The content of the invention
For this reason, it may be necessary to which providing a kind of car-mounted terminals such as car running computer that increase logs in safe intelligent automobile voice login methodAnd system.
To achieve the above object, a kind of intelligent automobile voice login method is inventor provided, including：
User speech information and device id are obtained, device id or mobile terminal of the device id including on-vehicle hostDevice id；
Parsing is carried out to user speech information and obtains vocal print feature, and vocal print and device id are sent to cloud server；
Cloud server is retrieved by carrying out vocal print feature and device id matching in database beyond the clouds, the high in the clouds dataStorehouse stores the vocal print feature obtained during user's registration, device id, username and password, when vocal print feature and device id matchingWhen, cloud server sends the feedback information for agreeing to log in on-vehicle host；When vocal print feature and device id are mismatched, then cloudServer is held to send the feedback information for refusing to log in on-vehicle host；
On-vehicle host performs the feedback information of cloud server.
Further optimization, the vocal print feature includes：Acoustic feature, lexical characteristics, prosodic features and channel information.
Further optimization, " cloud server is retrieved by carrying out vocal print and device id matching in database beyond the clouds "Including：
By in database beyond the clouds retrieve vocal print feature whether match, and device id whether be on-vehicle host equipmentOne in the device id of ID or mobile terminal.
Further optimization, when the cloud server is by user's registration, acquisition sound is carried out according to the content pronunciation of regulationLine feature, and store to cloud database.
Further optimization, when the cloud server is by user's registration, the content pronunciation set according to user is obtainedVocal print feature is taken, and is stored to cloud database.
The present inventor provides a kind of intelligent automobile voice login system of another scheme again, it is characterised in that including carCarry main frame, mobile device and cloud server；
The on-vehicle host is used for the device id for obtaining user speech information and on-vehicle host, and user speech information is enteredRow parsing obtains speech characteristic value, and speech characteristic value and device id are sent to cloud server, and the on-vehicle host is also usedIn the feedback information for performing cloud server；
The mobile terminal is used for the device id for obtaining user speech information and mobile terminal, and user speech information is enteredRow parsing obtains speech characteristic value, and speech characteristic value and device id are sent to cloud server；
The cloud server includes cloud database, and the cloud database stores the voice inputted during user's registrationCharacteristic value, device id, username and password, the cloud server are used in database beyond the clouds carry out vocal print feature and equipmentID matching retrievals, when vocal print feature and device id matching, cloud server sends the feedback letter for agreeing to log in on-vehicle hostBreath；When vocal print feature and device id are mismatched, then cloud server sends the feedback information for refusing to log in on-vehicle host.
Further optimization, the vocal print feature includes：Acoustic feature, lexical characteristics, prosodic features and channel information.
Further optimization, " cloud server is retrieved by carrying out vocal print and device id matching in database beyond the clouds "Including：
By in database beyond the clouds retrieve vocal print feature whether match, and device id whether be on-vehicle host equipmentOne in the device id of ID or mobile terminal.
Further optimization, when the cloud server is by user's registration, acquisition sound is carried out according to the content pronunciation of regulationLine feature, and store to cloud database.
Further optimization, when the cloud server is by user's registration, the content pronunciation set according to user is obtainedVocal print feature is taken, and is stored to cloud database.
The present inventor additionally provides a kind of intelligent automobile voice login system of another technical scheme, including on-vehicle host,Mobile terminal and cloud server；
The on-vehicle host is used to obtaining the device id of user speech information and on-vehicle host, and by voice messaging and equipmentID is sent to cloud server, and the on-vehicle host is additionally operable to perform the feedback information of cloud server；
The mobile terminal is used to obtaining the device id of user speech information and mobile terminal, and by voice messaging and movementThe device id of terminal is sent to cloud server；
The cloud server includes cloud database, and the cloud database stores the voice inputted during user's registrationCharacteristic value, device id, username and password, the cloud server are used to carry out voice messaging parsing acquisition vocal print feature,Vocal print feature and device id matching retrieval are carried out in database beyond the clouds, when vocal print feature and device id matching, cloud serviceDevice sends the feedback information for agreeing to log in on-vehicle host；When vocal print feature and device id are mismatched, then cloud server toOn-vehicle host sends the feedback information that refusal is logged in.
