WO2020006886A1 - 门禁系统的识别方法、装置、门禁系统及存储介质 - Google Patents
门禁系统的识别方法、装置、门禁系统及存储介质 Download PDFInfo
- Publication number
- WO2020006886A1 WO2020006886A1 PCT/CN2018/106772 CN2018106772W WO2020006886A1 WO 2020006886 A1 WO2020006886 A1 WO 2020006886A1 CN 2018106772 W CN2018106772 W CN 2018106772W WO 2020006886 A1 WO2020006886 A1 WO 2020006886A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- voiceprint
- voiceprint information
- access control
- control system
- random code
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/06—Decision making techniques; Pattern matching strategies
- G10L17/08—Use of distortion metrics or a particular distance between probe pattern and reference templates
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/06—Decision making techniques; Pattern matching strategies
- G10L17/14—Use of phonemic categorisation or speech recognition prior to speaker recognition or verification
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification techniques
- G10L17/22—Interactive procedures; Man-machine interfaces
Definitions
- the present application relates to the field of voiceprint recognition, and in particular, to a method, a device, an access control system, and a computer-readable storage medium for identifying an access control system.
- the main purpose of this application is to provide an identification method, device, access control system, and computer-readable storage medium for an access control system, which aims to solve the problem of inconvenient access identification of the current access control system.
- an identification device for an access control system including:
- a judging module configured to judge whether the mapped random code is consistent with the random code displayed by the access control system
- a comparison module configured to perform a voiceprint comparison between the voiceprint information and the found voiceprint template to determine whether the user and the owner of the voiceprint template are consistent;
- the present application also provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the identification method of the access control system as described above is implemented. A step of.
- This application obtains the voiceprint information sent by the user, and then converts the voiceprint information into corresponding text data including a mapping random code and a mapping fixed code; after excluding the voiceprint information sent by the user as the machine voice by mapping the random code , Use the mapping fixed code to find a single or a few voiceprint templates in the preset mapping data to determine whether the user is consistent with the owner of one of the voiceprint templates; when the user is consistent with the owner of one of the voiceprint templates To confirm that the access control system has been identified. Therefore, the entry and exit recognition control of the access control system is realized by the entry of voiceprint information, and the operation is convenient.
- mapping random code is consistent with the random code displayed by the access control system, it can prevent criminals from using the recorded machine voice to identify the access control system; it also finds the corresponding correspondence from the preset mapping data by mapping the fixed code.
- the few or single voiceprint templates also reduce the scope of voiceprint comparison and speed up the speed and accuracy of the access control system recognition.
- FIG. 1 is a schematic structural diagram of an access control system according to an embodiment of the present application.
- step S50 is a detailed flowchart of step S50 in the second embodiment of the identification method of the access control system of the present application.
- FIG. 5 is a schematic flowchart of a fourth embodiment of a method for identifying an access control system of this application.
- FIG. 6 is a functional module diagram of the identification device of the access control system of the present application.
- FIG. 1 is a schematic diagram of a hardware structure of an access control system provided by the present application.
- the access control system may include components such as a display module 10, a memory 20, and a processor 30.
- the processor 30 is connected to the memory 20 and the display module 10, respectively.
- a computer program is stored in the memory 20, and the computer program is executed by the processor 30 at the same time.
- the display module 10 is a display component of the access control system, and may be a capacitive screen component or a resistive screen component and the like.
- the display module 10 may extract data from the memory 20 and display the data as an image or a character string.
- the character string may be a random code or a prompt character for pass / fail recognition by the access control system.
- the character string may be composed of letters, Numbers and / or Chinese characters.
- Memory 20 which can be used to store software programs and various data, can be an SD card (Secure Digital Memory Card, secure digital card), memory modules, and more.
- the memory 20 may mainly include a storage program area and a storage data area, where the storage program area may store an operating system, at least one application required by a function (such as obtaining voiceprint information), and the like; the storage data area may include a database to store data Zones can store data or information created based on the use of the access control system.
- the memory 20 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.
- the processor 30 is the control center of the access control system. It uses various interfaces and lines to connect various parts of the entire access control system, and runs or executes software programs and / or modules stored in the memory 20 and calls the Data, perform various functions of the access control system and process data, so as to monitor the access control system as a whole.
- the processor 30 may include one or more processing units; optionally, the processor 30 may integrate an application processor and a modem processor, wherein the application processor mainly processes an operating system, a user interface, and an application program, etc.
- the tuning processor mainly handles wireless communication. It can be understood that the aforementioned modem processor may not be integrated into the processor 30.
- the above access control system may further include a circuit control module for connecting to a power source to ensure normal work of other components.
- the above access control system may further include a communication module.
- the communication module may not be turned on during the normal use of the access control system. Only when the administrator permission of the access control system is turned on, the voice database of the local criminal can be updated and new voices recorded. Pattern template and / or firmware upgrade; the communication module may also start and send early warning information to when the similarity between the voiceprint information and any special voiceprint template in the voiceprint database of the local criminal is greater than a first preset threshold Supervisor terminal.
- the above access control system may further include a microphone array, and the microphone array may be disposed near the display module and used to receive voice information entered by a user.
- the method includes:
- Step S10 Acquire voiceprint information issued by the user
- the acquisition of voiceprint information is the premise of voiceprint recognition. After obtaining the voice information from the user through the microphone array, it uses electroacoustic instruments to extract the sound wave spectrum carrying speech information from the voice information as voiceprint information. .
- step S10 It is detected by infrared whether there is a user in the area where the access control system is located; and when the infrared detects that a user is present in the area where the access control system is located, step S10 is performed.
- infrared human body sensing technology is used, which can use an infrared sensor to emit an infrared beam. When the infrared beam is blocked due to the passage of a person, it can be determined that someone is in front of the access control detection system.
- an infrared thermal imaging sensor can also be used to determine a person near the access control detection system when a person passes by and generates a specific infrared band signal due to human thermal radiation.
- the infrared array detects the presence of a user in the area of the access control system and triggers the microphone array of the access control system to acquire voiceprint information, which can reduce the working time of the microphone array and reduce the loss of the voiceprint recognition device of the access control system.
- Step S20 converting the voiceprint information into corresponding text data including a mapping random code and a mapping fixed code
- the processor of the access control system is mainly through ASR (Automatic Speech Recognition (automatic speech recognition technology) converts voiceprint information into text data, where the text data includes a mapping random code and a mapping fixed code. If the text data corresponding to the voiceprint information entered by the user includes only a mapping fixed code or a mapping random code, the identification of the access control system fails.
- ASR Automatic Speech Recognition
- the process of obtaining text data based on ASR technology is: converting the voiceprint information from the original non-stationary state to a stationary short-term voice signal, that is, frame, where each short-term voice signal period is called a frame; The short-term speech signal is then transformed into a multi-dimensional vector to form a feature vector sequence.
