CN105390136A - Vehicle control device and method used for user-adaptable service - Google Patents
Vehicle control device and method used for user-adaptable service Download PDFInfo
- Publication number
- CN105390136A CN105390136A CN201510514457.9A CN201510514457A CN105390136A CN 105390136 A CN105390136 A CN 105390136A CN 201510514457 A CN201510514457 A CN 201510514457A CN 105390136 A CN105390136 A CN 105390136A
- Authority
- CN
- China
- Prior art keywords
- user
- service
- adaptive type
- information
- voice messaging
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 230000003044 adaptive effect Effects 0.000 claims description 78
- 238000004458 analytical method Methods 0.000 claims description 27
- 239000000284 extract Substances 0.000 claims description 10
- 230000008451 emotion Effects 0.000 claims description 6
- 230000000052 comparative effect Effects 0.000 claims description 4
- 230000006870 function Effects 0.000 abstract description 37
- 230000008569 process Effects 0.000 description 15
- 238000010586 diagram Methods 0.000 description 8
- 230000036651 mood Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000001914 filtration Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000004378 air conditioning Methods 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000012880 independent component analysis Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000010223 real-time analysis Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 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
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000003245 working effect Effects 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 invention provides a voice pattern identifying drivers through voices and analyzing the drivers through different kinds of information. Firstly, a vehicle device control device and a method used for user-adaptable services are guided to the drivers by a vehicle. The vehicle device control device used for user-adaptable service comprises a characteristic information generation portion generating characteristic information of users according to voice information of the users, a voice information analyzing portion obtaining meaning information through analyzing the voice information, an adaptable service determination portion determining adaptable service for the users according to the characteristic information and the meaning information and a vehicle device control portion controlling control target equipment including the vehicle equipment so as to carry out adaptable service. The invention can realize a natural voice identification system, and provide convenience for the drivers to use common functions.
Description
Technical field
The present invention relates to and control the device and method of vehicle arrangement, particularly relate to and a kind ofly control the vehicle arrangement control device for the service of user's adaptive type and the method that vehicle arrangement makes to perform the service of user's adaptive type.
Background technology
Speech recognition is from the acoustic information extraction phoneme voice and language message and makes a series of processes of machine recognition and reaction.
Although generally believe that with voice dialogue be method natural, the easiest in the message exchange medium of the mankind and machine, human speech must be converted to the code that machine can process can talk with voice and machine.Speech recognition is exactly this process converting code to.
On vehicle, be suitable at present the speech recognition technology of development in recent years, therefore only need the voice command of driver can drive simple convenient means, such as lifting window, startup and stop wiper, open air-conditioning, open or close headlamp etc.
Current vehicle audio recognition methods is below described.
Current vehicle audio recognition methods to comprise when driver's voice send device start order by the step of microphones driver voice, by filter and simulating signal pre-service is the step of digital signal, the step by extraction property vector and classification voice pattern voice command recognition by analog to digital conversion, and according to the step of the voice command drived control object apparatus identified.
Current speech recognition with a small amount of vocabulary of speech engine identification even Large Copacity vocabulary, only can press putting call through immediately after connection (Push-to-Talk; PTT) during key, speech identifying function activates.
But what adopt at present is speech recognition system just forms corresponding scene according to this order one way system when words person gives an order to speech recognition system, cannot two-way exchange.
No. 2014-0051630th, KR published patent discloses a kind of method by speech recognition controlled vehicle image and sound guidance system.But the method is also provide speech identifying function by remote control speech recognition key, therefore cannot solve the problem.
Summary of the invention
Technical matters
For solving the problem, the object of the present invention is to provide and a kind ofly distinguish driver by speech recognition and analyzed the voice pattern of this driver by much information, first guided the vehicle arrangement control device of serving for user's adaptive type and the method for best-of-breed functionality by vehicle to driver.
Object of the present invention is not limited to above-mentioned object, and those skilled in the art clearly understand other objects NM by following record.
Technical scheme
For reaching above-mentioned purpose, the invention provides a kind of vehicle arrangement control device for the service of user's adaptive type, comprising: characteristic information generating unit, it generates the characteristic information of described user according to the voice messaging of user; Voice messaging analysis unit, it obtains implication information by resolving described voice messaging; Adaptive type service determination portion, it determines the adaptive type service to described user according to described characteristic information and described implication information; And vehicle arrangement control part, its control object equipment controlling to comprise vehicle arrangement makes to perform described adaptive type service.
Preferably, described characteristic information generating unit extracts resonance peak (formant) value, frequency values, speech energy value and linear predictive coding (linearpredictioncoding from described voice messaging, hereinafter referred to as ' LPC ') at least one value in value, and generate described characteristic information in real time according at least one value described.
Preferably, described characteristic information generating unit generates at least one information in the emotion information of the gender information of described user, the age information of described user and described user in real time as described characteristic information.