Further optimization, the vocal print feature includes：Acoustic feature, lexical characteristics, prosodic features and channel information.
Further optimization, " cloud server is retrieved by carrying out vocal print and device id matching in database beyond the clouds "Including：
Judge it is that on-vehicle host passage obtains voice messaging or mobile terminal leads to according to the channel information of vocal print featureRoad obtains voice messaging；
If on-vehicle host passage obtains voice messaging, then the device id for carrying out on-vehicle host carries out matching comparison and correspondinglyVocal print feature, grammar property and the prosodic features that on-vehicle host passage is obtained carry out matching comparison；
If mobile terminal passage obtains voice messaging, then the device id for moving terminal carries out matching comparison and correspondinglyVocal print feature, grammar property and the prosodic features that mobile terminal passage is obtained carry out matching comparison.
Further optimization, when the cloud server is by user's registration, acquisition sound is carried out according to the content pronunciation of regulationLine feature, and store to cloud database.
Further optimization, when the cloud server is by user's registration, the content pronunciation set according to user is obtainedVocal print feature is taken, and is stored to cloud database.
Prior art is different from, above-mentioned technical proposal obtains vocal print feature by carrying out parsing to user speech information, andBy the device id or the device id of mobile terminal of on-vehicle host, server progress retrieval matching, works as vocal print feature beyond the clouds, can be with login system, it is not necessary to user with the device id of the device id of on-vehicle host or mobile terminal after the match is successful simultaneouslyCarry out providing user name and login password and verified, it is to avoid occur password forget or password leakage situation and after producingReally.
Embodiment
To describe the technology contents of technical scheme in detail, feature, the objects and the effects being constructed, below in conjunction with specific realityApply example and coordinate accompanying drawing to be explained in detail.
Referring to Fig. 1, a kind of intelligent automobile voice login method of the present embodiment, including：
Step 110：User speech information and device id are obtained, the device id includes device id or the shifting of on-vehicle hostThe device id of dynamic terminal；
Step 120：Parsing is carried out to user speech information and obtains vocal print feature, and vocal print and device id are sent to high in the cloudsServer；
Step 130：Cloud server is retrieved by carrying out vocal print feature and device id matching in database beyond the clouds, describedCloud database stores the vocal print feature obtained during user's registration, device id, username and password, when vocal print feature and equipmentWhen ID is matched, cloud server sends the feedback information for agreeing to log in on-vehicle host；When vocal print feature and device id are mismatchedWhen, then cloud server sends the feedback information for refusing to log in on-vehicle host；
Step 140：On-vehicle host performs the feedback information of cloud server.
Vocal print not only has specificity, and the characteristics of have relative stability.After adult, the sound of people can keep long-termStablize relatively constant.It is demonstrated experimentally that no matter talker is deliberately to imitate other people sound and the tone, or whisper in sb.'s ear is softly talked, i.e.,Make to imitate remarkably true to life, its vocal print is but differed all the time.And Application on Voiceprint Recognition, one kind of biological identification technology.Application on Voiceprint Recognition is justIt is that acoustical signal is converted into electric signal, then is identified with computer.The generation of human language is Body Languages maincenter and pronunciationA complicated physiology physical process between organ, phonatory organ -- tongue, tooth, larynx, lung, nasal cavity that people uses in speechEveryone is widely different in terms of size and form, so the voiceprint map of any two people is all variant.
