CN108877813A - The methods, devices and systems of man-machine identification - Google Patents

Abstract

The invention discloses a kind of methods, devices and systems of man-machine identification.Wherein, which includes：Terminal, for acquiring user's voiceprint and behavioral data information；Server, it is connect with terminal, for being identified according to the user's voiceprint and behavioral data information that prestore model and upload to terminal, judge whether operation object belonging to user's voiceprint and behavioral data information is identical as the target object prestored in model, type according to target object exports corresponding judging result, wherein, judging result includes：Operation object is natural person；Or, operation object is robot.The present invention is solved as being directed to defect existing for man-machine identification, the technical problem for causing man-machine recognition accuracy low in the prior art.

Description

The methods, devices and systems of man-machine identification

Technical field

The present invention relates to Internet technology application fields, a kind of method, apparatus in particular to man-machine identification and are System.

Background technique

Behavior, biological characteristic are commonly used in currently available technology as man-machine identifying schemes, are limited to operating platform and user Experience limitation, can collected user behavior data have serious limitation, the behavior that extracts from collected behavioral data is raw Object feature can bring information gain be restricted.

Man-machine identification is carried out for behavior identifying code principle at present, wherein whether man-machine identification is for distinguishing current operation Occur for machine or manual operation occurs, in the prior art during doing man-machine identification, firstly, acquisition user behavior Data；Second, extract the behavioural characteristic in behavioral data；It is last to carry out man-machine judgement according to behavior feature, as shown in Figure 1, Fig. 1 is man-machine identification process schematic diagram in the prior art.But there are following defects for existing man-machine identification：

First, it is different that different terminals web, pc, h5, andriod, ios etc. acquire behavioural characteristic.For example, same On personal mobile phone, user's operation dummy keyboard speed from tap speed in physical keyboard can different, use mouse sliding and direct Mobile phone screen carry out track sliding speed, pause also can be different.

Second, under limited action collection, unique degree of behavioural characteristic is not strong enough.When uniqueness is not strong enough, attacker is used When identical or very close data carry out Replay Attack, man-machine identifying system due to can not identify cause man-machine identifying system without Method refuses the multiple request attack of machine.

Third, under limited action collection, confirmation is not obtained in the stability of behavioural characteristic.Also, due to behavioural characteristic Simulated cost is lower, according to behavior biological characteristic, is difficult inversely to position natural person, causes attacker unbridled.

By taking wechat voiceprint lock as an example, when wechat voiceprint lock starts and is arranged, wechat client shows a string of characters, and shows Prompt, prompt instruction client press at screen instruction and then read aloud character, and wechat client reads aloud character by enrolling user Vocal print report server, the corresponding voiceprint lock of the user is generated by server, so as to later period user log in wechat when by should Voiceprint lock logs in wechat.

From the foregoing, it will be observed that if recording user voice sample by sound pick-up outfit, if attacker is logged in using the sample sound Wechat reads aloud the corresponding user voice sample of verifying character by player, then wechat client after receiving a string of verifying characters End also due to can not identify natural person operation or machine operation, cause attacker by using sample sound log in it is micro- Letter.

For above-mentioned as being directed to defect existing for man-machine identification in the prior art, cause man-machine recognition accuracy is low to ask Topic, currently no effective solution has been proposed.

Summary of the invention

The embodiment of the invention provides a kind of methods, devices and systems of man-machine identification, at least to solve due to existing skill For defect existing for man-machine identification in art, the technical problem for causing man-machine recognition accuracy low.

According to an aspect of an embodiment of the present invention, a kind of system of man-machine identification is provided, including：Terminal, for adopting Collect user's voiceprint and behavioral data information；Server is connect with terminal, prestores the use that model uploads terminal for foundation Family voiceprint and behavioral data information are identified, judge operation object belonging to user's voiceprint and behavioral data information Whether identical as the target object prestored in model, the type according to target object exports corresponding judging result, wherein judgement As a result include：Operation object is natural person；Or, operation object is robot.

Optionally, terminal includes：Show screen, antenna and sensor group, wherein display screen, it is to be verified for showing Information, wherein information to be verified includes：Acoustic information to be verified and behavioural information to be verified；Sensor group, for acquiring user According to the corresponding user's voiceprint of acoustic information to be verified and the corresponding behavioral data information of behavioural information to be verified；Antenna, It is connect with sensor group, for user's voiceprint and behavioral data information to be sent to server.

Optionally, server includes：Memory and processor, wherein memory prestores model for storing；Processor, For judging user's voiceprint and behavioral data information according to storage model identification user's voiceprint and behavioral data information It is whether identical as the target object prestored in model, it is in the case where the judgment result is yes, defeated if target object is natural person It is natural person that judging result, which is operation object belonging to user's voiceprint and behavioral data information, out；If target object is machine People, then it is robot that export judging result, which be operation object belonging to user's voiceprint and behavioral data information, and is alerted.

Further, optionally, prestoring model includes：Natural person's model and robot model, wherein natural person's model, For parsing user's voiceprint and behavioral data information, judge whether user's voiceprint meets default natural person's vocal print respectively Feature, and judge whether behavioral data information meets default natural person's behavioural characteristic, in user's voiceprint and behavioral data In the case that operation object belonging to information is natural person, all requests that terminal is sent are received；Robot model, for parsing User's voiceprint and behavioral data information, judge whether user's voiceprint meets default robot vocal print feature respectively, with And judge whether behavioral data information meets default robot behavior feature, belonging to user's voiceprint and behavioral data information Operation object be robot in the case where, refusal processing terminal send all requests.

Optionally, processor is also used in judging result be operation belonging to user's voiceprint and behavioral data information In the case that object is natural person, whether judgement receives the corresponding user's voiceprint of operation object and row within a preset time It is to receive the corresponding user's voiceprint of operation object and behavioral data within a preset time in judging result for data information In the case where information, all requests of refusal processing terminal transmission.

Optionally, server further includes：Alarm, wherein alarm is connect with processor, for being in judging result In the case that operation object belonging to user's voiceprint and behavioral data information is robot, outputting alarm signal, so that Processor refuses all requests that processing terminal is sent.

According to another aspect of an embodiment of the present invention, a kind of man-machine knowledge method for distinguishing is additionally provided, including：Receive terminal hair The checking request sent；Information to be verified is sent to terminal, wherein information to be verified includes：Acoustic information to be verified and to be verified Behavioural information；Receive the user's voiceprint and behavioral data information of the correspondence information to be verified that terminal uploads；Judge user's sound Whether operation object belonging to line information and behavioral data information is identical as the target object prestored in model, and according to target pair The type of elephant exports corresponding judging result, wherein judging result includes：Operation object is natural person；Or, operation object is machine Device people.

Optionally, judge operation object belonging to user's voiceprint and behavioral data information whether with prestore in model Target object is identical to include：By prestoring model analyzing user voiceprint and behavioral data information, user's voiceprint is obtained In user's vocal print feature, user read aloud content information and user behavior characteristics；Judgement state user read aloud content information whether with Acoustic information to be verified in information to be verified is identical, and judge user behavior characteristics whether with it is to be tested in information to be verified It is identical to demonstrate,prove behavioural information；Include natural person and robot in target object, and reads aloud content information and user behavior spy in user It levies in situation identical with acoustic information to be verified and behavioural information to be verified, judges user's vocal print feature by prestoring model It whether is natural person's vocal print feature.