- text data is obtained through analysis and transformation of phonemes and states based on a preset model.
- the preset model can be a hidden Markov model and an artificial neural network. Wait.
- voiceprint information can also be pre-processed, such as mute cutting of voiceprint information at the end and end to reduce interference to subsequent steps. It should also be noted that the formation of feature vector sequences by extracting feature parameters needs to accurately and clearly express the information carried by the voiceprint, such as channel characteristics and auditory characteristics; feature parameters need to be independent and easy to calculate.
- the usage scenario of the access control system may be that the user walks into the infrared-detectable range, the microphone array starts to detect the user's sound, and the display module will randomly display a random code.
- the random code is a two-digit number as an example.
- the user will read the two digits displayed by the display module and say a fixed code that can confirm the identity of the user.
- the processor extracts the voiceprint information in the voice information, and then converts the voiceprint information into text data including mapped random codes and mapped fixed codes based on ASR technology.
- the fixed code is set based on the area to which different users actually belong, or is set based on different users.
- the fixed code corresponding to user A is AABB
- the fixed code corresponding to user B is ABCD
- the fixed code corresponding to user C is ABAB
- the fixed code corresponding to user D may be AABB. That is to say, the fixed code set based on different users can also be in a one-to-one or one-to-many relationship with the user, and can be set by referring to the remaining memory capacity of the memory or the requirements of the administrator.
- Step S30 Determine whether the mapped random code is consistent with the random code displayed by the access control system; if so, perform step S40;
- the mapped random code is converted by the user based on the voiceprint information displayed by the random code displayed on the display module, if the mapped random code is consistent with the random code displayed by the access control system at that time, it means that the user is not a machine voice, and the fixed code is indeed Read by the user according to the displayed random code.
- the setting of fixed code reading and comparison rules is mainly due to criminals recording and playing the user's voice information to wait for the opportunity to pass the access control system based on voiceprint recognition, and it is precisely because of the timeliness of random codes that they can It is guaranteed that the random code entered by the criminals in advance is not consistent with the random code displayed by the display module when waiting for the opportunity to pass through the access control system, reducing the entry of criminals, thereby improving the security of the access control system.
- time interval at which the display module of the access control system updates the random code can be set according to actual needs, for example, it is updated every 5 seconds.
- step S30 When the determination result of step S30 is "No", the access control system recognizes an error, and may send an error reminder to the user through the display module, and may also remind the user to re-enter the sound. Further, when the number of errors reaches a certain limit, the fixed code spoken by the user can be frozen, and the communication module can be triggered to seek further confirmation from the administrator.
- Step S40 using the mapping fixed code as a keyword, searching for a voiceprint template corresponding to the mapping fixed code from preset mapping data between the fixed code and the voiceprint template;
- the determination process is to preset the preset mapping data between the fixed code and the voiceprint template in the memory, and find the voiceprint template corresponding to the mapping fixed code in the preset mapping data by mapping the fixed code. It can be understood that, according to the setting of the corresponding relationship between different fixed codes and users, a voiceprint template corresponding to one mapped fixed code may be a few or one.
- Step S50 Perform a voiceprint comparison between the voiceprint information and the found voiceprint template to determine whether the user and the owner of the voiceprint template are consistent; if so, perform step S60;
- step S60 it is determined that the identification of the access control system passes.
- the similarity between the voiceprint information and the found voiceprint template may be calculated, and then it is determined whether the similarity is greater than a first preset threshold. If the similarity is greater than the first preset threshold, It is considered that the user corresponds to the owner corresponding to the found voiceprint template.
- voiceprint information sent by a user is obtained; the voiceprint information is converted into corresponding text data including a mapping random code and a mapping fixed code; and whether the mapped random code and the random code displayed by the access control system are determined. Consistent; when the mapped random code is consistent with the random code, using the mapped fixed code as a key, from the preset mapping data between the fixed code and the voiceprint template, find the sound corresponding to the mapped fixed code Compare the voiceprint information with the found voiceprint template to determine whether the user is the same as the owner of the voiceprint template; when the user is the owner of the voiceprint template When they are consistent, it is determined that the identification of the access control system passes.
- mapping random code is consistent with the random code displayed by the access control system, it can prevent criminals from using the recorded machine voice to identify the access control system; it also finds the corresponding correspondence from the preset mapping data by mapping the fixed code.
- the few or single voiceprint templates reduce the range of voiceprint comparison and speed up the speed and accuracy of the access control system recognition.
- the mapping relationship data is set in the memory of the access control system in advance, and fast voiceprint comparison can be performed without networking, which further improves the recognition speed.
- the step S50 when the voiceprint template found is greater than When equal to two, the step S50 includes:
- Step S51 Calculate the similarity between each voiceprint template and the voiceprint information
- mapping fixed code corresponds to multiple voiceprint templates
- the similarity between the voiceprint information and the voiceprint information needs to be calculated one by one to determine which of the multiple voiceprint templates is related to the user's Voiceprint information has the highest similarity.
- Step S52 Select the voiceprint template with the highest similarity to the voiceprint information from all the voiceprint templates as a reference voiceprint template;
- Step S53 Determine whether the similarity between the reference voiceprint template and the voiceprint information is greater than a first preset threshold, and when the similarity between the reference voiceprint template and the voiceprint information is greater than the When the first preset threshold value is determined, the user is consistent with the owner of the reference voiceprint template.
- the most similar voiceprint template and the similarity between the voiceprint template and the voiceprint information may be individually extracted as a reference, and the similarity between the reference voiceprint template and the voiceprint information may be compared with a first preset threshold, Only when the similarity is greater than the first preset threshold, it is determined that the user is consistent with the owner of the reference voiceprint template.
- the voiceprint information of some users has high similarity, but the users corresponding to the voiceprint information are not the same person. Therefore, it is necessary to set a first preset threshold to exclude users from corresponding to the reference voiceprint template.
- the voiceprint information between the attributors has a high degree of similarity, but the user and the attributor are not the same person.
- This solution provides an optional implementation of how to determine whether the user's identity is successfully identified based on multiple voiceprint templates corresponding to the mapping fixed code, which increases the utilization of the mapping fixed code and ensures the accuracy of the access control system identification.
- the method further includes:
- Step S71 intercept the first voiceprint information corresponding to the mapping random code and the second voiceprint information corresponding to the mapping fixed code from the voiceprint information;
- the identification by the access control system is opportunistic, and the voiceprint information sent by the user can be intercepted.
- the first voiceprint information corresponding to the random code and the second voiceprint information corresponding to the fixed code are mapped, and the similarity between the first voiceprint information and the second voiceprint information is compared.