Preferably, described vehicle arrangement control device also comprises: voice messaging selection portion, and it selects a voice messaging when receiving at least two voice messagings from multiple described voice messaging.
Preferably, described voice messaging selection portion selects a described voice messaging according at least one in the position of the size of voice messaging, the comparative result between the voice messaging of input and the voice messaging prestored, described user and multilayer perceptron (multilayerperceptron).
Preferably, described vehicle arrangement control device also comprises: voice messaging input part, and it receives described voice messaging from each seat of vehicle, and described vehicle arrangement control part controls vehicle arrangement and makes to perform described adaptive type service respectively by described each seat.
Preferably, described voice messaging input part comprises the directional microphone (directionalmicrophone) being arranged at described each seat.
Preferably, described vehicle arrangement control device also comprises: adaptive type service execution judging part, and it judges whether to perform described adaptive type service according to the information that described user inputs; And alternative service determination portion, it is not when judged result is for performing the service of described adaptive type, and the information inputted according to described user determines the alternative service of alternative described adaptive type service.
Preferably, the described vehicle arrangement that described vehicle arrangement control part controls is audio-visual navigation (AudioVideoNavigation; AVN) system.
Further, the invention provides a kind of vehicle arrangement control method for the service of user's adaptive type, comprising: the step generating the characteristic information of described user according to the voice messaging of user; The step of implication information is obtained by resolving described voice messaging; The step that the adaptive type of described user is served is determined according to described characteristic information and described implication information; And the control object equipment controlling to comprise vehicle arrangement makes the step performing the service of described adaptive type.
Preferably, the described step generated extracts resonance peak (formant) value, frequency values, speech energy value and linear predictive coding (linearpredictioncoding from described voice messaging, hereinafter referred to as ' LPC ') at least one value in value, and generate described characteristic information in real time according at least one value described.
Preferably, the described step of generation generates at least one information in the emotion information of the gender information of described user, the age information of described user and described user in real time as described characteristic information.
Preferably, also comprise before the described step of generation: the step selecting a voice messaging when receiving at least two voice messagings from multiple described voice messaging.
Preferably, the described step of selection selects a described voice messaging according at least one in the position of the size of voice messaging, the comparative result between the voice messaging of input and the voice messaging prestored, described user and multilayer perceptron (multilayerperceptron).
Preferably, also comprise before the described step of selection: the step receiving described voice messaging from each seat of vehicle, the described step of control specifically controls vehicle arrangement and makes to perform described adaptive type service respectively by described each seat.
Preferably, the described step of reception utilizes the directional microphone being arranged at described each seat.
Preferably, determine also to comprise between the described step of the adaptive type service of described user and the described step of control: the information inputted according to described user judges whether to perform the step that described adaptive type is served; And when judged result is not for performing the service of described adaptive type, the information inputted according to described user determines the step of the alternative service of alternative described adaptive type service.
Preferably, the described vehicle arrangement that the described step of control controls is image and sound guidance system (AudioVideoNavigation; AVN).
Technique effect
The present invention is distinguished driver by speech recognition and is analyzed the voice pattern of this driver by much information, first guides best-of-breed functionality by vehicle to driver, thus has following beneficial effect:
The first, the trend changed to two-way exchange mode gradually can be complied with and change into two-way communication from one way system, thus natural speech recognition system can be realized.
The second, system, according to driver's correspondingly recommendation function, is therefore convenient to driver and is used common function.
Accompanying drawing explanation
Fig. 1 is the concept map showing vehicle Speaker identification system according to an embodiment of the invention;
Fig. 2 is the process flow diagram of the first embodiment of the method for work of display vehicle Speaker identification system;
Fig. 3 is the process flow diagram of the second embodiment of the method for work of display vehicle Speaker identification system;
Fig. 4 summarizes display according to the preferred embodiment of the invention for the block diagram of the vehicle arrangement control device of user's adaptive type service;
Fig. 5 summarizes display according to the preferred embodiment of the invention for the process flow diagram of the vehicle arrangement control method of user's adaptive type service.
Embodiment
The preferred embodiments of the present invention are described in detail referring to accompanying drawing.First, it should be noted that in the inscape interpolation Reference numeral of each figure, even if identical inscape appears on different accompanying drawings also add identical Reference numeral as far as possible.And if judge to think illustrating to cause theme of the present invention and obscure related known structure or function when illustrating of the present invention, then omit relevant detailed description in detail.In addition, preferred embodiments of the present invention will be described below, but technical scheme of the present invention does not limit or is limited to this, and person of ordinary skill in the field can do various deformation and implement.
The invention is characterized in the technology trends according to changing to two-way communication mode, utilize speech recognition this Speaker identification function distinguishing driver and analyze the voice pattern of this driver, first for driver recommends optimal function, artificial intelligence trend can be complied with.
