when the passenger for being sitting in the region D issues the order about air-conditioner temperature by taking air-conditioning as an example
Only to change the air-conditioning gear of the air-conditioning device in the region D according to order.
It is explained above and distinguishes that the speech recognition system of words person provides the sheet of best information to driver by speech recognition
One embodiment of invention.Illustrate the preferred embodiment of the present invention that can be reasoned out from these embodiments below.
Fig. 4 is to summarize the display vehicle arrangement control dress according to the preferred embodiment of the invention for the service of user's adaptive type
The block diagram set.
Referring to Fig. 4, the vehicle arrangement control device 400 for user's adaptive type service include characteristic information generating unit 410,
Voice messaging analysis unit 420, adaptive type service determining section 430, vehicle arrangement control unit 440, power supply unit 450 and main control unit
460。
The function of power supply unit 450 is to provide power supply to each composition for constituting vehicle arrangement control device 400.Main control unit
460 function is all working respectively constituted that control constitutes vehicle arrangement control device 400.In view of can be by vehicle arrangement
Even control device 400 is arranged in AVN system, therefore the present embodiment does not have power supply unit 450 and main control unit 460.
The function of characteristic information generating unit 410 is the characteristic information that user is generated according to the voice messaging of user.
Characteristic information generating unit 410 from voice messaging extract formant (formant) value, frequency values, speech energy value and
At least one value in LPC value, can be according to the real-time formation characteristic information of at least one value.
The characteristic information for the user that characteristic information generating unit 410 generates in real time can be the gender information of user, user
In age information and the emotion information of user to planting a kind of information.
Characteristic information generating unit 410 corresponds to the concept of the analysis portion 130 in Fig. 1.
The function of voice messaging analysis unit 420 is to obtain meaning information by parsing the voice messaging of user.
The function of adaptive type service determining section 430 is the characteristic information and voice generated according to characteristic information generating unit 410
The meaning information that information analyzing section 420 obtains determines the adaptive type service to user.
The function of vehicle arrangement control unit 440 is that control object equipment of the control including vehicle arrangement makes execution suitable
Distribution type services the adaptive type service that determining section 430 determines.
The vehicle arrangement that vehicle arrangement control unit 440 controls can be audio-visual navigation (Audio Video Navigation；
AVN) system.
Vehicle arrangement control device 400 can also include voice messaging selector (not shown).
The function of voice messaging selector is selected from these voice messagings when receiving at least two voice messagings
One voice messaging.
Voice messaging selector can be believed according to the size of voice messaging, the voice messaging of input and pre-stored voice
The choosing of at least one of the position of comparison result, user between breath and multilayer perceptron (multilayer perceptron)
Select a voice messaging.
In the case where multilayer perceptron, voice messaging selector can select voice messaging in the following order.
Firstly, each area voice of the vehicle interior received by microphone is input to housebroken perceptron model
To extract driver information.
Then, if other regions have the voice of other characteristics of the characteristic in the region being different from the basis of driver,
The signal in the voice signal for the microphone that then will enter into driver region is determined as noise and filters.
Then, according to the voice for being input to all positions be separately input to the result that perceptron model obtains be filtered with
Obtain voice messaging.
Vehicle arrangement control device 400 can also include voice messaging input unit (not shown).
The function of voice messaging input unit is to receive at least one voice messaging.In particular, the function of voice messaging input unit
It is to receive voice messaging at each seat of vehicle.Voice messaging input unit can be with directional microphone prominent form in each seat
Position.
In this case, vehicle arrangement control unit 440 can control vehicle arrangement and to execute adaptive type clothes by each seat
Business.
Vehicle arrangement control device 400 can also include that adaptive type service execution judging part (not shown) and alternative service are true
Determine portion's (not shown).
The function of adaptive type service execution judging part is to be determined whether to execute adaptive type clothes according to the information of user's input
Business.
The function of alternative service determining section is when judging result is not execute adaptive type service, according to the letter of user's input
Breath determines the alternative service for substituting adaptive type service.