Further, optionally, user's vocal print feature includes：User voice frequency spectrum, user voice loudness and volume；User Behavioural characteristic includes：Touch track and touching frequency.

Optionally, exporting corresponding judging result according to the type of target object includes：If judge user's vocal print feature and Operation object belonging to user behavior characteristics is natural person, then exporting judging result is user's voiceprint and behavioral data information Affiliated operation object is natural person；If judging operation object belonging to user's vocal print feature and user behavior characteristics for machine People, then it is robot that export judging result, which be operation object belonging to user's voiceprint and behavioral data information,.

Further, optionally, judging operation object belonging to user's vocal print feature and user behavior characteristics for nature After people, method further includes：In the case that the operation object belonging to user's voiceprint and behavioral data information is natural person, Whether judgement receives the corresponding user's voiceprint of operation object and behavioral data information within a preset time；In judging result In the case where receiving the corresponding user's voiceprint of operation object and behavioral data information within a preset time, refusal processing The checking request that terminal is sent；It is not receive the corresponding user's voiceprint of operation object within a preset time in judging result In the case where behavioral data information, the checking request of processing terminal transmission.

It optionally, be operation object belonging to user's voiceprint and behavioral data information is machine in output judging result After people, method further includes：Outputting alarm signal, and refuse the checking request of processing terminal transmission.

According to another aspect of an embodiment of the present invention, another man-machine knowledge method for distinguishing is additionally provided, including：To server Send checking request；Receive the information to be verified that server is returned according to checking request, wherein information to be verified includes：It is to be tested Demonstrate,prove acoustic information and behavioural information to be verified；Acquire corresponding with acoustic information to be verified user's voiceprint, and acquire to Verify the corresponding behavioral data information of behavioural information；User's voiceprint and behavioral data information are sent to server.

Optionally, user's voiceprint corresponding with acoustic information to be verified, and acquisition behavioural information to be verified are acquired Corresponding behavioral data information includes：It acquires voice content when user reads aloud acoustic information to be verified, user voice frequency spectrum, use Family sound intensity and volume；Touching track and touching frequency when acquisition user only needs behavioural information to be verified.

Another aspect according to an embodiment of the present invention additionally provides a kind of device of man-machine identification, including：First receives mould Block, for receiving the checking request of terminal transmission；Sending module, for sending information to be verified to terminal, wherein letter to be verified Breath includes：Acoustic information to be verified and behavioural information to be verified；Second receiving module, the correspondence for receiving terminal upload are to be tested Demonstrate,prove the user's voiceprint and behavioral data information of information；Identification module, for judging user's voiceprint and behavioral data letter Whether operation object belonging to breath is identical as the target object prestored in model, and corresponding according to the output of the type of target object Judging result, wherein judging result includes：Operation object is natural person；Or, operation object is robot.

Another aspect according to an embodiment of the present invention additionally provides the device of another man-machine identification, including：First sends Module, for sending checking request to server；Receiving module, for receiving server according to the to be verified of checking request return Information, wherein information to be verified includes：Acoustic information to be verified and behavioural information to be verified；Acquisition module, for acquire with to Verify the corresponding user's voiceprint of acoustic information, and the corresponding behavioral data information of acquisition behavioural information to be verified；Second Sending module, for sending user's voiceprint and behavioral data information to server.

It is according to an embodiment of the present invention in another aspect, additionally provide a kind of storage medium, storage medium includes the journey of storage Sequence, wherein equipment where control storage medium executes above-mentioned man-machine knowledge method for distinguishing in program operation.

It is according to an embodiment of the present invention that in another aspect, additionally providing a kind of processor, processor is used to run program, In, program executes above-mentioned man-machine knowledge method for distinguishing when running.

In embodiments of the present invention, by terminal, for acquiring user's voiceprint and behavioral data information；Server, It connect with terminal, for being identified according to the user's voiceprint and behavioral data information that prestore model and upload to terminal, sentences Whether operation object belonging to disconnected user's voiceprint and behavioral data information is identical as the target object prestored in model, foundation The type of target object exports corresponding judging result, wherein judging result includes：Operation object is natural person；Or, operation pair As for robot, having achieved the purpose that protect machine request attack, imitated to realize and promote the technology of man-machine recognition accuracy Fruit, and then solve as being directed to defect existing for man-machine identification, the technology for causing man-machine recognition accuracy low in the prior art Problem.

Detailed description of the invention

The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings：

Fig. 1 is man-machine identification process schematic diagram in the prior art；

Fig. 2 is the structural schematic diagram according to the system of the man-machine identification of the embodiment of the present application one；

Fig. 3 is the execution flow chart according to the system of the man-machine identification of the embodiment of the present application one；

Fig. 4 is a kind of hardware block diagram of the server of man-machine knowledge method for distinguishing of the embodiment of the present application

Fig. 5 is the flow chart according to the man-machine knowledge method for distinguishing of the embodiment of the present application two；

Fig. 6 is the flow chart according to the man-machine knowledge method for distinguishing of the embodiment of the present application three；

Fig. 7 is the schematic diagram that terminal is shown in the man-machine knowledge method for distinguishing according to the embodiment of the present application three；

Fig. 8 is the structural schematic diagram according to the device of the man-machine identification of the embodiment of the present application four；

Fig. 9 is the structural schematic diagram according to the device of the man-machine identification of the embodiment of the present application five.

Specific embodiment

In order to make those skilled in the art more fully understand application scheme, below in conjunction in the embodiment of the present application Attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only The embodiment of the application a part, instead of all the embodiments.Based on the embodiment in the application, ordinary skill people Member's every other embodiment obtained without making creative work, all should belong to the model of the application protection It encloses.

It should be noted that the description and claims of this application and term " first " in above-mentioned attached drawing, " Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way Data be interchangeable under appropriate circumstances, so as to embodiments herein described herein can in addition to illustrating herein or Sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that cover Cover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to Step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, product Or other step or units that equipment is intrinsic.

This application involves technical term

Replay Attack：In the man-machine identification scene using biological characteristic, attacker can pass through certain tool (such as key It is smart) etc. record data of normal people, then carry out multiple attack；

Uniqueness：The biological characteristic of different people has enough othernesses；

Stability：Biological characteristic when different of the same person changes very little.

Embodiment 1

This application provides the systems of man-machine identification as shown in Figure 2.Fig. 2 is the man-machine knowledge according to the embodiment of the present application one The structural schematic diagram of other system, as shown in Fig. 2, the system of the man-machine identification includes：Terminal 22 and server 24,

Terminal 22, for acquiring user's voiceprint and behavioral data information；Server 24 connect with terminal 22, is used for Identified according to the user's voiceprint and behavioral data information for prestoring model and uploading to terminal, judge user's voiceprint with Whether operation object belonging to behavioral data information is identical as the target object prestored in model, and the type according to target object is defeated Corresponding judging result out, wherein judging result includes：Operation object is natural person；Or, operation object is robot.

Specifically, terminal 22 may include in the system of man-machine identification provided by the present application here：Smart phone (such as： Be applicable in the mobile phone of IOS system, android system or Windows system), tablet computer, laptop, PC The commercial affairs of (Personal Computer, abbreviation PC), palm or intelligent wearable device (such as Google glass, smartwatch, enhancing Real AR equipment or Virtual Reality equipment) etc. have the mobile terminal of communication and data processing, storage capacity.Server 24 is logical It crosses wired or wireless mode to connect with terminal 22, establishes communication link.