By the microphone on the microphone or mobile terminal in on-vehicle host obtain the voice messaging of user, andAnd by being parsed to voice messaging after, obtain vocal print feature, while obtain on-vehicle host device id or mobile terminalDevice id, beyond the clouds the cloud database in server retrieved, vocal print feature and device id are matched, work as vocal printFeature and device id are matched simultaneously when, then cloud server feeds back to on-vehicle host and logged in, conversely, when vocal print feature and settingWhen standby ID is mismatched, then, cloud server feeds back to on-vehicle host refusal login.So pass through the vocal print feature progress of userMatch somebody with somebody, and the device id or the device id of mobile terminal by on-vehicle host are matched, increase is accurate, without inputting user nameAnd login password, it is user-friendly, and simultaneously because the uniqueness of vocal print feature, each reason without contact so thatLog in safer, it is to avoid occur forgetting Password and the consequence caused by reason such as password leakage.
In the present embodiment, in order to the vocal print feature of simple and fast resolving user, the vocal print feature includes：SoundLearn feature, lexical characteristics, prosodic features and channel information.From the angle that can be modeled using mathematical method, vocal print is automaticThe feature that identification model can be used at present includes：(1) acoustic feature (cepstrum)；(2) lexical characteristics (the related word of speakerN-gram, phoneme n-gram)；(3) prosodic features (fundamental tone and energy " posture " that are described using n-gram)；(4) channel information(which kind of passage used)；N-Gram is a kind of language model commonly used in large vocabulary continuous speech recognition.By using mathematics sideMethod is modeled, parsing voice messaging that can be quickly, so as to obtain the vocal print feature of user.
It is in the present embodiment, described that " cloud server is examined by carrying out vocal print and device id matching in database beyond the cloudsRope " includes：By in database beyond the clouds retrieve vocal print feature whether match, and device id whether be on-vehicle host equipmentOne in the device id of ID or mobile terminal.By the device id or mobile terminal that carry out vocal print feature and on-vehicle hostDevice id in one matched so that matching more accurately and quickly.Stepped on when there is user using new intelligent automobileWhen recording car-mounted terminal, by inputting ID, user profile is obtained by sound pick-up and the vocal print feature of acquisition user is parsed, andAnd the device id of the mobile device of acquisition user is logged in, and when making the car-mounted terminal of the new intelligent automobile of User logs in, is more pacifiedIt is complete and quick.And when user logs in the car-mounted terminal of intelligent automobile using new mobile terminal, by inputting ID, pass throughSound pick-up obtains user profile and parses the vocal print feature for obtaining user, and obtains the device id progress of the on-vehicle host of userLog in, it is safer and quick when making the car-mounted terminal of the new intelligent automobile of User logs in.
In the present embodiment, in order to reach more preferable recognition effect, when the cloud server is by user's registration, according toDefined content pronunciation carries out acquisition vocal print feature, and stores to cloud database.Occurred by the content of setting, such as" cloud degree automobile changes ordinary ", " cloud degree pursued vehicle is remarkable ", the content provided of " cloud degree bus arrival is farther " are sent outSound and then or vocal print feature, stored into cloud database, beyond the clouds in database, sound-groove model can be smart one by oneAccurate foundation, during identification, by pronouncing by defined content, can so reach more preferable recognition effect.
In the present embodiment, the use of user for convenience, and increase the alternative of user, the cloud server leads toWhen crossing user's registration, the content pronunciation set according to user carries out acquisition vocal print feature, and stores to cloud database.In registrationWhen, it can be pronounced without user according to the content of regulation, and the voice content voluntarily set by user is pronounced,It is modeled in cloud database so that user, which uses, to be more convenient, and increases the alternative of user.
Referring to Fig. 2, in another embodiment, a kind of intelligent automobile voice login system, including on-vehicle host, movementTerminal and cloud server；
The on-vehicle host is used to obtaining the device id of user speech information and on-vehicle host, and by voice messaging and equipmentID is sent to cloud server, and the on-vehicle host is additionally operable to perform the feedback information of cloud server；
The mobile terminal is used to obtaining the device id of user speech information and mobile terminal, and by voice messaging and movementThe device id of terminal is sent to cloud server；
The cloud server includes cloud database, and the cloud database stores the voice inputted during user's registrationCharacteristic value, device id, username and password, the cloud server are used to carry out voice messaging parsing acquisition vocal print feature,Vocal print feature and device id matching retrieval are carried out in database beyond the clouds, when vocal print feature and device id matching, cloud serviceDevice sends the feedback information for agreeing to log in on-vehicle host；When vocal print feature and device id are mismatched, then cloud server toOn-vehicle host sends the feedback information that refusal is logged in.