In this case, vehicle arrangement control unit 440 makes vehicle arrangement execute alternative service by control.
Working method of the explanation for the vehicle arrangement control device 400 of user's adaptive type service below.
Fig. 5 is to summarize to show the vehicle arrangement controlling party according to the preferred embodiment of the invention for the service of user's adaptive type
The flow chart of method.It is illustrated referring to Fig. 5.
First in step S510, voice messaging input unit receives the voice messaging of user from each seat of vehicle.
Then in step S530, characteristic information generating unit 410 is believed according to the characteristic that the voice messaging of user generates user
Breath.And in step S520, voice messaging analysis unit 420 obtains meaning information by the voice messaging of parsing user.Step
S530 can be performed simultaneously with step S520, but can also be executed before step S520 or step S520 after execute.
Then in step S540, adaptive type services determining section 430 to be believed according to the characteristic that characteristic information generating unit 410 generates
Breath determines the adaptive type service to user with the meaning information that voice messaging analysis unit 420 obtains.
Then in step S550, vehicle arrangement control unit 440 controls the control object equipment including vehicle arrangement
So that executing the adaptive type service that adaptive type service determining section 430 determines.
In addition after step S510, voice messaging selector can be from these languages when receiving at least two voice messagings
A voice messaging is selected in message breath.The above-mentioned steps of voice messaging selector can be between step S510 and step S520
It is executed between execution or step S510 and step S530.
In addition between step S540 and step S550, adaptive type service execution judging part is sentenced according to the information that user inputs
It is disconnected whether to execute adaptive type service.Then, alternative service determining section judging result be do not execute adaptive type service when, according to
The information of user's input determines the alternative service for substituting adaptive type service.
All constituent elements for describing the composition embodiment of the present invention above are combined into one or combine work, but the present invention
It is not limited to these embodiments.I.e. within the scope of the purpose of the present invention, all constituent elements can be by more than one selectivity
Ground combination work.Also, its all constituent element can occur in the form of an independent hardware respectively, but the also property of can choose
Ground combines part or all of each component, by have for execute part that one or more hardware combinations are realized or
The computer program of the program module of repertoire is realized.Also, this computer program can be stored in USB storage, CD
The computer-readable recording medium such as disk, flash disk (Flash Memory) (Computer Readable Media), by counting
Calculate it is machine-readable take and execute, to realize the embodiment of the present invention.The recording medium of computer program may include magnetic recording medium, light
Recording medium, carrier wave (Carrier Wave) medium etc..
Also, include the case where all terms including technology or scientific words in illustrating without separately defining following table
Show and is generally understood the identical meaning with general technical staff of the technical field of the invention.Usually used dictionary definition
Term should be interpreted that the meaning consistent with the meaning of the context of the relevant technologies must not if undefined in the present invention
It is construed to ideal or excessively formality the meaning.
The above description is only an example illustrates technical solution of the present invention, and those of ordinary skill in the art are not departing from this
A variety of amendments, change and replacement can be carried out in the range of inventive nature characteristic.Therefore, disclosed embodiment of this invention and attached drawing
And non-limiting technical solution of the present invention, but for illustrating, technical solution of the present invention be not limited to these embodiments and
Attached drawing.Protection scope of the present invention is determined by technical solution, is both contained in the present invention with all technical solutions of its equivalency range
Technical solution in.
Claims (11)
1. a kind of vehicle arrangement control device for user's adaptive type service characterized by comprising
Characteristic information generating unit generates the characteristic information of the user according to the voice messaging of user；
Voice messaging analysis unit obtains meaning information by parsing the voice messaging；
Adaptive type services determining section, is determined according to the characteristic information and the meaning information and is taken to the adaptive type of the user
Business；
Vehicle arrangement control unit controls the control object equipment including vehicle arrangement and to execute the adaptive type clothes
Business；
Adaptive type service execution judging part judges whether to execute the adaptive type service according to the information that the user inputs；
And
Alternative service determining section, when judging result is not execute adaptive type service, according to the letter of user input
Breath determines the alternative service for substituting the adaptive type service.