Wherein, the system of man-machine identification provided by the embodiments of the present application is said for being suitable for authentication environment Bright, terminal 22 is (subsequent to be referred to as with smart phone：Mobile phone) for be illustrated, user using mobile phone log in APP when, defeated Enter account and during password waits for APP verifying, if account and password are inconsistent, APP shows login failure, multiple in user In the case where the equal login failure of typing account number cipher, APP shows password retrieval function, prompts user in such a way that password is given for change Again it logs in, after mobile phone detects that user clicks password retrieval function item, mobile phone sends checking request, server to server After receiving checking request, information to be verified is sent to mobile phone, wherein the information to be verified includes：Acoustic information to be verified and Behavioural information to be verified；Mobile phone after receiving the information to be verified, by the sensor that is connect in mobile phone with display screen and Microphone acquires user's voiceprint and behavioral data information, and after acquisition user's voiceprint and behavioral data information, User's voiceprint and behavioral data information are uploaded to server, so that server is according to user's voiceprint and behavior number It is believed that breath is identified, judge that user's voiceprint and behavioral data information operate generation or robot manipulation for natural person It generates；It is generated if robot manipulation, then forbids the checking request of mobile phone, password, which gives process for change, to be terminated；And it is when recognizing When natural person operates generation, the checking request of mobile phone is received, user enters password resetting process, and resets by password, again Log in APP.

For example, the information to be verified that mobile phone receives includes：Character string 123456 and " pin current mark bright area, it is bright Character string is read, and slides the sliding block for marking bright area from left to right after reading aloud ", at this point, the microphone detection of mobile phone is It is no to collect sound, and detect and show whether the sliding block in mark bright area slides from left to right in mobile phone display screen；When mobile phone is aobvious When display screen detects pressing mark bright area, microphone detection receives the sound that user reads aloud 123456；And it is used in microphone detection After 123456 end are read aloud at family, mobile phone display screen detects that the sliding block of mark bright area slides from left to right, and mobile phone will collect " 123456 " acoustic information and " pressing mark bright area and mobile phone display screen detect mark bright area sliding block from left to right The behavioral data information of sliding " is uploaded to server, and server identifies " 123456 " acoustic information and " pressing by prestoring model Mark bright area and mobile phone display screen detect that the sliding block of mark bright area slides from left to right " behavioral data information, judgement adopts " 123456 " acoustic information that collects and " pressing mark bright area and mobile phone display screen detect the sliding block of mark bright area by a left side Whether the behavioral data information of sliding to the right " is that robot is sent, wherein target object includes in the present embodiment：Natural person and Robot, when detecting " 123456 " acoustic information and " pressing mark bright area and mobile phone display screen detect mark bright area Sliding block slides from left to right " behavioral data information be natural person operate generate when, receive mobile phone send checking request；Work as inspection It measures " 123456 " acoustic information and " pressing mark bright area and mobile phone display screen detects the sliding block of mark bright area by left-hand When the behavioral data information of right sliding " is that robot manipulation generates, the checking request that the termination mobile phone is sent, and refusing should All requests of the subsequent transmission of mobile phone.To avoid the generation of Replay Attack.

It should be noted why the present embodiment will identify together it is due to each by voiceprint and behavioural information The uniqueness of voice line since the immutable and uniqueness of vocal print has determined the stability of voice print verification, and passes through behavior The stability of data comprehensively considers promotion verifying stability, and generates since behavioural characteristic is easy to be imitated by robot, passes through The double standards of vocal print feature and behavioural characteristic, the user's voiceprint and behavioral data information upload to terminal side are known Not, the accuracy rate of man-machine identification is improved.

And it is worth noting that not needing the vocal print history note for saving any user in server side in the present embodiment Record is connect only by prestoring the learning ability of model by obtaining the vocal print feature identification of vocal print feature and robot of natural person The user's voiceprint received, and the behavioural characteristic by prestoring the identification of the behavioural characteristic in the model good robot of natural person, The storage pressure for alleviating server, has saved the system resource of server, has improved operation efficiency.

Here the application above-mentioned example is only subject to the system for the man-machine identification for realizing that the present embodiment passes through, and does not limit specifically It is fixed.

In the embodiment of the present application, by terminal, for acquiring user's voiceprint and behavioral data information；Server, It connect with terminal, for being identified according to the user's voiceprint and behavioral data information that prestore model and upload to terminal, sentences Whether operation object belonging to disconnected user's voiceprint and behavioral data information is identical as the target object prestored in model, foundation The type of target object exports corresponding judging result, wherein judging result includes：Operation object is natural person；Or, operation pair As for robot, having achieved the purpose that protect machine request attack, imitated to realize and promote the technology of man-machine recognition accuracy Fruit, and then solve as being directed to defect existing for man-machine identification, the technology for causing man-machine recognition accuracy low in the prior art Problem.

Specifically, the system of man-machine identification provided in this embodiment is specific as follows during realizing man-machine identification：

Optionally, terminal 22 includes：Show screen, antenna and sensor group, wherein display screen, it is to be tested for showing Demonstrate,prove information, wherein information to be verified includes：Acoustic information to be verified and behavioural information to be verified；Sensor group is used for acquiring Family is according to the corresponding user's voiceprint of acoustic information to be verified and the corresponding behavioral data information of behavioural information to be verified；It Line is connect with sensor group, for user's voiceprint and behavioral data information to be sent to server 24.

Specifically, being still illustrated taking the above example as an example, by taking password is given for change as an example, initiated in user's using terminal 22 In the case that password is given for change, terminal 22 sends password to server 24 and gives request for change, receives what server 24 was sent by antenna Information to be verified, the user's voiceprint and behavioral data that the sensor group acquisition user of terminal 22 generates according to information to be verified User's voiceprint and behavioral data information finally will be sent to server 24 by antenna by information.

Wherein, the sensor group of terminal 22 may include in the present embodiment：Microphone, pressure sensor, track sensing The sensors such as device, gyroscope, for acquiring the behavior of user voice and user.

Optionally, server 24 includes：Memory and processor, wherein memory prestores model for storing；Processing Device, for judging user's voiceprint and behavioral data according to storage model identification user's voiceprint and behavioral data information Whether information is identical as the target object prestored in model, in the case where the judgment result is yes, if target object is natural person, It is natural person that then to export judging result, which be operation object belonging to user's voiceprint and behavioral data information,；If target object is Robot, then it is robot that export judging result, which be operation object belonging to user's voiceprint and behavioral data information, and is accused It is alert.

Specifically, server 24 includes memory and processor in the system of man-machine identification provided in this embodiment, In, memory storage prestores model, and processor prestores the user's voiceprint and behavior number of model identification terminal upload by this It is believed that breath, and model identifies user's voiceprint and whether the affiliated operation object of behavioral data information is to prestore in model by prestoring Target object, if the operation object is that the target object prestored in model exports if the target object is natural person Judging result is that operation object belonging to user's voiceprint and behavioral data information is natural person；If target object is machine People, then it is robot that export judging result, which be operation object belonging to user's voiceprint and behavioral data information, and is alerted.