- Step S72 calculating a similarity between the first voiceprint information and the second voiceprint information
- step S73 it is determined whether the similarity between the first voiceprint information and the second voiceprint information is greater than a second preset threshold; if yes, step S30 is performed; if not, step S80 is performed;
- the similarity between the first voiceprint information corresponding to the mapped random code and the second voiceprint information corresponding to the fixed mapping code is issued, only when the similarity between the first voiceprint information and the second voiceprint information is greater than a second preset threshold At this time, it can be determined that the user who sends the first voiceprint message and the user who sends the second voiceprint message are the same person.
- the second preset threshold may be the same as the first preset threshold, or may be higher than the first preset threshold.
- step S80 it is determined that the identification of the access control system is incorrect.
- the access control recognition fails, and the access control system can issue an error reminder through the display module.
- Step S91 Determine, based on the voiceprint information, an interval between the user sending the first voiceprint information corresponding to the mapping random code and sending the second voiceprint information corresponding to the mapping fixed code;
- an interval time between the first voiceprint information and the second voiceprint information is obtained.
- the interval time refers to a time difference between a time point when the first voiceprint information ends and a time point when the second voiceprint information is sent out.
- Step S92 it is determined whether the interval time is less than a preset interval threshold; if yes, step S30 is performed; if not, step S80 is performed;
- step S80 it is determined that the identification of the access control system is incorrect.
- the preset interval threshold is compared with the interval time to improve the recognition speed of the access control system.
- the scheme of comparing the similarity between the first voiceprint information and the second voiceprint information in the third embodiment with the second preset threshold may also be compared with the interval time and the preset interval threshold in this embodiment.
- the combination of the solutions ensures the security of the identification of the access control system and improves the identification speed.
- a voiceprint database for local criminals may also be provided, in which special voiceprint templates are stored, and each special voiceprint template may correspond to an illegal criminal.
- the voiceprint information is compared with all the special voiceprint templates in the voiceprint database of the local criminal, as long as the voiceprint issued by the user If the similarity between the information and any one of the special voiceprint templates is greater than the first preset threshold, the communication module can be triggered to network and send an early warning message to the manager.
- step S20 may be continued.
- the communication module is triggered only for criminals to conduct network operations, which reduces the interference of other communication equipment on the access control system through the communication module and ensures the smooth operation of the access control system.
- the present application further provides an identification device for an access control system.
- the device may be a processor for access control identification or an access control device.
- the device includes:
- An acquisition module 10 configured to acquire voiceprint information issued by a user
- a conversion module 20 configured to convert the voiceprint information into corresponding text data including a mapping random code and a mapping fixed code
- a determining module 30, configured to determine whether the mapped random code is consistent with a random code displayed by the access control system
- a searching module 40 configured to: when the mapped random code is consistent with the random code, use the mapped fixed code as a keyword, and search for the fixed mapping from preset mapping data between the fixed code and the voiceprint template; The corresponding voiceprint template;
- a comparison module 50 configured to perform a voiceprint comparison between the voiceprint information and the found voiceprint template to determine whether the user and the owner of the voiceprint template are consistent;
- a determining module 60 is configured to determine that the identification of the access control system passes when the user is consistent with an owner of the voiceprint template.
- the comparison module when the number of voiceprint templates found is greater than or equal to two, the comparison module includes:
- a computing unit configured to separately calculate the similarity between each voiceprint template and the voiceprint information
- a selecting unit configured to select, from all voiceprint templates, the voiceprint template with the highest similarity to the voiceprint information as a reference voiceprint template
- a determining unit configured to determine that the user is consistent with the owner of the reference voiceprint template when the similarity between the reference voiceprint template and the voiceprint information is greater than the first preset threshold.
- the device further includes an interception module and a calculation module.
- an interception module and a calculation module.
- the intercepting module is configured to intercept the first voiceprint information corresponding to the mapping random code and the second voiceprint information corresponding to the mapping fixed code from the voiceprint information;
- the calculation module is configured to calculate a similarity between the first voiceprint information and the second voiceprint information
- the determining module is further configured to determine whether a similarity between the first voiceprint information and the second voiceprint information is greater than a second preset threshold; and when the first voiceprint information and the second voiceprint information When the similarity between the voiceprint information is greater than a second preset threshold, the step of determining whether the mapped random code is consistent with the random code displayed by the access control system is continued; when the first voiceprint information and the second When the similarity between the voiceprint information is less than or equal to a second preset threshold, the determining module is triggered to execute a step of determining that the access control system recognizes an error.
- the determining module is further configured to determine, according to the voiceprint information, that the user sends the first voiceprint information corresponding to the mapping random code to the second voiceprint corresponding to the mapping fixed code. Interval between voiceprint messages;
- the judging module is further configured to judge whether the interval time is less than a preset interval threshold; and when the interval time is less than a preset interval threshold, perform a judgment whether the mapped random code is consistent with a random code displayed by the access control system When the interval time is greater than or equal to a preset interval threshold, triggering the determining module to execute a step of determining that the access control system recognizes an error.
- the transformation module includes:
- a preprocessing unit configured to preprocess the voiceprint information
- An extraction unit configured to extract feature parameters from the pre-processed voiceprint information, and generate a feature vector sequence according to all the feature parameters
- a recognition unit is configured to extract and identify the feature vector sequence based on a preset model to obtain text data including a mapping random code and a mapping fixed code.
- the apparatus further includes a sending module; in the apparatus,
- the comparison module is further configured to compare the voiceprint information with a special voiceprint template stored in a voiceprint database of a local criminal;
- the sending module is configured to send an early warning message to a manager when the similarity between the voiceprint information and any special voiceprint template in the voiceprint database of the local criminal is greater than a first preset threshold.
- the apparatus further includes:
- a detection module for detecting whether a user appears in the area where the access control system is located by infrared; and when the infrared detects that a user is present in the area where the access control system is located, triggering the acquisition module to execute the sound generated by the user Grain information steps.
- the present application also proposes a computer-readable storage medium on which a computer program is stored.
- the computer-readable storage medium may be the memory 20 in the access control system 100 of FIG. 1, or may be a ROM (Read-Only Memory (Read Only Memory) / RAM (Random Access Memory, at least one of a random access memory, a magnetic disk, and an optical disk.