Fig. 1 is the concept map showing vehicle Speaker identification system according to an embodiment of the invention.
As shown in Figure 1, the present invention is by speech recognition words person and analyzes driver's friendly vehicle interior system of the voice pattern of each driver.
Private car is generally shared by many people, and the present invention stores the characteristics of speech sounds of each driver and analyzes the voice pattern of each driver.The voice pattern of driver can be search for ground, recently call catalog, audio-frequency function etc. recently.Driver by bus after distinguish that who driver words person is when being spoken by microphone, vehicle confirms the function being applicable to this driver's voice pattern, guarantees the faster easier function commonly used close to driver.
The function of input part 110 is the voice commands receiving driver.Input part 110 can be microphone.
Identification part 120 is identified by the voice signal that input part 110 inputs.By sound (SpeechToText, hereinafter referred to as ' STT ') operation of larding speech with literary allusions, identification part 120 judges that what voice what receive is.
Analysis portion 130 distinguishes words person, analyzes sex and age bracket, by obtaining everyone characteristic of resonance peak (formant) value identification by the database (Database) of study.
Analysis portion 130 by the real time discriminating user statistically such as resonance peak (formant) value, basic frequency value, speech energy value, linear predictive coding (linearpredictioncoding, hereinafter referred to as ' LPC ') value of words person's voice sex/age/mood/state etc.
Storage part 160 stores the characteristic of each driver collected by analysis portion 130.Now, storage part 160 stores the result that voice command sound that 120 pairs, identification part driver says is larded speech with literary allusions.
Handling part 140 utilizes the DB plan stored by driver to forward next scene to, and whether inquiry driver will forward the corresponding scene that vehicle is recommended to, first guides driver's common function to make vehicle.
Such as, handling part 140 builds the destination of first recommending driver A often to go at special time period by vehicle to driver, the radio broadcasting often listened, digital media broadcast (DigitalMediaBroadcasting; DMB) scene of channel etc., or driver's music of often listening in the music playing vehicle storage by priority, or grasp age of driver and the people playing this age bracket when being listened to music by server likes the music of listening.
Handling part 140 searches for (searching) data, to guarantee, when processing the function needed for user, especially can provide the adaptive type facilitating functions information of the user personality of the most applicable analysis portion 130 real-time analysis when user does not specify precise information.
Handling part 140 provides and selects music, recommends the adaptive type facilitating functions information such as radio broadcasting, recommending digital multimedia broadcasting, searching facility.
Whether efferent 150 will make the system in vehicle by the scene work drawn according to the customizing messages of driver by loudspeaker inquiry driver.
Efferent 150 transmits handling part 140 to user and processes the result drawn.
Fig. 2 is the process flow diagram of the first embodiment of the method for work of display vehicle Speaker identification system.
When user says " broadcast " by microphone (MIC), in step S210, input part 110 obtains this voice signal.
Then, in step S220, identification part 120 performs speech identifying function, by STT, broadcasting command is converted to RTA Radio Text.
Then, in step S230, analysis portion 130 utilizes the DB of study and resonance peak, basic frequency value, speech energy value, LPC value etc. analyze the voice command of driver and store.
Then in step S240, handling part 140 confirms current time, and generate waveform (wave) file according to the analysis result of the recognition result of identification part 120, analysis portion 130, the information etc. that is stored in storage part 160, broadcast with the TBS that the frequency guiding this time driver to listen to is FM95.1.
Then in step s 250, efferent 150 with loudspeaker export " listening to FM95.1TBS Traffic Announcement? "
When receiving the information of the content agreeing to be exported by loudspeaker from user, in step S260, audio-visual navigation (AudioVideoNavigation, hereinafter referred to as ' the AVN ') system of vehicle exports FM95.1TBS Traffic Announcement.
On the contrary, if receive the information of the content not agreeing to be exported by loudspeaker from user, then in step S270, handling part 140 exports " please say channel " by efferent 150 loudspeaker.Then, in step S270, when driver orders required frequency by input part 110, handling part 140 exports the broadcast of corresponding frequencies.
Speech recognition system differentiate press that driver characteristics is recommended after driver and driver refuses when, as above inquire required function and by this function operation to driver.
Speech recognition system has the situation of erroneous judgement driver, therefore in AVN function, add driver's recognition function On/Off (ON/OFF) function, when makeing mistakes, driver is set to closedown (OFF) and makes speech recognition system nonrecognition words person.
Speech recognition system of the present invention dependent carries out speech recognition, but by sharing the information realization Vehicular system more friendly to driver with other modules.
Fig. 3 is the process flow diagram of the second embodiment of the method for work of display vehicle Speaker identification system.