Further, optionally, prestoring model includes：Natural person's model and robot model, wherein natural person's model, For parsing user's voiceprint and behavioral data information, judge whether user's voiceprint meets default natural person's vocal print respectively Feature, and judge whether behavioral data information meets default natural person's behavioural characteristic, in user's voiceprint and behavioral data In the case that operation object belonging to information is natural person, all requests that terminal is sent are received；Robot model, for parsing User's voiceprint and behavioral data information, judge whether user's voiceprint meets default robot vocal print feature respectively, with And judge whether behavioral data information meets default robot behavior feature, belonging to user's voiceprint and behavioral data information Operation object be robot in the case where, refusal processing terminal send all requests.

Specifically, setting up natural person's model and robot model in prestoring model, natural person's model is stored with nature People's vocal print feature and natural person's behavioural characteristic, similarly, in robot model are stored with robot vocal print feature and robot row It is characterized, during the user's voiceprint and behavioral data information that identification terminal 22 uploads, server 24 is pre- by this Deposit user's voiceprint that the other terminal 22 of natural person's model and robot model in model uploads and behavioral data information be by Natural person generates or is generated by robot, while after obtaining judging result, corresponding natural person and robot configured with pair The processing strategie answered, for example, in the case that the operation object belonging to user's voiceprint and behavioral data information is natural person, Receive all requests that terminal is sent；The operation object belonging to user's voiceprint and behavioral data information is the feelings of robot Under condition, all requests of refusal processing terminal transmission.

Optionally, processor is also used in judging result be operation belonging to user's voiceprint and behavioral data information In the case that object is natural person, whether judgement receives the corresponding user's voiceprint of operation object and row within a preset time It is to receive the corresponding user's voiceprint of operation object and behavioral data within a preset time in judging result for data information In the case where information, all requests of refusal processing terminal transmission.

Specifically, copying user's sound by sample sound in order to avoid the sample sound that machine records user (natural person) The case where line information, occurs, and due to initiating to request if it is robot, robot can initiate repeatedly to request (example in a short time Thousands of requests such as are sent in 1 minute, server 24 will be considered that attacker initiates Replay Attack at this time), therefore server 24 Processor receive terminal 22 upload user's voiceprint and behavioral data information after, in addition to identifying user's vocal print Information and behavioral data information are that except natural person generates, can also detect and whether repeatedly receive the terminal 22 within a preset time The request of transmission, or, having the identification record of the natural person in the historical record of processor, if there is the identification of the natural person It records or detects and have received the request that the multiple terminal 22 is sent in a short time, then it is assumed that the request that the terminal 22 is sent is What robot initiated, it is corresponding that refusal is handled to all requests that the terminal is sent.Wherein, the short time in the present embodiment can be with Including the detection in 1 minute, the present embodiment is only illustrated for 1 minute, to realize man-machine identification provided in this embodiment System subject to, specifically without limitation.

Optionally, server 24 further includes：Alarm, wherein alarm is connect with processor, in judging result In the case where being robot for operation object belonging to user's voiceprint and behavioral data information, outputting alarm signal, so that Obtain all requests that processor refusal processing terminal is sent.

Specifically, further including alarm in server 24 provided in this embodiment, which connect with processor, is sentencing Disconnected result is outputting alarm letter in the case that operation object belonging to user's voiceprint and behavioral data information is robot Number, and 22 typing of terminal is alerted and is recorded, so that server 24 is refused in the request that the subsequently received terminal 22 is sent The request of terminal transmission is handled absolutely.

To sum up, the system of man-machine identification provided in this embodiment, as shown in figure 3, Fig. 3 is according to the embodiment of the present application one The execution flow chart of the system of man-machine identification.

Step1, server send information to be verified, which is used to indicate user and reads aloud in information to be verified Character string, and corresponding behavior operation is executed according to information to be verified, which may include：Number, letter, text One of word, picture or audio or at least two combination.

Step2, audio and behavioral data of the client of terminal according to the information collection user to be verified, and by user's Audio and behavioral data are sent to server；

Step3, the audio and behavioral data that server judges user are (that is, user's voiceprint and row in the present embodiment For data information) whether meet biological characteristic (that is, prestoring the natural person's model and robot mould of model storage in the present embodiment Type judges whether the audio of user and behavioral data meet biological characteristic by above-mentioned model)；It is yes in judging result In the case of execute Step4, execute Step5 if the determination result is NO；

Step4, judges whether the vocal print meets biological characteristic, in the case where the judgment result is yes, executes Step6, Judging result be it is no in the case where execute Step5；

Step5, the request are the request that robot initiates, and belong to query-attack；

Step6 judges whether it is Replay Attack, in the case where the judgment result is yes, Step7 is executed, in judging result Be it is no in the case where execute Step5；

Step7, server determine the request for normal request.

Embodiment 2

According to the embodiment of the present application, a kind of embodiment of the method for man-machine identification is additionally provided, it should be noted that in attached drawing Process the step of illustrating can execute in a computer system such as a set of computer executable instructions, although also, Logical order is shown in flow charts, but in some cases, can be executed with the sequence for being different from herein it is shown or The step of description.

Embodiment of the method provided by the embodiment of the present application one can be in mobile terminal, terminal or similar fortune It calculates and is executed in device.For running on the server, Fig. 4 is a kind of service of man-machine knowledge method for distinguishing of the embodiment of the present application The hardware block diagram of device.As shown in figure 4, server 40 may include one or more (only showing one in figure) processors 402 (processing units that processor 402 can include but is not limited to Micro-processor MCV or programmable logic device FPGA etc.) are used Memory 404 in storing data and the transmission module 406 for communication function.Those of ordinary skill in the art can manage Solution, structure shown in Fig. 4 are only to illustrate, and do not cause to limit to the structure of above-mentioned electronic device.For example, server 40 may be used also Including component more perhaps more less than shown in Fig. 4 or with the configuration different from shown in Fig. 4.

Memory 404 can be used for storing the software program and module of application software, as man-machine in the embodiment of the present application Know the corresponding program instruction/module of method for distinguishing, processor 402 by software program that operation is stored in memory 404 with And module realizes the man-machine knowledge method for distinguishing of above-mentioned application program thereby executing various function application and data processing. Memory 404 may include high speed random access memory, may also include nonvolatile memory, as one or more magnetic storage fills It sets, flash memory or other non-volatile solid state memories.In some instances, memory 404 can further comprise relative to place The remotely located memory of device 402 is managed, these remote memories can pass through network connection to server 40.The reality of above-mentioned network Example includes but is not limited to internet, intranet, local area network, mobile radio communication and combinations thereof.

Transmitting device 406 is used to that data to be received or sent via a network.Above-mentioned network specific example may include The wireless network that the communication providers of server 40 provide.In an example, transmitting device 406 includes a network adapter (Network Interface Controller, NIC), can be connected by base station with other network equipments so as to interconnection Net is communicated.In an example, transmitting device 406 can be radio frequency (Radio Frequency, RF) module, be used for Wirelessly communicated with internet.

Under above-mentioned running environment, this application provides man-machine knowledge method for distinguishing as shown in Figure 5.In server side, Fig. 5 It is the flow chart according to the man-machine knowledge method for distinguishing of the embodiment of the present application two.

Step S502 receives the checking request that terminal is sent；

Step S504 sends information to be verified to terminal, wherein information to be verified includes：Acoustic information to be verified and to Verify behavioural information；

Step S506 receives the user's voiceprint and behavioral data information of the correspondence information to be verified that terminal uploads；

Step S508, judge operation object belonging to user's voiceprint and behavioral data information whether with prestore in model Target object it is identical, and export corresponding judging result according to the type of target object, wherein judging result includes：Operation Object is natural person；Or, operation object is robot.