2. the vehicle arrangement control device according to claim 1 for the service of user's adaptive type, it is characterised in that:
The characteristic information generating unit extracts resonance peak, frequency values, speech energy value and linear prediction from the voice messaging
At least one of encoded radio value, and the characteristic information is generated according at least one described value in real time.
3. the vehicle arrangement control device according to claim 1 for the service of user's adaptive type, it is characterised in that:
The characteristic information generating unit generates the gender information of the user, the age information of the user and the user in real time
At least one of emotion information information as the characteristic information.
4. the vehicle arrangement control device according to claim 1 for the service of user's adaptive type, which is characterized in that also wrap
It includes:
Voice messaging selector selects a language when receiving at least two voice messagings from multiple voice messagings
Message breath.
5. the vehicle arrangement control device according to claim 4 for the service of user's adaptive type, it is characterised in that:
The voice messaging selector according to the size of voice messaging, the voice messaging of input and pre-stored voice messaging it
Between comparison result, the user position and the one voice messaging of at least one of multilayer perceptron selection.
6. the vehicle arrangement control device according to claim 1 for the service of user's adaptive type, which is characterized in that also wrap
It includes:
Voice messaging input unit receives the voice messaging from each seat of vehicle,
The vehicle arrangement control unit control vehicle arrangement to execute the adaptive type service respectively by each seat.
7. the vehicle arrangement control device according to claim 6 for the service of user's adaptive type, it is characterised in that:
The voice messaging input unit includes the directional microphone for being set to each seat.
8. the vehicle arrangement control device according to claim 1 for the service of user's adaptive type, it is characterised in that:
The vehicle arrangement of the vehicle arrangement control unit control is image and sound guidance system.
9. a kind of vehicle arrangement control method for user's adaptive type service characterized by comprising
The step of generating the characteristic information of the user according to the voice messaging of user；
By parsing the step of voice messaging obtains meaning information；
The step of servicing the adaptive type of the user is determined according to the characteristic information and the meaning information；
Control includes the steps that the control object equipment including vehicle arrangement to execute the adaptive type service；
Judge whether the step of executing adaptive type service according to the information that the user inputs；And
When judging result is not execute adaptive type service, determined according to the information of user input described for substituting
The step of alternative service of adaptive type service.
10. the vehicle arrangement control method according to claim 9 for the service of user's adaptive type, it is characterised in that:
The step generated generates the sense of the gender information of the user, the age information of the user and the user in real time
At least one of feelings information information is as the characteristic information.
11. the vehicle arrangement control method according to claim 9 for the service of user's adaptive type, which is characterized in that also
Include:
The step of receiving the voice messaging from each seat of vehicle,
The step of control is specifically to control vehicle arrangement to execute the adaptive type service respectively by each seat.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR10-2014-0116184 | 2014-09-02 | ||