- the computer-readable storage medium includes several instructions to enable a terminal device (such as a mobile phone, computer, server, and access control system) with a processor , Or network equipment, etc.) perform the methods described in the embodiments of the present application.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Security & Cryptography (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Collating Specific Patterns (AREA)
- Telephonic Communication Services (AREA)
Abstract
本申请公开了一种门禁系统的识别方法、装置、门禁系统及存储介质,方法包括步骤:获取用户发出的声纹信息;将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;判断所述映射随机码与所述门禁系统显示的随机码是否一致;当所述映射随机码与所述随机码一致时,以所述映射固定码为关键词,从固定码与声纹模板间的预设映射数据中,查找所述映射固定码对应的声纹模板;将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致;当所述用户与该声纹模板的归属者一致时,确定所述门禁系统识别通过。
Description
本申请要求于2018年07月06日提交中国专利局、申请号为201810737583.4、发明名称为“门禁系统的识别方法、装置、门禁系统及存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及声纹识别领域,尤其涉及门禁系统的识别方法、装置、门禁系统及计算机可读存储介质。
背景技术
为了提高小区住户的居住安全以及办公区域的财产安全,现在大部分小区和办公楼都设置了门禁系统。目前的门禁系统主要是通过输入密码或刷卡等形式进行用户确认,当输入密码正确或刷卡完成后才会打开闸门,但当用户双手拎着很多物品时,执行密码输入或刷卡操作十分不便。
申请内容
本申请的主要目的在于提供一种门禁系统的识别方法、装置、门禁系统及计算机可读存储介质,旨在解决目前门禁系统出入识别十分不便的问题。
为实现上述目的,本申请提供一种门禁系统的识别方法,所述方法包括步骤:
获取用户发出的声纹信息;
将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;
判断所述映射随机码与所述门禁系统显示的随机码是否一致;
当所述映射随机码与所述随机码一致时,以所述映射固定码为关键词,从固定码与声纹模板间的预设映射数据中,查找所述映射固定码对应的声纹模板;
将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致;
当所述用户与该声纹模板的归属者一致时,确定所述门禁系统识别通过。
为实现上述目的,本申请还提供一种门禁系统的识别装置,包括:
获取模块,用于获取用户发出的声纹信息;
转化模块,用于将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;
判断模块,用于判断所述映射随机码与所述门禁系统显示的随机码是否一致;
查找模块,用于当所述映射随机码与所述随机码一致时,以所述映射固定码为关键词,从固定码与声纹模板间的预设映射数据中,查找所述映射固定码对应的声纹模板;
比对模块,用于将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致;
确定模块,用于当所述用户与该声纹模板的归属者一致时,确定所述门禁系统识别通过。
为实现上述目的,本申请还提供一种门禁系统,所述门禁系统包括:显示模块、存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序,所述计算机程序被所述处理器执行时实现如上所述的门禁系统的识别方法的步骤。
为实现上述目的,本申请还提供一种计算机可读存储介质,所述计算机可读存储介质上存储有计算机程序,所述计算机程序被处理器执行时实现如如上所述的门禁系统的识别方法的步骤。
本申请通过获取用户发出的声纹信息,接着将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;在通过映射随机码排除用户发出的声纹信息为机器语音后,利用映射固定码找到预设映射数据中单个或少数几个声纹模板,以确定用户是否与其中某个声纹模板的归属者一致;在用户与其中某个声纹模板的归属者一致时,确认门禁系统识别通过。从而利用声纹信息的录入实现了门禁系统的出入识别控制,操作方便。再者,由于通过映射随机码与门禁系统显示的随机码进行了一致性匹配,能防止有不法分子利用录制的机器语音进行门禁系统识别;还通过映射固定码从预设映射数据中查找到对应的少数几个或单个声纹模板,也缩小了声纹比对的范围,加快了门禁系统识别的速度与准确性。
附图说明
图1是本申请实施例方案涉及的门禁系统的结构示意图;
图2为本申请门禁系统的识别方法第一实施例的流程示意图;
图3为本申请门禁系统的识别方法第二实施例中步骤S50的细化流程示意图;
图4为本申请门禁系统的识别方法第三实施例的流程示意图;
图5为本申请门禁系统的识别方法第四实施例的流程示意图;
图6为本申请门禁系统的识别装置的一功能模块示意图。
本申请目的的实现、功能特点及优点将结合实施例,参照附图做进一步说明。
具体实施方式
应当理解,此处所描述的实施例仅仅用以解释本申请,并不用于限定本申请。
请参看图1,图1为本申请所提供的门禁系统的硬件结构示意图。所述门禁系统可以包括显示模块10、存储器20以及处理器30等部件。在所述门禁系统中,所述处理器30分别与所述存储器20以及所述显示模块10连接,所述存储器20上存储有计算机程序,所述计算机程序同时被处理器30执行。
显示模块10,是门禁系统的显示部件,可以是电容屏组件或电阻屏组件等等。所述显示模块10可以从存储器20中提取数据,并将数据显示为图像或字符串,所述字符串可以为随机码或门禁系统识别通过/失败的提示字符,所述字符串可以由字母、数字和/或中文字符组成。
存储器20,可用于存储软件程序以及各种数据,可以是SD卡(Secure Digital Memory
Card,安全数码卡)、内存条等等。存储器20可主要包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声纹信息的获取)等;存储数据区可包括数据库,存储数据区可存储根据门禁系统的使用所创建的数据或信息等。此外,存储器20可以包括高速随机存取存储器,还可以包括非易失性存储器,例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。