Specifically, corresponding server 24 in 1, man-machine knowledge method for distinguishing provided in this embodiment can fit in conjunction with the embodiments For in server, man-machine knowledge method for distinguishing provided by the embodiments of the present application to be said for being suitable for authentication environment Bright, terminal is (subsequent to be referred to as with smart phone：Mobile phone) for be illustrated, user using mobile phone log in APP when, inputting During account and password wait for APP verifying, if account and password are inconsistent, APP shows login failure, repeatedly records in user In the case where entering the equal login failure of account number cipher, APP shows password retrieval function, prompts user's weight in such a way that password is given for change New login, after mobile phone detects that user clicks password retrieval function item, mobile phone sends checking request to server, and server exists After receiving checking request, information to be verified is sent to mobile phone, wherein the information to be verified includes：Acoustic information to be verified and to Verify behavioural information；Mobile phone passes through the sensor connecting in mobile phone with display screen and wheat after receiving the information to be verified Gram elegance collection user voiceprint and behavioral data information, and after acquisition user's voiceprint and behavioral data information, it will User's voiceprint and behavioral data information are uploaded to server, so that server is according to user's voiceprint and behavioral data Information is identified, judges that user's voiceprint and behavioral data information operate generation for natural person or robot manipulation produces It is raw；It is generated if robot manipulation, then forbids the checking request of mobile phone, password, which gives process for change, to be terminated；It and is certainly when recognizing When right people operates generation, the checking request of mobile phone is received, user enters password resetting process, and resets by password, steps on again Land APP.

For example, the information to be verified that mobile phone receives includes：Character string 123456 and " pin current mark bright area, it is bright Character string is read, and slides the sliding block for marking bright area from left to right after reading aloud ", at this point, the microphone detection of mobile phone is It is no to collect sound, and detect and show whether the sliding block in mark bright area slides from left to right in mobile phone display screen；When mobile phone is aobvious When display screen detects pressing mark bright area, microphone detection receives the sound that user reads aloud 123456；And it is used in microphone detection After 123456 end are read aloud at family, mobile phone display screen detects that the sliding block of mark bright area slides from left to right, and mobile phone will collect " 123456 " acoustic information and " pressing mark bright area and mobile phone display screen detect mark bright area sliding block from left to right The behavioral data information of sliding " is uploaded to server, and server identifies " 123456 " acoustic information and " pressing by prestoring model Mark bright area and mobile phone display screen detect that the sliding block of mark bright area slides from left to right " behavioral data information, judgement adopts " 123456 " acoustic information that collects and " pressing mark bright area and mobile phone display screen detect the sliding block of mark bright area by a left side Whether the behavioral data information of sliding to the right " is that robot is sent, wherein target object includes in the present embodiment：Natural person and Robot, when detecting " 123456 " acoustic information and " pressing mark bright area and mobile phone display screen detect mark bright area Sliding block slides from left to right " behavioral data information be natural person operate generate when, receive mobile phone send checking request；Work as inspection It measures " 123456 " acoustic information and " pressing mark bright area and mobile phone display screen detects the sliding block of mark bright area by left-hand When the behavioral data information of right sliding " is that robot manipulation generates, the checking request that the termination mobile phone is sent, and refusing should All requests of the subsequent transmission of mobile phone.To avoid the generation of Replay Attack.

It should be noted why the present embodiment will identify together it is due to each by voiceprint and behavioural information The uniqueness of voice line since the immutable and uniqueness of vocal print has determined the stability of voice print verification, and passes through behavior The stability of data comprehensively considers promotion verifying stability, and generates since behavioural characteristic is easy to be imitated by robot, passes through The double standards of vocal print feature and behavioural characteristic, the user's voiceprint and behavioral data information upload to terminal side are known Not, the accuracy rate of man-machine identification is improved.

And it is worth noting that not needing the vocal print history note for saving any user in server side in the present embodiment Record is connect only by prestoring the learning ability of model by obtaining the vocal print feature identification of vocal print feature and robot of natural person The user's voiceprint received, and the behavioural characteristic by prestoring the identification of the behavioural characteristic in the model good robot of natural person, The storage pressure for alleviating server, has saved the system resource of server, has improved operation efficiency.

Here the application above-mentioned example is only subject to the man-machine knowledge method for distinguishing for realizing that the present embodiment passes through, and does not limit specifically It is fixed.

In the embodiment of the present application, the checking request sent by receiving terminal；Information to be verified is sent to terminal, In, information to be verified includes：Acoustic information to be verified and behavioural information to be verified；Receive the correspondence information to be verified that terminal uploads User's voiceprint and behavioral data information；Whether judge operation object belonging to user's voiceprint and behavioral data information It is identical as the target object prestored in model, and corresponding judging result is exported according to the type of target object, wherein judgement knot Fruit includes：Operation object is natural person；Or, operation object is robot, achieve the purpose that protect machine request attack, thus The technical effect for promoting man-machine recognition accuracy is realized, and then is solved as being directed to existing for man-machine identification in the prior art Defect, the technical problem for causing man-machine recognition accuracy low.

Optionally, judge in step S508 operation object belonging to user's voiceprint and behavioral data information whether in advance Deposit that the target object in model is identical to include：

Step1 is obtained in user's voiceprint by prestoring model analyzing user voiceprint and behavioral data information User's vocal print feature, user read aloud content information and user behavior characteristics；

Step2, whether user in judgement to read aloud content information identical as the acoustic information to be verified in information to be verified if being stated, with And judge whether user behavior characteristics are identical as the behavioural information to be verified in information to be verified；

Step3 includes natural person and robot in target object, and reads aloud content information and user behavior characteristics in user In situation identical with acoustic information to be verified and behavioural information to be verified, judge that user's vocal print feature is by prestoring model No is natural person's vocal print feature.

Further, optionally, user's vocal print feature includes：User voice frequency spectrum, user voice loudness and volume；User Behavioural characteristic includes：Touch track and touching frequency.

It should be noted that man-machine knowledge method for distinguishing provided by the present application only includes with user's vocal print feature：User voice Frequency spectrum, user voice loudness and volume；User behavior characteristics include：It is illustrated for touching track and touching frequency, with reality Subject to existing man-machine knowledge method for distinguishing provided in this embodiment, specifically without limitation.

Optionally, exporting corresponding judging result according to the type of target object in step S508 includes：

Step4, if judging operation object belonging to user's vocal print feature and user behavior characteristics for natural person, output is sentenced Disconnected result is that operation object belonging to user's voiceprint and behavioral data information is natural person；

Step5, if judging operation object belonging to user's vocal print feature and user behavior characteristics for robot, output is sentenced Disconnected result is that operation object belonging to user's voiceprint and behavioral data information is robot.

Further, optionally, judge that operation object belonging to user's vocal print feature and user behavior characteristics is in Step4 After natural person, man-machine knowledge method for distinguishing provided by the embodiments of the present application further includes：

Step 1, in the case that the operation object belonging to user's voiceprint and behavioral data information is natural person, judgement The corresponding user's voiceprint of operation object and behavioral data information whether are received within a preset time；

It step 2, is to receive the corresponding user's voiceprint of operation object and behavior within a preset time in judging result In the case where data information, the checking request of refusal processing terminal transmission；

It step 3, is not receive the corresponding user's voiceprint of operation object and row within a preset time in judging result In the case where data information, the checking request of processing terminal transmission.