When saying broadcast or DMB with voice in step S310, analysis portion 130 extracts the characteristic of user by the step S320 of speech analysis.Then, handling part 140 selects the broadcast of the most applicable user personality.
If be in during phonetic search music the state be connected with server, then analysis portion 130 by voice extract user characteristic and from server provide different sexes/age/music list liked of the crowd of state select.
Even if be not connected with server, the general characteristic (sex liked, age bracket, mood) about music file also can be stored in storage part 160 by analysis portion 130, plays think optimal music according to the voice status of user.
When by navigating search facility, if handling part 140 searches for such as periphery dining room by the characteristics of speech sounds of user, then show the dining room that each sex/age bracket is liked at first.When using Xi Li (SIRI), thering is provided the result of being searched for by YELP with during phonetic search periphery dining room to user, also utilizing the dining room that other information searches such as sex/age bracket and the more approximate people of this user like when searching for periphery dining room.
If there is network to connect, then handling part 140 utilizes voice command and user personality information carry out web search and information is supplied to user.First congee shop is searched for, if think that first user's suffer from the disease is into flu, then search internal medicine by analyzing user speech during search periphery hospital when learning that health state of user is not good by speech analysis.
And when with phonetic search concert or concert, the concert that the people that handling part 140 is similar to environment by web search and this user like or concert.
Below sum up content described above.
In step S310, when user gives orders or instructions, input part 110 obtains corresponding voice messaging.Such as, when user says " broadcast " or " DMB ", input part 110 obtains this voice messaging.
Then, analysis portion 130 in step s 320 real-time analysis voice messaging and judge in step S330 ~ S350 sex/age/state etc.
By voice messaging, analysis portion 130 judges that words person is the male sex or women in step S330.Then, analysis portion 130 judges words person's age bracket (such as, two teens, three teens, four teens etc.) in step S340.Then, analysis portion 130 judges that in step S350 the mood/state of words person is good bad.
Then, in step S360, handling part 140 is according to the radio station of the most applicable user of analysis result search of analysis portion 130.Such as, handling part 140 searches for the radio station that the two good teens male sex of mood/state like listening, or the three teens male sex of search mood/be not in good state like the radio station of listening, or the good teens women of search mood/state likes the radio station of listening, two teens women of or search mood/be not in good state like the radio station of listening.
Then, in step S370, efferent 150 is by the result (execution respective service) of loudspeaker output processing part 140.
The invention is not restricted to the driver in vehicle, and not only consider the content of speaking in all speech recognition situations when the information of search, also consider words person's characteristic simultaneously.
Be different from prior art, the present invention grasps the User Status that may change by real-time analysis voice and correspondingly provides user's adaptive type information.
Below illustrate that speech recognition of the present invention is suitable for the method for back-propagation algorithm (BackPropagationAlgorithm).
According to general noise filtering methods, opening voice identification microphone also sends voice for speech recognition after a predetermined time afterwards, and signal speech recognition being advanced into microphone is judged as noise in vehicle and this noise in trap signal.
Although have the directional microphone arranged towards driver in vehicle, but because the signal of input blink before being given orders or instructions by voice is judged as noise, therefore when speech recognition is given orders or instructions time point also given orders or instructions in other seats in addition to a driver, voice mix mutually, and therefore phonetic recognization rate declines.
Therefore, four seating areas of the present invention respectively in vehicle arrange directional microphone, with the input signal of the microphone in driver region for benchmark, other region microphone signals are determined as noise and filter.The characteristic of the driver in real time discriminating driver region in the process of processing signals, to guarantee that multimedia equipment provides the information of applicable driver.
Below this is described in further details, in below illustrating, driver's seat is defined as a-quadrant, front passenger's seat is defined as B region, the rear side seat of the back seat of driver's seat and front passenger's seat is defined as C region and D region respectively.
When driver starts speech identifying function, the microphone in A, B, C, D region is opened, by the voice signal in microphones four regions simultaneously.When the vehicle noise not being human speech is input to the microphone in four regions, its input value is almost identical, therefore by a-quadrant filter vehicle noise figure.Further, the voice in four regions are analyzed.First analyzing other speech vector value of representative in four regions, take a-quadrant as benchmark, and when from B, C, D extracted region to the vector value represented with a-quadrant different sexes, the signal corresponding to this vector value is filtered in a-quadrant.Age, mood/state etc. is analyzed in the same way after having analyzed sex.
Although voice signal maximum in a-quadrant is the voice signal of driver, when also having the voice signal in B, C, D region, only cannot extract the sound of a-quadrant driver, therefore adopting the method.
Now mutual relationship (CORRELATION), independent component analysis (IndependentComponentAnalysis can be passed through; ICA) other algorithm judgment signal outside technology, Wave beam forming (BEAMFORMING) technology are independent still has approximation.