| KR1020140116184A KR102249392B1 (en) | 2014-09-02 | 2014-09-02 | Apparatus and method for controlling device of vehicle for user customized service |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105390136A CN105390136A (en) | 2016-03-09 |
| CN105390136B true CN105390136B (en) | 2019-05-21 |
Family
ID=55422356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510514457.9A Active CN105390136B (en) | 2014-09-02 | 2015-08-20 | Vehicle arrangement control device and method for user's adaptive type service |
Country Status (2)
| Country | Link |
|---|---|
| KR (1) | KR102249392B1 (en) |
| CN (1) | CN105390136B (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105976815A (en) * | 2016-04-22 | 2016-09-28 | 乐视控股（北京）有限公司 | Vehicle voice recognition method and vehicle voice recognition device |
| KR102497299B1 (en) | 2016-06-29 | 2023-02-08 | 삼성전자주식회사 | Electronic apparatus and method for controlling the electronic apparatus |
| CN107590120A (en) * | 2016-07-07 | 2018-01-16 | 深圳狗尾草智能科技有限公司 | Artificial intelligence process method and device |
| KR102568143B1 (en) * | 2016-09-23 | 2023-08-18 | 주식회사 케이티 | Method and device for providing customized service mode |
| KR102329888B1 (en) | 2017-01-09 | 2021-11-23 | 현대자동차주식회사 | Speech recognition apparatus, vehicle having the same and controlling method of speech recognition apparatus |
| KR101883301B1 (en) | 2017-01-11 | 2018-07-30 | (주)파워보이스 | Method for Providing Personalized Voice Recognition Service Using Artificial Intellignent Speaker Recognizing Method, and Service Providing Server Used Therein |
| KR20180106196A (en) | 2017-03-17 | 2018-10-01 | 현대자동차주식회사 | Apparatus and method for optimizing navigation performance |
| KR102437833B1 (en) | 2017-06-13 | 2022-08-31 | 현대자동차주식회사 | Apparatus for selecting at least one task based on voice command, a vehicle including the same and a method thereof |
| KR102707293B1 (en) * | 2018-03-29 | 2024-09-20 | 삼성전자주식회사 | The apparatus for processing user voice input |
| KR102562227B1 (en) * | 2018-06-12 | 2023-08-02 | 현대자동차주식회사 | Dialogue system, Vehicle and method for controlling the vehicle |
| CN110503947B (en) * | 2018-05-17 | 2024-06-18 | 现代自动车株式会社 | Dialogue system, vehicle including the same, and dialogue processing method |
| KR102114843B1 (en) * | 2018-11-26 | 2020-05-26 | 한국생산기술연구원 | Interactive Module Service System and Method for Custom Assembly Vehicle Industry based on emotion |
| KR102275873B1 (en) * | 2018-12-18 | 2021-07-12 | 한국전자기술연구원 | Apparatus and method for speaker recognition |
| KR102235091B1 (en) * | 2019-02-21 | 2021-04-02 | 주식회사 에스디아이컴퍼니 | Fabrics |
| JP7211856B2 (en) * | 2019-03-11 | 2023-01-24 | 本田技研工業株式会社 | AGENT DEVICE, AGENT SYSTEM, SERVER DEVICE, CONTROL METHOD FOR AGENT DEVICE, AND PROGRAM |
| JP2020154013A (en) * | 2019-03-18 | 2020-09-24 | 株式会社Ｓｕｂａｒｕ | Caution evocation device for vehicle, caution evocation method for vehicle and program |
| CN114049894A (en) * | 2022-01-11 | 2022-02-15 | 广州小鹏汽车科技有限公司 | Voice interaction method and device, vehicle and storage medium |
| CN115170239A (en) * | 2022-07-14 | 2022-10-11 | 艾象科技(深圳)股份有限公司 | Commodity customization service system and commodity customization service method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102800315A (en) * | 2012-07-13 | 2012-11-28 | 上海博泰悦臻电子设备制造有限公司 | Vehicle-mounted voice control method and system |
| CN102802114A (en) * | 2012-06-20 | 2012-11-28 | 北京语言大学 | Method and system for screening seat by using voices |
| CN103137125A (en) * | 2011-11-30 | 2013-06-05 | 北京德信互动网络技术有限公司 | Intelligent electronic device based on voice control and voice control method |
| CN103137043A (en) * | 2011-11-23 | 2013-06-05 | 财团法人资讯工业策进会 | Advertisement display system and advertisement display method in combination with search engine service |
| CN103324729A (en) * | 2013-06-27 | 2013-09-25 | 北京小米科技有限责任公司 | Method and device for recommending multimedia resources |