处理器30,是门禁系统的控制中心,利用各种接口和线路连接整个门禁系统的各个部分,通过运行或执行存储在存储器20内的软件程序和/或模块,以及调用存储在存储器20内的数据,执行门禁系统的各种功能和处理数据,从而对门禁系统进行整体监控。处理器30可包括一个或多个处理单元;可选地,处理器30可集成应用处理器和调制解调处理器,其中,应用处理器主要处理操作系统、用户界面和应用程序等,调制解调处理器主要处理无线通信。可以理解的是,上述调制解调处理器也可以不集成到处理器30中。
尽管图1未示出,但上述门禁系统还可以包括电路控制模块,用于与电源连接,保证其他部件的正常工作。上述门禁系统还可以包括通信模块,所述通信模块在门禁系统正常使用过程中可以不开启,仅当门禁系统的管理员权限打开时,才能够进行本地犯罪人员声纹库更新、录入新的声纹模板和/或固件升级;所述通信模块还可以在声纹信息与本地犯罪人员声纹库中任一个特殊声纹模板间的相似度大于第一预设阈值时,启动并发送预警信息给管理者终端。上述门禁系统还可以包括麦克风阵列,所述麦克风阵列可以设置在显示模块附近,用于接收用户录入的声音信息。本领域技术人员可以理解,图1中示出的门禁系统结构并不构成对门禁系统的限定,可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件布置。
基于上述硬件结构,提出本申请方法各个实施例。
参见图2,在本申请门禁系统的识别方法的第一实施例中,包括:
步骤S10,获取用户发出的声纹信息;
声纹信息的获取是声纹识别的前提,其主要是在通过麦克风阵列获取用户发出的声音信息后,利用电声学仪器从声音信息中提取携带言语信息的声波频谱,以此作为声纹信息。
可选地,在所述步骤S10之前,还包括:
通过红外线检测所述门禁系统所在区域内是否有用户出现;并当所述红外线检测到所述门禁系统所在区域内有用户出现时,才执行步骤S10。这其中运用到了红外线人体感应技术,可以是利用红外传感器发出红外光束,当红外光束由于人的经过被遮挡时,可以确定门禁检测系统前有人。或者,还可以通过红外热成像传感器,当有人经过时,由于人体热辐射产生特定的红外线波段信号,确定门禁检测系统附近有人。通过当红外线检测到门禁系统所在区域内有用户出现时,才触发门禁系统的麦克风阵列进行声纹信息的获取,能够减少麦克风阵列的工作时间,降低门禁系统声纹识别设备的损耗。
步骤S20,将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;
门禁系统的处理器主要是通过ASR(Automatic Speech
Recognition,自动语音识别技术)将声纹信息转化为文本数据,其中所述文本数据包括映射随机码以及映射固定码。如果用户录入的声纹信息对应的文本数据中仅包括映射固定码或映射随机码时,门禁系统的识别不通过。
基于ASR技术得到文本数据的过程是:将声纹信息从原先的非平稳态转化为平稳态的短时语音信号,即分帧,其中每一个短时语音信号的时段称为一帧;然后将短时语音信号转变为多维向量,以形成特征向量序列,最后基于预设模型通过音素和状态的分析转化识别得到文本数据,其中预设模型可以是隐马尔可夫模型和人工神经网络等等。此外,在进行分帧之前,还可以进行声纹信息的预处理,例如将首尾端的声纹信息进行静音切除,以降低对后续步骤造成的干扰。还需要说明的是,通过提取特征参数形成特征向量序列需要准确清晰的表达声纹所携带的信息,如声道特性、听觉特性;特征参数之间则需要相互独立、易于计算。
在本实施例中,门禁系统的使用情景可以是用户走到红外线可检测范围内,麦克风阵列开始侦测用户发出的声音,同时显示模块会随机显示随机码,以随机码为两位数字为例,用户会读出显示模块显示的两位数字,并说出可以确认用户身份的固定码。麦克风阵列获取了用户读出的声音后,由处理器提取声音信息中的声纹信息,然后再基于ASR技术将声纹信息转化为包括映射随机码和映射固定码的文本数据。
需要说明的是,在使用过程中,固定码是基于不同用户实际所属区域设定的,或者是基于不同用户设定的。
以基于不同用户实际所属区域设定固定码为例,假设固定码由楼栋号、单元号、房间号以及用户序号顺序排列组合而成,如果用户1住在13栋2单元27楼F号,则固定码可以是“130227F01”或者“130227F”,其中“13”表示用户1的楼栋号,02表示单元号,27F表示房间号,“01”表示用户序号。在房间号后是否还携带有用户序号可以根据用户家中的用户数量进行确认,也可以直接不将用户序号加入固定码的构建规则中,由多名用户使用同一固定码。也就是说,基于不同用户实际所属区域设定的固定码可以是固定码与用户之间一对一或一对多的关系。
以基于不同用户设定固定码为例,用户A对应的固定码为AABB,用户B对应的固定码为ABCD,用户C对应的固定码为ABAB,用户D对应的固定码也可以为AABB。也就是说,基于不同用户设定固定码也可以与用户之间是一对一或一对多的关系,可以参照存储器的剩余内存容量或管理者的要求进行设定。
可以理解的是,当固定码与用户之间处于一对多的关系时,可以设置用户数量阈值,当固定码对应的用户数量达到设定的用户数量阈值时,则在门禁系统更新声纹模板时,停止为新的用户匹配该固定码。通过在固定码与用户之间的对应关系为一对多时,设置用户数量阈值,能增加单个固定码的使用率,同时保证在声纹比对时,将用户的声纹信息与预设映射数据中所有的声纹模板的对比缩减为一对少数几个,缩减了声纹比对的样本量。
步骤S30,判断所述映射随机码与所述门禁系统显示的随机码是否一致;若是,则执行步骤S40;
由于映射随机码是用户根据显示模块显示的随机码读出的声纹信息转化而成的,如果映射随机码与门禁系统当时显示的随机码一致时,表示用户并非机器语音,该固定码确实是用户根据显示的随机码读出的。可以理解的是,固定码读取及比较规则的设置主要是由于有不法分子通过录音并播放用户的声音信息伺机通过基于声纹识别的门禁系统,而正是由于随机码的时效性特点,能保证不法分子事先录入的随机码与伺机通过门禁系统时显示模块显示的随机码不一致,减少了不法分子的进入,从而提高了门禁系统的安全性。
需要说明的是,门禁系统的显示模块更新随机码的时间间隔可以根据实际需要进行设置,例如5秒更新一次。
当步骤S30的判断结果为“否”时,门禁系统识别错误,可以通过显示模块发送错误提醒给用户,还可以提醒用户重新录入声音。进一步地,还可以当该错误次数达到一定限制时,冻结该名用户说出的固定码,并触发通信模块启动,以寻求管理者进行进一步确认。
步骤S40,以所述映射固定码为关键词,从固定码与声纹模板间的预设映射数据中,查找所述映射固定码对应的声纹模板;
在排除了用户发出的声音为机器语音之后,可以进一步确认用户是否为门禁系统所保护区域范围内的用户。其确定的过程是在存储器中预先存储了固定码与声纹模板间的预设映射数据,通过映射固定码查找预设映射数据中映射固定码对应的声纹模板。可以理解的是,根据不同固定码与用户之间的对应关系的设置,一个映射固定码对应的声纹模板可以是少数几个或一个。
步骤S50,将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致;若是,则执行步骤S60;
步骤S60,确定所述门禁系统识别通过。
将查找到的声纹模板与用户发出的声纹信息进行声纹比对即可确定门禁系统是否识别通过。在进行声纹比对时,可以是计算声纹信息与查找到的声纹模板间的相似度,然后判断该相似度是否大于第一预设阈值,如果该相似度大于第一预设阈值,则认为用户与查找到的声纹模板对应的归属者一致。