Can carry out by four microphones the individual characteristic filtering to grasp words person, the information filtering noise grasped individual characteristic and obtain can be utilized, improve discrimination with this.
Below multilayer perceptron (multilayerperceptron) is described.
The theoretical emotion for identifying voice (judging the content of voice when receiving voice) or differentiation people of the existing perceptron relevant to voice (perceptron).
Multilayer perceptron (multilayerperceptron) is the neural network between input layer and output layer with more than one middle layer.Network presses input layer, concealment layer, output layer direction connects, and is there is not connection in each layer and direct feedforward (Feedforward) network that be connected of output layer to input layer.
Vehicle generally has four seats, in vehicle, the user of speech recognition system is generally driver, use in the process of speech recognition system driver, when the passenger at other seats gives orders or instructions, the voice of many people are overlapping, the therefore order of speech recognition system None-identified driver.The speech recognition system generally used at present is that setting does not have the interval of voice and this interval is identified as noise and in the structure of the interval filtered noise of phonetic entry before speech recognition interval.
The present invention utilizes perceptron theory to extract the characteristic of voice to identify words person's characteristic, provides the technology of applicable information according to these data in real time to words person.By perceptron, 1. can provide adaptive type information by the characteristic of each words person, 2. can identify words person position and the function needed for the person of this position is provided.Below to 1. and being 2. described in further details.
1. provide adaptive type information according to words person's characteristic
When utilizing multilayer perceptron construction system, even if the voice of many people occur that superposition also can extract the voice of driver.The method is not limited to driver, can also be used for identifying other people.Such as, only extract the characteristics of speech sounds of a-quadrant, ignore the voice signal in all the other B, C, D regions.
The major premise of perceptron is the state being formed with the algorithm utilizing backpropagation (BACKPROPAGATION) technique drill in advance according to a large amount of DB.
About perceptron modeling, such as by analysis two teens and a large amount of voice extraction properties (resonance peak, basic frequency, energy value, LPC value etc.) of the good Soul women of state being inputted by input end, export (OUTPUT) object and be two teens and state good Soul women time, perceptron inside configuration determines suitable weighting (WEIGHT) value through backpropagation (BACKPROPAGATION) process.When as above the people of multifrequency nature being trained, no matter input any voice and can both find characteristic in trained structure.LPC value is linear prediction symbolism value, is based on the one in the phonic symbol mode of mankind's generation model, has the vector of 26 dimensions.
During 26 dimensional vector value of the resonance peak of a large amount of voice of input special object, basic frequency, LPC model, by reverse expansion process, operation that suitable weighted value specifies (two teens and the good Soul women of state, three teens and area, the Qing Shang road the be not in good state male sex are repeated to multiple object ...).
By above-mentioned training process, no matter receiving any voice can both by being input to the characteristic perceptron structure of the eigen vector modeling of these voice being learnt to words person.
With putting call through immediately after connection (Push-to-Talk, hereinafter referred to as ' PTT ') as seat selection reference.If there are four PTT keys, then according to its position, by the microphones that receives the position of corresponding PTT to phonetic decision be the voice of Water demand, all the other are judged as noise and filter.Undertaken identifying by the voice after filtration and provide best information for words person, such as words person gives an order to media product and searches periphery dining room, then first search out the periphery dining room of applicable words person's characteristic.
Arrange above description and can derive following characteristic.
First, differentiate PTT position and extract the vector 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 that other characteristic values in A microphone signal are judged as noise and filter.
Then, utilize the data execution speech recognition of only extracting a-quadrant voice and obtaining, judge the implication of voice.
Then, the information of the order of the most applicable a-quadrant words person is provided.
2. identify words person position and the function needed for words person of this position is provided
With putting call through immediately after connection (Push-to-Talk, hereinafter referred to as ' PTT ') as seat selection reference.If there are four PTT keys, then according to its position, by the microphones that receives the position of corresponding PTT to phonetic decision be the voice of Water demand, all the other are judged as noise and filter.For air-conditioning, when the passenger being sitting in D region sends the order about air-conditioner temperature, only can change the air-conditioning gear of the aircondition in D region according to order.
Be explained above and distinguish that the speech recognition system of words person provides one embodiment of the present of invention of best information to driver by speech recognition.The preferred embodiments of the present invention can reasoned out from these embodiments are below described.
Fig. 4 summarizes display according to the preferred embodiment of the invention for the block diagram of the vehicle arrangement control device of user's adaptive type service.
With reference to Fig. 4, the vehicle arrangement control device 400 for the service of user's adaptive type comprises characteristic information generating unit 410, voice messaging analysis unit 420, adaptive type service determination portion 430, vehicle arrangement control part 440, power supply unit 450 and master control part 460.