Specifically, server includes memory and processor, wherein memory storage prestores model, and processor is by being somebody's turn to do The user's voiceprint and behavioral data information of the upload of model identification terminal are prestored, and by prestoring model identification user's voiceprint It whether is the target object prestored in model with the affiliated operation object of behavioral data information, if the operation object is to prestore model In target object, if the target object is natural person, exporting judging result is that user's voiceprint and behavioral data are believed Operation object belonging to breath is natural person；If target object is robot, exporting judging result is user's voiceprint and row It is robot for operation object belonging to data information.

It optionally, is operation object belonging to user's voiceprint and behavioral data information in Step5 output judging result After robot, man-machine knowledge method for distinguishing provided by the embodiments of the present application further includes：

Step6, outputting alarm signal, and refuse the checking request of processing terminal transmission.

Specifically, further including alarm in server provided in this embodiment, which connect with processor, is judging It as a result is outputting alarm signal in the case that operation object belonging to user's voiceprint and behavioral data information is robot, And the terminal typing is alerted and is recorded, so that in the request that the subsequently received terminal 22 is sent, the processing of server refusal The request that the terminal is sent.

Embodiment 3

This application provides man-machine knowledge method for distinguishing as shown in FIG. 6.In terminal side, Fig. 6 is according to the embodiment of the present application The flow chart of three man-machine knowledge method for distinguishing.

Step S602 sends checking request to server；

Step S604 receives the information to be verified that server is returned according to checking request, wherein information to be verified includes： Acoustic information to be verified and behavioural information to be verified；

Step S606 acquires user's voiceprint corresponding with acoustic information to be verified, and acquisition behavior letter to be verified Cease corresponding behavioral data information；

Step S608 sends user's voiceprint and behavioral data information to server.

Specifically, the terminal in corresponding embodiment 1, man-machine knowledge method for distinguishing provided in this embodiment can be adapted for terminal Side, terminal may include in man-machine knowledge method for distinguishing provided by the present application：Smart phone (such as：It is applicable in IOS system, Android The mobile phone of system or Windows system), tablet computer, laptop, PC (PersonalComputer, referred to as PC), (such as Google glass, smartwatch, augmented reality AR equipment or Virtual Reality are set for palm commercial affairs or intelligent wearable device It is standby) etc. have the mobile terminal of communication and data processing, storage capacity.Server 24 passes through wired or wireless mode and terminal Connection, establishes communication link.

Wherein, man-machine knowledge method for distinguishing provided by the embodiments of the present application is said for being suitable for authentication environment Bright, terminal 22 is (subsequent to be referred to as with smart phone：Mobile phone) for be illustrated, user using mobile phone log in APP when, defeated Enter account and during password waits for APP verifying, if account and password are inconsistent, APP shows login failure, multiple in user In the case where the equal login failure of typing account number cipher, APP shows password retrieval function, prompts user in such a way that password is given for change Again it logs in, after mobile phone detects that user clicks password retrieval function item, mobile phone sends checking request, server to server After receiving checking request, information to be verified is sent to mobile phone, wherein the information to be verified includes：Acoustic information to be verified and Behavioural information to be verified；Mobile phone after receiving the information to be verified, by the sensor that is connect in mobile phone with display screen and Microphone acquires user's voiceprint and behavioral data information, and after acquisition user's voiceprint and behavioral data information, User's voiceprint and behavioral data information are uploaded to server.

For example, terminal to server sends password and gives request for change in the case where user's using terminal initiation password is given for change, The information to be verified that server is sent is received by antenna, what the sensor group acquisition user of terminal generated according to information to be verified User's voiceprint and behavioral data information finally will be sent to service by antenna by user's voiceprint and behavioral data information Device.

In the embodiment of the present application, by sending checking request to server；Server is received to return according to checking request Information to be verified, wherein information to be verified includes：Acoustic information to be verified and behavioural information to be verified；Acquisition with it is to be verified The corresponding user's voiceprint of acoustic information, and the corresponding behavioral data information of acquisition behavioural information to be verified；To server User's voiceprint and behavioral data information are sent, has achieved the purpose that protect machine request attack, has promoted people to realize The technical effect of machine recognition accuracy, and then solve as being directed to defect existing for man-machine identification in the prior art, lead to people The low technical problem of machine recognition accuracy.

Optionally, user's voiceprint corresponding with acoustic information to be verified, and acquisition behavioural information to be verified are acquired Corresponding behavioral data information includes：It acquires voice content when user reads aloud acoustic information to be verified, user voice frequency spectrum, use Family sound intensity and volume；Touching track and touching frequency when acquisition user only needs behavioural information to be verified.

Specifically, as shown in fig. 7, Fig. 7 is that terminal is shown in the man-machine knowledge method for distinguishing according to the embodiment of the present application three Schematic diagram.After the information to be verified that terminal receives that server is sent, the character in information to be verified is shown, and need user Mike's icon of pressing, and then after detecting that user presses Mike's icon, detect that user reads aloud the sound of character, terminal Sound and pressing operation are uploaded to the verifying of server waiting for server, device to be serviced returns after being proved to be successful and is proved to be successful boundary Face will re-initiate verifying process if the sound for detecting that the terminal uploads can not identify.

It should be noted that for the various method embodiments described above, for simple description, therefore, it is stated as a series of Combination of actions, but those skilled in the art should understand that, the application is not limited by the described action sequence because According to the application, some steps may be performed in other sequences or simultaneously.Secondly, those skilled in the art should also know It knows, the embodiments described in the specification are all preferred embodiments, related actions and modules not necessarily the application It is necessary.

Through the above description of the embodiments, those skilled in the art can be understood that according to above-mentioned implementation The man-machine knowledge method for distinguishing of example can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hard Part, but the former is more preferably embodiment in many cases.Based on this understanding, the technical solution of the application substantially or Say that the part that contributes to existing technology can be embodied in the form of software products, which is stored in In one storage medium (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that a terminal device (can be hand Machine, computer, server or network equipment etc.) execute method described in each embodiment of the application.

Embodiment 4

It according to the embodiment of the present application, additionally provides a kind of for implementing the device of above-mentioned man-machine knowledge method for distinguishing, is servicing Device side, as shown in figure 8, Fig. 8 is the device packet according to the structural schematic diagram of the device of the man-machine identification of the embodiment of the present application four It includes：

First receiving module 81, for receiving the checking request of terminal transmission；Sending module 82, for terminal send to Verification information, wherein information to be verified includes：Acoustic information to be verified and behavioural information to be verified；Second receiving module 83 is used In the user's voiceprint and behavioral data information that receive the correspondence information to be verified that terminal uploads；Identification module 84, for sentencing Whether operation object belonging to disconnected user's voiceprint and behavioral data information is identical as the target object prestored in model, and according to Corresponding judging result is exported according to the type of target object, wherein judging result includes：Operation object is natural person；Or, operation Object is robot.