本实施例通过获取用户发出的声纹信息;将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;判断所述映射随机码与所述门禁系统显示的随机码是否一致;当所述映射随机码与所述随机码一致时,以所述映射固定码为关键词,从固定码与声纹模板间的预设映射数据中,查找所述映射固定码对应的声纹模板;将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致;当所述用户与该声纹模板的归属者一致时,确定所述门禁系统识别通过。从而利用声纹信息的录入实现了门禁系统的出入识别控制,操作方便。再者,由于通过映射随机码与门禁系统显示的随机码进行了一致性匹配,能防止有不法分子利用录制的机器语音进行门禁系统识别;还通过映射固定码从预设映射数据中查找到对应的少数几个或单个声纹模板,缩小了声纹比对的范围,加快了门禁系统识别的速度与准确性。此外,还将映射关系数据预先设置在了门禁系统的存储器中,不需要联网即可进行快速声纹比对,也进一步提高了识别速度。
进一步地,参见图3,基于本申请门禁系统的识别方法的第一实施例提出本申请门禁系统的识别方法的第二实施例,在本实施例中,当查找到的所述声纹模板大于或等于2个时,所述步骤S50包括:
步骤S51,分别计算每个声纹模板与所述声纹信息间的相似度;
在本实施例中,由于映射固定码对应有多个声纹模板,因此需要逐一通过计算与声纹信息间的相似度,来确定所述多个声纹模板中哪一个声纹模板与用户的声纹信息的相似度最高。
步骤S52,从所有声纹模板中选择与所述声纹信息间的所述相似度最高的所述声纹模板作为参考声纹模板;
步骤S53,判断所述参考声纹模板与所述声纹信息间的相似度是否大于第一预设阈值,其中,当所述参考声纹模板与所述声纹信息间的相似度大于所述第一预设阈值时,确定所述用户与该参考声纹模板的归属者一致。
可以将最为相似的声纹模板以及该声纹模板与声纹信息间的相似度单独拎出来作为参考,并将参考声纹模板与声纹信息间的相似度与第一预设阈值进行比较,仅当相似度大于第一预设阈值时,才确定用户与该参考声纹模板的归属者一致。
需要说明的是,有部分用户的声纹信息之间相似度很高,但实质上声纹信息所对应的用户并非同一个人,因此需要设置第一预设阈值,排除用户与参考声纹模板对应的归属者之间的声纹信息的相似度高,但用户与归属者之间并非同一人的情况。本方案给出了如何根据映射固定码对应的多个声纹模板确定用户身份是否识别成功的可选实施方案,增加了映射固定码的利用率,同时保证了门禁系统识别的准确性。
进一步地,参见图4,基于本申请门禁系统的识别方法的第一实施例提出本申请门禁系统的识别方法的第三实施例,在本实施例中,所述步骤S20之后还包括:
步骤S71,从所述声纹信息中截取所述映射随机码对应的第一声纹信息以及所述映射固定码对应的第二声纹信息;
为了防止在门禁系统识别过程中,不法分子自己读出随机码,然后播放录入的用户发出的映射随机码对应的声音,以伺机通过门禁系统的识别,还可以从用户发出的声纹信息中截取映射随机码对应的第一声纹信息以及映射固定码对应的第二声纹信息,并比对第一声纹信息与第二声纹信息间的相似度。
步骤S72,计算所述第一声纹信息与所述第二声纹信息间的相似度;
步骤S73,判断所述第一声纹信息与所述第二声纹信息间的相似度是否大于第二预设阈值;若是,则执行步骤S30;若否,则执行步骤S80;
可以理解地是,即便不法分子自己读出随机码,然后再播放录入的映射固定码对应的声音,但由于是两个不同的人发出的声音,其声纹信息必然不同,因此可以通过计算发出映射随机码对应的第一声纹信息与发出映射固定码对应的第二声纹信息间的相似度,只有当第一声纹信息与第二声纹信息间的相似度大于第二预设阈值时,才能确定发出第一声纹信息的用户与发出第二声纹信息的用户为同一个人。其中,第二预设阈值可以与第一预设阈值相同,也可以高于第一预设阈值。
步骤S80,确定所述门禁系统识别错误。
若第一声纹信息与第二声纹信息间的相似度小于或等于第二预设阈值,表示发出第一声纹信息的用户与发出第二声纹信息的用户并非同一人,用户此次门禁识别失败,门禁系统可以通过显示模块发出错误提醒。通过比较映射随机码对应的第一声纹信息与映射固定码对应的第二声纹信息间的相似度,减少了门禁系统识别过程中的漏洞,保证了财产安全以及居民居住安全。
参见图5,进一步地,基于本申请门禁系统的识别方法的第一实施例提出本申请门禁系统的识别方法的第四实施例,在本实施例中,所述步骤S20之后还包括:
步骤S91,根据所述声纹信息,确定所述用户发出所述映射随机码对应的第一声纹信息至发出所述映射固定码对应的第二声纹信息间的间隔时间;
本实施例获取了第一声纹信息与第二声纹信息发出的间隔时间,该间隔时间是指第一声纹信息结束的时间点到第二声纹信息发出的时间点之间的时间差。
步骤S92,判断所述间隔时间是否小于预设间隔阈值;若是,则执行步骤S30;若否,则执行步骤S80;
步骤S80,确定所述门禁系统识别错误。
如果间隔时间过长,会耽误后续用户使用门禁系统,通过预设间隔阈值与间隔时间进行比较设置,提高了门禁系统的识别速度。
可选地,还可以将第三实施例中第一声纹信息与第二声纹信息间的相似度与第二预设阈值进行比较的方案与本实施例中间隔时间与预设间隔阈值比较的方案结合,保证门禁系统识别的安全性,同时提高识别速度。
可选地,在其他实施例中,还可以设置本地犯罪人员声纹库,其中存储有特殊声纹模板,每个特殊声纹模板可以对应一名不法犯罪分子。通过在获取了用户发出的声纹信息,或者门禁系统识别通过/失败后,将声纹信息与本地犯罪人员声纹库中的所有特殊声纹模板进行声纹比对,只要用户发出的声纹信息与特殊声纹模板中任一个的相似度大于第一预设阈值,则可以触发通信模块联网并发送预警信息至管理者。其中,在获取了用户发出的声纹信息后,且用户发出的声纹信息与所有声纹模板间的相似度均小于或等于第一预设阈值时,可以继续执行步骤S20。
通过本地犯罪人员声纹库的设置,从门禁识别的角度给公安机关提供了不法犯罪分子的线索,另一方面由于犯罪人员声纹库设置在本地存储器中,只有当确定发出声纹信息的用户为不法犯罪分子时才触发通信模块进行联网操作,减少了其他通信设备通过通信模块对门禁系统造成干扰,保证了门禁系统的顺利运行。
参见图6,本申请还提供一种门禁系统的识别装置,所述装置可以是用于门禁识别的处理器,也可以是门禁设备,在一实施例中,所述装置包括:
获取模块10,用于获取用户发出的声纹信息;
转化模块20,用于将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;
判断模块30,用于判断所述映射随机码与所述门禁系统显示的随机码是否一致;
查找模块40,用于当所述映射随机码与所述随机码一致时,以所述映射固定码为关键词,从固定码与声纹模板间的预设映射数据中,查找所述映射固定码对应的声纹模板;
比对模块50,用于将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致;
确定模块60,用于当所述用户与该声纹模板的归属者一致时,确定所述门禁系统识别通过。
在另一实施例中,当查找到的所述声纹模板大于或等于2个时,所述比对模块包括:
计算单元,用于分别计算每个声纹模板与所述声纹信息间的相似度;
选择单元,用于从所有声纹模板中选择与所述声纹信息间的所述相似度最高的所述声纹模板作为参考声纹模板;
判断单元,用于判断所述参考声纹模板与所述声纹信息间的相似度是否大于第一预设阈值;
确定单元,用于当所述参考声纹模板与所述声纹信息间的相似度大于所述第一预设阈值时,确定所述用户与该参考声纹模板的归属者一致。
在又一实施例中,所述装置还包括截取模块和计算模块,在所述装置中,
所述截取模块,用于从所述声纹信息中截取所述映射随机码对应的第一声纹信息以及所述映射固定码对应的第二声纹信息;