| CN103491411A (en) * | 2013-09-26 | 2014-01-01 | 深圳Tcl新技术有限公司 | Method and device based on language recommending channels |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5152570B2 (en) * | 2008-03-06 | 2013-02-27 | 株式会社デンソー | Automotive user hospitality system |
| JP5182113B2 (en) * | 2009-01-16 | 2013-04-10 | 三菱自動車工業株式会社 | Control device for in-vehicle equipment |
| WO2014002128A1 (en) * | 2012-06-25 | 2014-01-03 | 三菱電機株式会社 | On-board information device |
| KR101467298B1 (en) * | 2012-11-16 | 2014-12-03 | 에스케이플래닛 주식회사 | System and method for recommending contents in vehicle |
| KR20140067687A (en) * | 2012-11-27 | 2014-06-05 | 현대자동차주식회사 | Car system for interactive voice recognition |
-
2014
- 2014-09-02 KR KR1020140116184A patent/KR102249392B1/en active IP Right Grant
-
2015
- 2015-08-20 CN CN201510514457.9A patent/CN105390136B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103137043A (en) * | 2011-11-23 | 2013-06-05 | 财团法人资讯工业策进会 | Advertisement display system and advertisement display method in combination with search engine service |
| CN103137125A (en) * | 2011-11-30 | 2013-06-05 | 北京德信互动网络技术有限公司 | Intelligent electronic device based on voice control and voice control method |
| CN102802114A (en) * | 2012-06-20 | 2012-11-28 | 北京语言大学 | Method and system for screening seat by using voices |
| CN102800315A (en) * | 2012-07-13 | 2012-11-28 | 上海博泰悦臻电子设备制造有限公司 | Vehicle-mounted voice control method and system |
| CN103324729A (en) * | 2013-06-27 | 2013-09-25 | 北京小米科技有限责任公司 | Method and device for recommending multimedia resources |
| CN103491411A (en) * | 2013-09-26 | 2014-01-01 | 深圳Tcl新技术有限公司 | Method and device based on language recommending channels |
Non-Patent Citations (1)
| Title |
|---|
| 《多类噪声环境下的语音端点检测》;汤霖等;《计算机工程与应用》;20121031;第48卷(第29期);第114-118、156页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105390136A (en) | 2016-03-09 |
| KR102249392B1 (en) | 2021-05-07 |
| KR20160027728A (en) | 2016-03-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105390136B (en) | Vehicle arrangement control device and method for user's adaptive type service | |
| US11184412B1 (en) | Modifying constraint-based communication sessions | |
| EP3090429B1 (en) | Modifying operations based on acoustic ambience classification | |
| WO2022060970A1 (en) | Dialog management for multiple users | |
| KR20220054602A (en) | Systems and methods that support selective listening | |
| CN109189980A (en) | The method and electronic equipment of interactive voice are carried out with user | |
| CN109074806A (en) | Distributed audio output is controlled to realize voice output | |
| DE112021001064T5 (en) | Device-directed utterance recognition | |
| CN113168832A (en) | Alternating response generation | |
| CN107819929A (en) | It is preferred that the identification and generation of emoticon | |
| CN111145721A (en) | Personalized prompt language generation method, device and equipment | |
| CN109712615A (en) | System and method for detecting the prompt in dialogic voice | |
| KR20210070213A (en) | Voice user interface | |
| US11393473B1 (en) | Device arbitration using audio characteristics | |
| CN116417003A (en) | Voice interaction system, method, electronic device and storage medium | |
| US11749282B1 (en) | Goal-oriented dialog system | |
| CN110286745A (en) | Dialog process system, the vehicle with dialog process system and dialog process method | |
| CN109274922A (en) | A kind of Video Conference Controlling System based on speech recognition | |
| US20240071408A1 (en) | Acoustic event detection | |
| DE112022000504T5 (en) | Interactive content delivery | |
| US20240062164A1 (en) | Data ingestion and understanding for natural language processing systems | |
| CN109791764A (en) | Communication based on speech | |
| CN103390406A (en) | Speaker authentication method, preparation method of speaker authentication and electronic device | |
| CN117882131A (en) | Multiple wake word detection | |
| US11798538B1 (en) | Answer prediction in a speech processing system |
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 |