In the embodiment of the present application, the checking request sent by receiving terminal；Information to be verified is sent to terminal, In, information to be verified includes：Acoustic information to be verified and behavioural information to be verified；Receive the correspondence information to be verified that terminal uploads User's voiceprint and behavioral data information；Whether judge operation object belonging to user's voiceprint and behavioral data information It is identical as the target object prestored in model, and corresponding judging result is exported according to the type of target object, wherein judgement knot Fruit includes：Operation object is natural person；Or, operation object is robot, achieve the purpose that protect machine request attack, thus The technical effect for promoting man-machine recognition accuracy is realized, and then is solved as being directed to existing for man-machine identification in the prior art Defect, the technical problem for causing man-machine recognition accuracy low.

Embodiment 5

According to the embodiment of the present application, additionally provide it is a kind of for implementing the device of above-mentioned man-machine knowledge method for distinguishing, in terminal Side, as shown in figure 9, Fig. 9 is according to the structural schematic diagram of the device of the man-machine identification of the embodiment of the present application five, which includes：

First sending module 91, for sending checking request to server；Receiving module 92, for receiving server foundation The information to be verified that checking request returns, wherein information to be verified includes：Acoustic information to be verified and behavioural information to be verified； Acquisition module 93, for acquiring user's voiceprint corresponding with acoustic information to be verified, and acquisition behavioural information to be verified Corresponding behavioral data information；Second sending module 94, for sending user's voiceprint and behavioral data information to server.

In the embodiment of the present application, by sending checking request to server；Server is received to return according to checking request Information to be verified, wherein information to be verified includes：Acoustic information to be verified and behavioural information to be verified；Acquisition with it is to be verified The corresponding user's voiceprint of acoustic information, and the corresponding behavioral data information of acquisition behavioural information to be verified；To server User's voiceprint and behavioral data information are sent, has achieved the purpose that protect machine request attack, has promoted people to realize The technical effect of machine recognition accuracy, and then solve as being directed to defect existing for man-machine identification in the prior art, lead to people The low technical problem of machine recognition accuracy.

Embodiment 6

According to the embodiment of the present application in another aspect, additionally providing a kind of storage medium, storage medium includes the journey of storage Sequence, wherein equipment where control storage medium executes above-described embodiment 2 and implements the man-machine identification in 3 in program operation Method.

Embodiment 7

According to the embodiment of the present application in another aspect, additionally providing a kind of processor, processor is used to run program, In, program executes above-described embodiment 2 and implements the man-machine knowledge method for distinguishing in 3 when running.

Embodiment 8

Embodiments herein additionally provides a kind of storage medium.Optionally, in the present embodiment, above-mentioned storage medium can For saving program code performed by man-machine knowledge method for distinguishing provided by above-described embodiment one.

Optionally, in the present embodiment, above-mentioned storage medium can be located in computer network in computer terminal group In any one terminal, or in any one mobile terminal in mobile terminal group.

Optionally, in the present embodiment, storage medium is arranged to store the program code for executing following steps：It connects Receive the checking request that terminal is sent；Information to be verified is sent to terminal, wherein information to be verified includes：Acoustic information to be verified With behavioural information to be verified；Receive the user's voiceprint and behavioral data information of the correspondence information to be verified that terminal uploads；Sentence Whether operation object belonging to disconnected user's voiceprint and behavioral data information is identical as the target object prestored in model, and according to Corresponding judging result is exported according to the type of target object, wherein judging result includes：Operation object is natural person；Or, operation Object is robot.

Optionally, in the present embodiment, storage medium is arranged to store the program code for executing following steps：Sentence Whether operation object is identical as the target object prestored in model belonging to disconnected user's voiceprint and behavioral data information includes： By prestoring model analyzing user voiceprint and behavioral data information, obtain user's vocal print feature in user's voiceprint, User reads aloud content information and user behavior characteristics；Judgement state user read aloud content information whether with it is to be tested in information to be verified It is identical to demonstrate,prove acoustic information, and judges whether user behavior characteristics are identical as the behavioural information to be verified in information to be verified；? Target object includes natural person and robot, and reads aloud content information and user behavior characteristics in user and believe with sound to be verified It ceases in situation identical with behavioural information to be verified, judges whether user's vocal print feature is that natural person's vocal print is special by prestoring model Sign.

Further, optionally, in the present embodiment, storage medium is arranged to store the journey for executing following steps Sequence code：User's vocal print feature includes：User voice frequency spectrum, user voice loudness and volume；User behavior characteristics include：Touching Track and touching frequency.

Optionally, in the present embodiment, storage medium is arranged to store the program code for executing following steps：According to Exporting corresponding judging result according to the type of target object includes：If judging belonging to user's vocal print feature and user behavior characteristics Operation object is natural person, then export judging result be operation object belonging to user's voiceprint and behavioral data information be from Right people；If judging operation object belonging to user's vocal print feature and user behavior characteristics for robot, exporting judging result is Operation object belonging to user's voiceprint and behavioral data information is robot.

Further, optionally, in the present embodiment, storage medium is arranged to store the journey for executing following steps Sequence code：After judging operation object belonging to user's vocal print feature and user behavior characteristics for natural person, method further includes： In the case that the operation object belonging to user's voiceprint and behavioral data information is natural person, judgement is within a preset time It is no to receive the corresponding user's voiceprint of operation object and behavioral data information；It is to receive within a preset time in judging result In the case where the corresponding user's voiceprint of operation object and behavioral data information, the verifying that refusal processing terminal is sent is asked It asks；It is not receive the corresponding user's voiceprint of operation object and behavioral data information within a preset time in judging result In the case of, the checking request of processing terminal transmission.

Optionally, in the present embodiment, storage medium is arranged to store the program code for executing following steps：? Output judging result is after operation object belonging to user's voiceprint and behavioral data information is robot, and method is also wrapped It includes：Outputting alarm signal, and refuse the checking request of processing terminal transmission.

Above-mentioned the embodiment of the present application serial number is for illustration only, does not represent the advantages or disadvantages of the embodiments.

In above-described embodiment of the application, all emphasizes particularly on different fields to the description of each embodiment, do not have in some embodiment The part of detailed description, reference can be made to the related descriptions of other embodiments.

In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others Mode is realized.Wherein, the apparatus embodiments described above are merely exemplary, such as the division of the unit, only A kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or components can combine or Person is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual Between coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication link of unit or module It connects, can be electrical or other forms.

The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme 's.

It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list Member both can take the form of hardware realization, can also realize in the form of software functional units.

If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product When, it can store in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words It embodies, which is stored in a storage medium, including some instructions are used so that a computer Equipment (can for personal computer, server or network equipment etc.) execute each embodiment the method for the application whole or Part steps.And storage medium above-mentioned includes：USB flash disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic or disk etc. be various to can store program code Medium.

The above is only the preferred embodiment of the application, it is noted that for the ordinary skill people of the art For member, under the premise of not departing from the application principle, several improvements and modifications can also be made, these improvements and modifications are also answered It is considered as the protection scope of the application.

Claims (17)

1. a kind of system of man-machine identification, which is characterized in that including：

Terminal, for acquiring user's voiceprint and behavioral data information；

Server is connect with the terminal, for according to the user's voiceprint for prestoring model the terminal being uploaded with The behavioral data information is identified, judges operation object belonging to user's voiceprint and the behavioral data information Whether identical as the target object prestored in model, the type according to the target object exports corresponding judging result, Wherein, the judging result includes：The operation object is natural person；Or, the operation object is robot.