所述计算模块,用于计算所述第一声纹信息与所述第二声纹信息间的相似度;
所述判断模块,还用于判断所述第一声纹信息与所述第二声纹信息间的相似度是否大于第二预设阈值;并当所述第一声纹信息与所述第二声纹信息间的相似度大于第二预设阈值时,继续执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述第一声纹信息与所述第二声纹信息间的相似度小于或等于第二预设阈值时,触发所述确定模块执行确定所述门禁系统识别错误的步骤。
在另一实施例中,所述确定模块,还用于根据所述声纹信息,确定所述用户发出所述映射随机码对应的第一声纹信息至发出所述映射固定码对应的第二声纹信息间的间隔时间;
所述判断模块,还用于判断所述间隔时间是否小于预设间隔阈值;当所述间隔时间小于预设间隔阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述间隔时间大于或等于预设间隔阈值时,触发所述确定模块执行确定所述门禁系统识别错误的步骤。
在另一实施例中,所述转化模块包括:
预处理单元,用于对所述声纹信息进行预处理;
提取单元,用于从预处理过的所述声纹信息中提取特征参数,并根据所有特征参数生成特征向量序列;
识别单元,用于基于预设模型对所述特征向量序列进行提取和识别,以得到包括映射随机码以及映射固定码的文本数据。
在另一实施例中,所述装置还包括发送模块;在所述装置中,
所述比对模块,还用于将所述声纹信息与本地犯罪人员声纹库中存储的特殊声纹模板进行比对;
所述发送模块,用于当所述声纹信息与所述本地犯罪人员声纹库中任一个特殊声纹模板间的相似度大于第一预设阈值时,发出预警信息至管理者。
在另一实施例中,所述装置还包括:
检测模块,用于通过红外线检测所述门禁系统所在区域内是否有用户出现;并当所述红外线检测到所述门禁系统所在区域内有用户出现时,触发所述获取模块执行获取用户发出的声纹信息的步骤。
本申请还提出一种计算机可读存储介质,其上存储有计算机程序。所述计算机可读存储介质可以是图1的门禁系统100中的存储器20,也可以是如ROM(Read-Only
Memory,只读存储器)/RAM(Random Access
Memory,随机存取存储器)、磁碟、光盘中的至少一种,所述计算机可读存储介质包括若干指令用以使得一台具有处理器的终端设备(可以是手机,计算机,服务器,门禁系统,或者网络设备等)执行本申请各个实施例所述的方法。
需要说明的是,在本文中,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者服务端不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者服务端所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括该要素的过程、方法、物品或者服务端中还存在另外的相同要素。
上述本申请实施例序号仅仅为了描述,不代表实施例的优劣。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。
以上仅为本申请的可选实施例,并非因此限制本申请的专利范围,凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换,或直接或间接运用在其他相关的技术领域,均同理包括在本申请的专利保护范围内。
Claims (21)
- 一种门禁系统的识别方法,其中,所述方法包括步骤:获取用户发出的声纹信息;将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;判断所述映射随机码与所述门禁系统显示的随机码是否一致;当所述映射随机码与所述随机码一致时,以所述映射固定码为关键词,从固定码与声纹模板间的预设映射数据中,查找所述映射固定码对应的声纹模板;将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致;当所述用户与该声纹模板的归属者一致时,确定所述门禁系统识别通过。
- 根据权利要求1所述的门禁系统的识别方法,其中,当查找到的所述声纹模板大于或等于2个时,所述将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致的步骤包括:分别计算每个声纹模板与所述声纹信息间的相似度;从所有声纹模板中选择与所述声纹信息间的所述相似度最高的所述声纹模板作为参考声纹模板;判断所述参考声纹模板与所述声纹信息间的相似度是否大于第一预设阈值;当所述参考声纹模板与所述声纹信息间的相似度大于所述第一预设阈值时,确定所述用户与该参考声纹模板的归属者一致。
- 根据权利要求2所述的门禁系统的识别方法,其中,所述将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据的步骤之后还包括:从所述声纹信息中截取所述映射随机码对应的第一声纹信息以及所述映射固定码对应的第二声纹信息;计算所述第一声纹信息与所述第二声纹信息间的相似度;判断所述第一声纹信息与所述第二声纹信息间的相似度是否大于第二预设阈值;当所述第一声纹信息与所述第二声纹信息间的相似度大于第二预设阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述第一声纹信息与所述第二声纹信息间的相似度小于或等于第二预设阈值时,确定所述门禁系统识别错误;或者,根据所述声纹信息,确定所述用户发出所述映射随机码对应的第一声纹信息至发出所述映射固定码对应的第二声纹信息间的间隔时间;判断所述间隔时间是否小于预设间隔阈值;当所述间隔时间小于预设间隔阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述间隔时间大于或等于预设间隔阈值时,确定所述门禁系统识别错误。
- 根据权利要求2所述的门禁系统的识别方法,其中,所述方法还包括:将所述声纹信息与本地犯罪人员声纹库中存储的特殊声纹模板进行比对;当所述声纹信息与所述本地犯罪人员声纹库中任一个特殊声纹模板间的相似度大于第一预设阈值时,发出预警信息至管理者。
- 根据权利要求1所述的门禁系统的识别方法,其中,所述将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据的步骤之后还包括:从所述声纹信息中截取所述映射随机码对应的第一声纹信息以及所述映射固定码对应的第二声纹信息;计算所述第一声纹信息与所述第二声纹信息间的相似度;判断所述第一声纹信息与所述第二声纹信息间的相似度是否大于第二预设阈值;当所述第一声纹信息与所述第二声纹信息间的相似度大于第二预设阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述第一声纹信息与所述第二声纹信息间的相似度小于或等于第二预设阈值时,确定所述门禁系统识别错误;或者,根据所述声纹信息,确定所述用户发出所述映射随机码对应的第一声纹信息至发出所述映射固定码对应的第二声纹信息间的间隔时间;判断所述间隔时间是否小于预设间隔阈值;当所述间隔时间小于预设间隔阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述间隔时间大于或等于预设间隔阈值时,确定所述门禁系统识别错误。
- 根据权利要求1所述的门禁系统的识别方法,其中,所述将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据的步骤包括:对所述声纹信息进行预处理;从预处理过的所述声纹信息中提取特征参数,并根据所有特征参数生成特征向量序列;基于预设模型对所述特征向量序列进行提取和识别,以得到包括映射随机码以及映射固定码的文本数据。