Vocal print not only has specificity, and the characteristics of have relative stability.After adult, the sound of people can keep long-termStablize relatively constant.It is demonstrated experimentally that no matter talker is deliberately to imitate other people sound and the tone, or whisper in sb.'s ear is softly talked, i.e.,Make to imitate remarkably true to life, its vocal print is but differed all the time.And Application on Voiceprint Recognition, one kind of biological identification technology.Application on Voiceprint Recognition is justIt is that acoustical signal is converted into electric signal, then is identified with computer.The generation of human language is Body Languages maincenter and pronunciationA complicated physiology physical process between organ, phonatory organ -- tongue, tooth, larynx, lung, nasal cavity that people uses in speechEveryone is widely different in terms of size and form, so the voiceprint map of any two people is all variant.
By the microphone on the microphone or mobile terminal in on-vehicle host obtain the voice messaging of user, andAnd after on-vehicle host or mobile terminal to voice messaging by parsing, vocal print feature is obtained, while obtaining vehicle-mounted masterThe device id of machine or the device id of mobile terminal, beyond the clouds the cloud database in server retrieved, to vocal print featureMatched with device id, when vocal print feature and device id while when matching, then cloud server feeds back to on-vehicle host progressLog in, conversely, when vocal print feature and device id are mismatched, then, cloud server feeds back to on-vehicle host refusal login.SoMatched by the vocal print feature of user, and the device id progress of the device id or mobile terminal by on-vehicle hostMatch somebody with somebody, increase is accurate, user-friendly without inputting user name and login password, and simultaneously because vocal print feature it is uniqueProperty, each reason without contact so that log in safer, it is to avoid occur forgetting Password with caused by the reason such as password leakageConsequence.
In the present embodiment, the vocal print feature includes：Acoustic feature, lexical characteristics, prosodic features and channel information.
It is in the present embodiment, described that " cloud server is examined by carrying out vocal print and device id matching in database beyond the cloudsRope " includes：By in database beyond the clouds retrieve vocal print feature whether match, and device id whether be on-vehicle host equipmentOne in the device id of ID or mobile terminal.When there is user using new intelligent automobile login car-mounted terminal, pass throughID is inputted, user profile is obtained by sound pick-up and the vocal print feature of acquisition user is parsed, and obtains the movement of userThe device id of equipment is logged in, safer and quick when making the car-mounted terminal of the new intelligent automobile of User logs in.And work as userWhen logging in the car-mounted terminal of intelligent automobile using new mobile terminal, by inputting ID, user is obtained by sound pick-up and believedThe vocal print feature for obtaining user is ceased and parsed, and obtains the device id of the on-vehicle host of user and is logged in, makes User logs inIt is safer and quick during the car-mounted terminal of new intelligent automobile.
In the present embodiment, when the cloud server is by user's registration, obtained according to the content pronunciation of regulationVocal print feature, and store to cloud database.
In the present embodiment, when the cloud server is by user's registration, the content set according to user pronounces to carry outVocal print feature is obtained, and is stored to cloud database.