2. system according to claim 1, which is characterized in that the terminal includes：Show screen, antenna and sensor Group, wherein

The display screen, for showing information to be verified, wherein the information to be verified includes：Acoustic information to be verified and Behavioural information to be verified；

The sensor group, for acquiring user according to the corresponding user's voiceprint of the acoustic information to be verified, and The corresponding behavioral data information of the behavioural information to be verified；

The antenna is connect with the sensor group, for sending user's voiceprint and the behavioral data information To the server.

3. system according to claim 1, which is characterized in that the server includes：Memory and processor, wherein

The memory described prestores model for storing；

The processor is sentenced for identifying user's voiceprint and the behavioral data information according to the storage model Break user's voiceprint and the behavioral data information it is whether identical as the target object prestored in model, judging It as a result is in the case where being, if the target object is the natural person, exporting judging result is user's voiceprint It is natural person with operation object belonging to the behavioral data information；If the target object is the robot, output is sentenced Disconnected result is that operation object belonging to user's voiceprint and the behavioral data information is robot, and is alerted.

4. system according to claim 3, which is characterized in that the model that prestores includes：Natural person's model and robot Model, wherein

Natural person's model judges the use for parsing user's voiceprint and the behavioral data information respectively Whether family voiceprint meets default natural person's vocal print feature, and judges whether the behavioral data information meets default nature The case where people's behavioural characteristic, the operation object belonging to user's voiceprint and the behavioral data information is natural person Under, receive all requests that the terminal is sent；

The robot model judges the use for parsing user's voiceprint and the behavioral data information respectively Whether family voiceprint meets default robot vocal print feature, and judges whether the behavioral data information meets default machine The case where people's behavioural characteristic, the operation object belonging to user's voiceprint and the behavioral data information is robot Under, refusal handles all requests that the terminal is sent.

5. system according to claim 3 or 4, which is characterized in that the processor, be also used to be in the judging result In the case that operation object belonging to user's voiceprint and the behavioral data information is natural person, judge when default It is interior whether to receive the corresponding user's voiceprint of the operation object and behavioral data information, it is described in judging result In the case where receiving the corresponding user's voiceprint of the operation object and behavioral data information in preset time, refusal processing All requests that the terminal is sent.

6. system according to claim 3, which is characterized in that the server further includes：Alarm, wherein

The alarm, is connected to the processor, for being user's voiceprint and the row in the judging result In the case where being robot for operation object belonging to data information, outputting alarm signal, so that at the processor refusal Manage all requests that the terminal is sent.

7. a kind of man-machine knowledge method for distinguishing, which is characterized in that including：

Receive the checking request that terminal is sent；

Information to be verified is sent to the terminal, wherein the information to be verified includes：Acoustic information to be verified and row to be verified For information；

Receive the user's voiceprint and behavioral data information of the correspondence information to be verified that the terminal uploads；

Judge operation object belonging to user's voiceprint and the behavioral data information whether with the mesh that prestores in model It is identical to mark object, and export corresponding judging result according to the type of the target object, wherein the judging result includes： The operation object is natural person；Or, the operation object is robot.

8. the method according to the description of claim 7 is characterized in that judgement user's voiceprint and the behavior number It is believed that ceasing, whether affiliated operation object is identical as the target object prestored in model to include：

By prestoring user's voiceprint described in model analyzing and the behavioral data information, obtain in user's voiceprint User's vocal print feature, user read aloud content information and user behavior characteristics；

Whether user described in judgement to read aloud content information identical as the acoustic information to be verified in the information to be verified if being stated, and Judge whether the user behavior characteristics are identical as the behavioural information to be verified in the information to be verified；

Include the natural person and the robot in the target object, and reads aloud content information and the use in the user In family behavioural characteristic situation identical with the acoustic information to be verified and the behavioural information to be verified, prestored by described Model judges whether user's vocal print feature is natural person's vocal print feature.

9. according to the method described in claim 8, it is characterized in that, the type output according to the target object is corresponding Judging result includes：

If judging, operation object belonging to user's vocal print feature and the user behavior characteristics for natural person, exports judgement It as a result is natural person for operation object belonging to user's voiceprint and the behavioral data information；

If judging, operation object belonging to user's vocal print feature and the user behavior characteristics for the robot, is exported Judging result is that operation object belonging to user's voiceprint and the behavioral data information is robot.

10. according to the method described in claim 9, it is characterized in that, in judgement user's vocal print feature and the use Operation object belonging to the behavioural characteristic of family be natural person after, the method also includes：

In the case that the operation object belonging to user's voiceprint and the behavioral data information is natural person, judge Whether the operation object corresponding user voiceprint and the behavioral data information are received in preset time；

It is that the corresponding user's voiceprint of the operation object and the row are received in the preset time in judging result In the case where data information, refusal handles the checking request that the terminal is sent；

It is not receive in the preset time the corresponding user's voiceprint of the operation object and described in judging result In the case where behavioral data information, the checking request that the terminal is sent is handled.

11. according to the method described in claim 9, it is characterized in that, being user's vocal print letter in the output judging result Breath and the behavioral data information belonging to operation object be robot after, the method also includes：

Outputting alarm signal, and refuse to handle the checking request that the terminal is sent.

12. a kind of man-machine knowledge method for distinguishing, which is characterized in that including：

Checking request is sent to server；

Receive the information to be verified that the server is returned according to the checking request, wherein the information to be verified includes：To Verify acoustic information and behavioural information to be verified；

Acquisition user's voiceprint corresponding with the acoustic information to be verified, and the acquisition behavioural information correspondence to be verified Behavioral data information；

User's voiceprint and the behavioral data information are sent to the server.

13. according to the method for claim 12, which is characterized in that the acquisition is corresponding with the acoustic information to be verified User's voiceprint, and acquire the corresponding behavioral data information of the behavioural information to be verified and include：

Acquisition user reads aloud voice content, user voice frequency spectrum, user voice loudness and the sound when acoustic information to be verified Amount；

Touching track and touching frequency when acquisition user only needs the behavioural information to be verified.

14. a kind of device of man-machine identification, which is characterized in that including：

First receiving module, for receiving the checking request of terminal transmission；

Sending module, for sending information to be verified to the terminal, wherein the information to be verified includes：Sound to be verified Information and behavioural information to be verified；

Second receiving module, for receiving user's voiceprint and the behavior of the correspondence information to be verified that the terminal uploads Data information；

Identification module, for judge operation object belonging to user's voiceprint and the behavioral data information whether in advance The target object deposited in model is identical, and exports corresponding judging result according to the type of the target object, wherein described to sentence Disconnected result includes：The operation object is natural person；Or, the operation object is robot.

15. a kind of device of man-machine identification, which is characterized in that including：

First sending module, for sending checking request to server；

Receiving module, the information to be verified returned for receiving the server according to the checking request, wherein described to be tested Demonstrate,proving information includes：Acoustic information to be verified and behavioural information to be verified；

Acquisition module for acquiring user's voiceprint corresponding with the acoustic information to be verified, and acquires described to be tested Demonstrate,prove the corresponding behavioral data information of behavioural information；

Second sending module, for sending user's voiceprint and the behavioral data information to the server.

16. a kind of storage medium, which is characterized in that the storage medium includes the program of storage, wherein run in described program When control the storage medium where equipment perform claim require any one of 7 to 13 described in man-machine knowledge method for distinguishing.

17. a kind of processor, which is characterized in that the processor is for running program, wherein right of execution when described program is run Benefit require any one of 7 to 13 described in man-machine knowledge method for distinguishing.