The function of power supply unit 450 provides power supply to each formation forming vehicle arrangement control device 400.The function of master control part 460 is all workings controlling each formation forming vehicle arrangement control device 400.Consider and vehicle arrangement control device 400 can be arranged on AVN system, therefore not have power supply unit 450 also harmless with master control part 460 for the present embodiment.
The function of characteristic information generating unit 410 is the characteristic informations generating user according to the voice messaging of user.
Characteristic information generating unit 410 extracts at least one value resonance peak (formant) value, frequency values, speech energy value and LPC value from voice messaging, can according to the real-time formation characteristic information of this at least one value.
The characteristic information of the user that characteristic information generating unit 410 generates in real time can be the gender information of user, in the emotion information of the age information of user and user to planting a kind of information.
Characteristic information generating unit 410 is the concepts of the analysis portion 130 corresponded in Fig. 1.
The function of voice messaging analysis unit 420 is the voice messaging acquisition implication information by resolving user.
The function of adaptive type service determination portion 430 is adaptive type services that the implication information obtained according to characteristic information and the voice messaging analysis unit 420 of characteristic information generating unit 410 generation determines to user.
The function of vehicle arrangement control part 440 is adaptive type services that the control object equipment controlling to comprise vehicle arrangement makes execution adaptive type service determination portion 430 determine.
The vehicle arrangement that vehicle arrangement control part 440 controls can be audio-visual navigation (AudioVideoNavigation; AVN) system.
Vehicle arrangement control device 400 can also comprise voice messaging selection portion (not shown).
The function of voice messaging selection portion from these voice messagings, selects a voice messaging when receiving at least two voice messagings.
Voice messaging selection portion can select a voice messaging according at least one in the position of the comparative result between the voice messaging of the size of voice messaging, input and the voice messaging prestored, user and multilayer perceptron (multilayerperceptron).
When based on multilayer perceptron, voice messaging selection portion can select voice messaging in the following order.
First, by by microphones to each area voice of vehicle interior be input to housebroken perceptron model to extract driver information.
Then, if other regions have the voice of other characteristics of the characteristic being different from driver the region being benchmark, then this signal be input in the voice signal of the microphone in driver region is determined as noise and filters.
Then, be input to according to the voice being input to all positions the result that perceptron model draws respectively to carry out filtering to obtain voice messaging.
Vehicle arrangement control device 400 can also comprise voice messaging input part (not shown).
The function of voice messaging input part receives at least one voice messaging.Especially, the function of voice messaging input part is each seat receiving speech information at vehicle.Voice messaging input part can be arranged at each seat with directional microphone form.
In this case, vehicle arrangement control part 440 can control vehicle arrangement and makes to perform adaptive type service by each seat.
Vehicle arrangement control device 400 can also comprise adaptive type service execution judging part (not shown) and alternative service determination portion (not shown).
The function of adaptive type service execution judging part judges whether to perform adaptive type service according to the information of user's input.
The function of alternative service determination portion is when judged result is not for performing adaptive type service, determines the alternative service of alternative adaptive type service according to the information of user's input.
In this case, vehicle arrangement control part 440 makes vehicle arrangement perform alternative service by control.
The method of work of the vehicle arrangement control device 400 being used for the service of user's adaptive type is below described.
Fig. 5 summarizes display according to the preferred embodiment of the invention for the process flow diagram of the vehicle arrangement control method of user's adaptive type service.Be described referring to Fig. 5.
First, in step S510, voice messaging input part receives the voice messaging of user from each seat of vehicle.
Then, in step S530, characteristic information generating unit 410 generates the characteristic information of user according to the voice messaging of user.And in step S520, voice messaging analysis unit 420 obtains implication information by the voice messaging of resolving user.Step S530 can perform with step S520 simultaneously, but also can perform before step S520 or perform after step S520.
Then, in step S540, the characteristic information that adaptive type service determination portion 430 generates according to characteristic information generating unit 410 and the implication information that voice messaging analysis unit 420 obtains determine the adaptive type service to user.
Then in step S550, the adaptive type service that the control object equipment that vehicle arrangement control part 440 controls to comprise vehicle arrangement makes execution adaptive type service determination portion 430 determine.
In addition after step S510, when voice messaging selection portion receives at least two voice messagings, a voice messaging can be selected from these voice messagings.The above-mentioned steps of voice messaging selection portion can perform or perform between step S510 and step S530 between step S510 and step S520.
In addition between step S540 and step S550, adaptive type service execution judging part judges whether to perform adaptive type service according to the information that user inputs.Then, alternative service determination portion, when judged result is not for performing adaptive type service, determines the alternative service of alternative adaptive type service according to the information of user's input.