In another embodiment intelligent automobile voice login system, including on-vehicle host, mobile terminal and cloud server；
The on-vehicle host is used to obtaining the device id of user speech information and on-vehicle host, and by voice messaging and equipmentID is sent to cloud server, and the on-vehicle host is additionally operable to perform the feedback information of cloud server；
The mobile terminal is used to obtaining the device id of user speech information and mobile terminal, and by voice messaging and movementThe device id of terminal is sent to cloud server；
The cloud server includes cloud database, and the cloud database stores the voice inputted during user's registrationCharacteristic value, device id, username and password, the cloud server are used to carry out voice messaging parsing acquisition vocal print feature,Vocal print feature and device id matching retrieval are carried out in database beyond the clouds, when vocal print feature and device id matching, cloud serviceDevice sends the feedback information for agreeing to log in on-vehicle host；When vocal print feature and device id are mismatched, then cloud server toOn-vehicle host sends the feedback information that refusal is logged in.
Vocal print not only has specificity, and the characteristics of have relative stability.After adult, the sound of people can keep long-termStablize relatively constant.It is demonstrated experimentally that no matter talker is deliberately to imitate other people sound and the tone, or whisper in sb.'s ear is softly talked, i.e.,Make to imitate remarkably true to life, its vocal print is but differed all the time.And Application on Voiceprint Recognition, one kind of biological identification technology.Application on Voiceprint Recognition is justIt is that acoustical signal is converted into electric signal, then is identified with computer.The generation of human language is Body Languages maincenter and pronunciationA complicated physiology physical process between organ, phonatory organ -- tongue, tooth, larynx, lung, nasal cavity that people uses in speechEveryone is widely different in terms of size and form, so the voiceprint map of any two people is all variant.
By the microphone on the microphone or mobile terminal in on-vehicle host obtain the voice messaging of user, andThe device id of on-vehicle host or the device id of mobile terminal are obtained, is sent to cloud server, beyond the clouds to language in serverMessage breath carries out parsing and obtains vocal print feature, and the cloud database in server is retrieved beyond the clouds, to vocal print feature withDevice id is matched, when vocal print feature and device id while when matching, then cloud server feeds back to on-vehicle host and stepped onRecord, conversely, when vocal print feature and device id are mismatched, then, cloud server feeds back to on-vehicle host refusal login.So lead toThe vocal print feature for crossing user is matched, and the device id or the device id of mobile terminal by on-vehicle host are matched,Increase is accurate, user-friendly without inputting user name and login password, and simultaneously because the uniqueness of vocal print feature,Each reason without contact so that log in safer, it is to avoid occur forgetting Password with caused by the reason such as password leakage afterReally.
In the present embodiment, the vocal print feature includes：Acoustic feature, lexical characteristics, prosodic features and channel information.
It is in the present embodiment, described that " cloud server is examined by carrying out vocal print and device id matching in database beyond the cloudsRope " includes：By in database beyond the clouds retrieve vocal print feature whether match, and device id whether be on-vehicle host equipmentOne in the device id of ID or mobile terminal.When there is user using new intelligent automobile login car-mounted terminal, pass throughID is inputted, user profile is obtained by sound pick-up and the vocal print feature of acquisition user is parsed, and obtains the movement of userThe device id of equipment is logged in, safer and quick when making the car-mounted terminal of the new intelligent automobile of User logs in.And work as userWhen logging in the car-mounted terminal of intelligent automobile using new mobile terminal, by inputting ID, user is obtained by sound pick-up and believedThe vocal print feature for obtaining user is ceased and parsed, and obtains the device id of the on-vehicle host of user and is logged in, makes User logs inIt is safer and quick during the car-mounted terminal of new intelligent automobile.In the present embodiment, the cloud server passes through user's registrationWhen, acquisition vocal print feature is carried out according to the content pronunciation of regulation, and store to cloud database.
In the present embodiment, when the cloud server is by user's registration, the content set according to user pronounces to carry outVocal print feature is obtained, and is stored to cloud database.
Although the various embodiments described above are described, those skilled in the art once know basic woundThe property made concept, then can make other change and modification to these embodiments, so embodiments of the invention are the foregoing is only,Not thereby the scope of patent protection of the present invention, the equivalent structure that every utilization description of the invention and accompanying drawing content are made are limitedOr equivalent flow conversion, or other related technical fields are directly or indirectly used in, similarly it is included in the patent of the present inventionWithin protection domain.