More than describe and form all inscapes of the embodiment of the present invention and to be combined into one or in conjunction with work, but the present invention is not limited to these embodiments.Namely, within the scope of object of the present invention, its all inscape can by more than one optionally in conjunction with work.And, its all inscape can respectively with one independently hardware form occur, but also optionally can combining part or all of each inscape, being realized by the computer program with the program module for performing the part or all of function that one or more hardware combinations realizes.And, this computer program can be stored in the computer-readable recording mediums (ComputerReadableMedia) such as USB storage, CD disk, flash disk (FlashMemory), read by computing machine and perform, to realize embodiments of the invention.The recording medium of computer program can comprise magnetic recording medium, optical recording media, carrier wave (CarrierWave) medium etc.
Further, all terms comprising technology or scientific words represent the meaning identical with the usual understanding of general technical staff of the technical field of the invention when nothing in illustrating defines separately.The term of normally used dictionary definition, should be interpreted as the meaning consistent with the meaning of the context of correlation technique, if undefined in the present invention, shall not be construed as the meaning of ideal or excessive formality.
More than illustrate and just illustrate technical scheme of the present invention, those of ordinary skill in the art can carry out multiple correction, change and replacement in the scope not departing from intrinsic propesties of the present invention.Therefore, disclosed embodiment of this invention and accompanying drawing non-limiting technical scheme of the present invention, but for illustration of, technical scheme of the present invention is not defined in these embodiments and accompanying drawing.Protection scope of the present invention is determined by technical scheme, is all contained in technical scheme of the present invention with all technical schemes of its equivalency range.
Claims (13)
1., for a vehicle arrangement control device for user's adaptive type service, it is characterized in that, comprising:
Characteristic information generating unit, it generates the characteristic information of described user according to the voice messaging of user;
Voice messaging analysis unit, it obtains implication information by resolving described voice messaging;
Adaptive type service determination portion, it determines the adaptive type service to described user according to described characteristic information and described implication information; And
Vehicle arrangement control part, its control object equipment controlling to comprise vehicle arrangement makes to perform described adaptive type service.
2. the vehicle arrangement control device for the service of user's adaptive type according to claim 1, is characterized in that:
Described characteristic information generating unit extracts at least one value resonance peak, frequency values, speech energy value and linear predictive coding value from described voice messaging, and generates described characteristic information in real time according at least one value described.
3. the vehicle arrangement control device for the service of user's adaptive type according to claim 1, is characterized in that:
Described characteristic information generating unit generates at least one information in the emotion information of the gender information of described user, the age information of described user and described user in real time as described characteristic information.
4. the vehicle arrangement control device for the service of user's adaptive type according to claim 1, is characterized in that, also comprise:
Voice messaging selection portion, it selects a voice messaging when receiving at least two voice messagings from multiple described voice messaging.
5. the vehicle arrangement control device for the service of user's adaptive type according to claim 4, is characterized in that:
Described voice messaging selection portion selects a described voice messaging according at least one in the position of the size of voice messaging, the comparative result between the voice messaging of input and the voice messaging prestored, described user and multilayer perceptron.
6. the vehicle arrangement control device for the service of user's adaptive type according to claim 1, is characterized in that, also comprise:
Voice messaging input part, it receives described voice messaging from each seat of vehicle,
Described vehicle arrangement control part controls vehicle arrangement and makes to perform described adaptive type service respectively by described each seat.
7. the vehicle arrangement control device for the service of user's adaptive type according to claim 6, is characterized in that:
Described voice messaging input part comprises the directional microphone being arranged at described each seat.
8. the vehicle arrangement control device for the service of user's adaptive type according to claim 1, is characterized in that, also comprise:
Adaptive type service execution judging part, it judges whether to perform described adaptive type service according to the information that described user inputs; And
Alternative service determination portion, it is not when judged result is for performing the service of described adaptive type, and the information inputted according to described user determines the alternative service of alternative described adaptive type service.
9. the vehicle arrangement control device for the service of user's adaptive type according to claim 1, is characterized in that:
The described vehicle arrangement that described vehicle arrangement control part controls is image and sound guidance system.
10., for a vehicle arrangement control method for user's adaptive type service, it is characterized in that, comprising:
The step of the characteristic information of described user is generated according to the voice messaging of user;
The step of implication information is obtained by resolving described voice messaging;
The step that the adaptive type of described user is served is determined according to described characteristic information and described implication information; And
The control object equipment controlling to comprise vehicle arrangement makes the step performing the service of described adaptive type.
The 11. vehicle arrangement control methods for the service of user's adaptive type according to claim 10, is characterized in that:
The described step generated generates at least one information in the emotion information of the gender information of described user, the age information of described user and described user in real time as described characteristic information.
The 12. vehicle arrangement control methods for the service of user's adaptive type according to claim 10, is characterized in that, also comprise:
The step of described voice messaging is received from each seat of vehicle,
The described step controlled specifically controls vehicle arrangement and makes to perform described adaptive type service respectively by described each seat.
The 13. vehicle arrangement control methods for the service of user's adaptive type according to claim 10, is characterized in that, also comprise:
The information inputted according to described user judges whether the step performing the service of described adaptive type; And
When judged result is not for performing the service of described adaptive type, the information inputted according to described user determines the step of the alternative service of alternative described adaptive type service.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020140116184A KR102249392B1 (en) | 2014-09-02 | 2014-09-02 | Apparatus and method for controlling device of vehicle for user customized service |
| KR10-2014-0116184 | 2014-09-02 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105390136A true CN105390136A (en) | 2016-03-09 |
| CN105390136B 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) |
Cited By (4)
| 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 |
| WO2018006470A1 (en) * | 2016-07-07 | 2018-01-11 | 深圳狗尾草智能科技有限公司 | Artificial intelligence processing method and device |
| CN110503947A (en) * | 2018-05-17 | 2019-11-26 | 现代自动车株式会社 | Conversational system, the vehicle including it and dialog process method |
| CN111681651A (en) * | 2019-03-11 | 2020-09-18 | 本田技研工业株式会社 | Agent device, agent system, server device, agent device control method, and storage medium |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102497299B1 (en) | 2016-06-29 | 2023-02-08 | 삼성전자주식회사 | Electronic apparatus and method for controlling the electronic apparatus |
| 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 |
| KR20190114325A (en) | 2018-03-29 | 2019-10-10 | 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 |
| 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 |
| 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 |
| 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 |
| 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 |
| DE112012006617B4 (en) * | 2012-06-25 | 2023-09-28 | Hyundai Motor Company | 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 |
|---|
| 汤霖等: "《多类噪声环境下的语音端点检测》", 《计算机工程与应用》 * |
Cited By (5)
| 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 |
| WO2018006470A1 (en) * | 2016-07-07 | 2018-01-11 | 深圳狗尾草智能科技有限公司 | Artificial intelligence processing method and device |
| CN110503947A (en) * | 2018-05-17 | 2019-11-26 | 现代自动车株式会社 | Conversational system, the vehicle including it and dialog process method |
| CN111681651A (en) * | 2019-03-11 | 2020-09-18 | 本田技研工业株式会社 | Agent device, agent system, server device, agent device control method, and storage medium |
| CN111681651B (en) * | 2019-03-11 | 2024-01-02 | 本田技研工业株式会社 | Agent device, agent system, server device, method for controlling agent device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105390136B (en) | 2019-05-21 |
| KR20160027728A (en) | 2016-03-10 |
| KR102249392B1 (en) | 2021-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105390136A (en) | Vehicle control device and method used for user-adaptable service | |
| CN110070868B (en) | Voice interaction method and device for vehicle-mounted system, automobile and machine readable medium | |
| EP3090429B1 (en) | Modifying operations based on acoustic ambience classification | |
| CN102842306B (en) | Sound control method and device, voice response method and device | |
| CN110032660A (en) | Personalized audio content is generated based on mood | |
| CN106816149A (en) | The priorization content loading of vehicle automatic speech recognition system | |
| US20220139389A1 (en) | Speech Interaction Method and Apparatus, Computer Readable Storage Medium and Electronic Device | |
| CN107819929A (en) | It is preferred that the identification and generation of emoticon | |
| DE102018125966A1 (en) | SYSTEM AND METHOD FOR RECORDING KEYWORDS IN A ENTERTAINMENT | |
| CN107554456A (en) | Vehicle-mounted voice control system and its control method | |
| WO2022001347A1 (en) | In-vehicle voice instruction control method, and related device | |
| CN105575383A (en) | Apparatus and method for controlling target information voice output through using voice characteristics of user | |
| CN110286745A (en) | Dialog process system, the vehicle with dialog process system and dialog process method | |
| CN103685783A (en) | Information processing system and storage medium | |
| CN105005276A (en) | Methods for providing operator support utilizing a vehicle telematics service system | |
| CN110520323A (en) | For controlling method, apparatus, mobile subscriber equipment and the computer program of vehicle audio frequency system | |
| CN113454717A (en) | Speech recognition apparatus and method | |
| CN111081244B (en) | Voice interaction method and device | |
| CN115148197A (en) | Voice wake-up method, device, storage medium and system | |
| CN114245280A (en) | Scene self-adaptive hearing aid audio enhancement system based on neural network | |
| CN111178081A (en) | Semantic recognition method, server, electronic device and computer storage medium | |
| Priyanka et al. | Multi-channel speech enhancement using early and late fusion convolutional neural networks | |
| CN115050375A (en) | Voice operation method and device of equipment and electronic equipment | |
| Nakagome et al. | Efficient and Stable Adversarial Learning Using Unpaired Data for Unsupervised Multichannel Speech Separation. | |
| US20240079007A1 (en) | System and method for detecting a wakeup command for a voice assistant |
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 |