- 根据权利要求1所述的门禁系统的识别方法,其中,所述方法还包括:将所述声纹信息与本地犯罪人员声纹库中存储的特殊声纹模板进行比对;当所述声纹信息与所述本地犯罪人员声纹库中任一个特殊声纹模板间的相似度大于第一预设阈值时,发出预警信息至管理者。
- 根据权利要求7所述的门禁系统的识别方法,其中,所述将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据的步骤之后还包括:从所述声纹信息中截取所述映射随机码对应的第一声纹信息以及所述映射固定码对应的第二声纹信息;计算所述第一声纹信息与所述第二声纹信息间的相似度;判断所述第一声纹信息与所述第二声纹信息间的相似度是否大于第二预设阈值;当所述第一声纹信息与所述第二声纹信息间的相似度大于第二预设阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述第一声纹信息与所述第二声纹信息间的相似度小于或等于第二预设阈值时,确定所述门禁系统识别错误;或者,根据所述声纹信息,确定所述用户发出所述映射随机码对应的第一声纹信息至发出所述映射固定码对应的第二声纹信息间的间隔时间;判断所述间隔时间是否小于预设间隔阈值;当所述间隔时间小于预设间隔阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述间隔时间大于或等于预设间隔阈值时,确定所述门禁系统识别错误。
- 根据权利要求1所述的门禁系统的识别方法,其中,所述获取用户发出的声纹信息的步骤之前,所述方法还包括:通过红外线检测所述门禁系统所在区域内是否有用户出现;当所述红外线检测到所述门禁系统所在区域内有用户出现时,执行获取用户发出的声纹信息的步骤。
- 一种门禁系统的识别装置,其中,包括:获取模块,设置为获取用户发出的声纹信息;转化模块,设置为将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;判断模块,设置为判断所述映射随机码与所述门禁系统显示的随机码是否一致;查找模块,设置为当所述映射随机码与所述随机码一致时,以所述映射固定码为关键词,从固定码与声纹模板间的预设映射数据中,查找所述映射固定码对应的声纹模板;比对模块,设置为将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致;以及确定模块,设置为当所述用户与该声纹模板的归属者一致时,确定所述门禁系统识别通过。
- 一种门禁系统,其中,所述门禁系统包括:显示模块、存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序,所述计算机程序被所述处理器执行时实现以下步骤:获取用户发出的声纹信息;将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;判断所述映射随机码与所述门禁系统显示的随机码是否一致;当所述映射随机码与所述随机码一致时,以所述映射固定码为关键词,从固定码与声纹模板间的预设映射数据中,查找所述映射固定码对应的声纹模板;将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致;当所述用户与该声纹模板的归属者一致时,确定所述门禁系统识别通过。
- 根据权利要求11所述的门禁系统,其中,所述计算机程序被所述处理器执行时还实现以下步骤:当查找到的所述声纹模板大于或等于2个时,分别计算每个声纹模板与所述声纹信息间的相似度;从所有声纹模板中选择与所述声纹信息间的所述相似度最高的所述声纹模板作为参考声纹模板;判断所述参考声纹模板与所述声纹信息间的相似度是否大于第一预设阈值;当所述参考声纹模板与所述声纹信息间的相似度大于所述第一预设阈值时,确定所述用户与该参考声纹模板的归属者一致。
- 根据权利要求12所述的门禁系统,其中,所述计算机程序被所述处理器执行时还实现以下步骤:从所述声纹信息中截取所述映射随机码对应的第一声纹信息以及所述映射固定码对应的第二声纹信息;计算所述第一声纹信息与所述第二声纹信息间的相似度;判断所述第一声纹信息与所述第二声纹信息间的相似度是否大于第二预设阈值;当所述第一声纹信息与所述第二声纹信息间的相似度大于第二预设阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述第一声纹信息与所述第二声纹信息间的相似度小于或等于第二预设阈值时,确定所述门禁系统识别错误;或者,根据所述声纹信息,确定所述用户发出所述映射随机码对应的第一声纹信息至发出所述映射固定码对应的第二声纹信息间的间隔时间;判断所述间隔时间是否小于预设间隔阈值;当所述间隔时间小于预设间隔阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述间隔时间大于或等于预设间隔阈值时,确定所述门禁系统识别错误。
- 根据权利要求11所述的门禁系统,其中,所述计算机程序被所述处理器执行时还实现以下步骤:对所述声纹信息进行预处理;从预处理过的所述声纹信息中提取特征参数,并根据所有特征参数生成特征向量序列;基于预设模型对所述特征向量序列进行提取和识别,以得到包括映射随机码以及映射固定码的文本数据。
- 根据权利要求11所述的门禁系统,其中,所述计算机程序被所述处理器执行时还实现以下步骤:将所述声纹信息与本地犯罪人员声纹库中存储的特殊声纹模板进行比对;当所述声纹信息与所述本地犯罪人员声纹库中任一个特殊声纹模板间的相似度大于第一预设阈值时,发出预警信息至管理者。
- 根据权利要求15所述的门禁系统,其中,所述计算机程序被所述处理器执行时还实现以下步骤:从所述声纹信息中截取所述映射随机码对应的第一声纹信息以及所述映射固定码对应的第二声纹信息;计算所述第一声纹信息与所述第二声纹信息间的相似度;判断所述第一声纹信息与所述第二声纹信息间的相似度是否大于第二预设阈值;当所述第一声纹信息与所述第二声纹信息间的相似度大于第二预设阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述第一声纹信息与所述第二声纹信息间的相似度小于或等于第二预设阈值时,确定所述门禁系统识别错误;或者,根据所述声纹信息,确定所述用户发出所述映射随机码对应的第一声纹信息至发出所述映射固定码对应的第二声纹信息间的间隔时间;判断所述间隔时间是否小于预设间隔阈值;当所述间隔时间小于预设间隔阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述间隔时间大于或等于预设间隔阈值时,确定所述门禁系统识别错误。
- 一种计算机可读存储介质,其中,所述计算机可读存储介质上存储有计算机程序,所述计算机程序被处理器执行时实现以下步骤:获取用户发出的声纹信息;将所述声纹信息转化为对应的包括映射随机码以及映射固定码的文本数据;判断所述映射随机码与所述门禁系统显示的随机码是否一致;当所述映射随机码与所述随机码一致时,以所述映射固定码为关键词,从固定码与声纹模板间的预设映射数据中,查找所述映射固定码对应的声纹模板;将所述声纹信息与查找到的声纹模板进行声纹比对,以确定所述用户与该声纹模板的归属者是否一致;当所述用户与该声纹模板的归属者一致时,确定所述门禁系统识别通过。
- 根据权利要求17所述的计算机可读存储介质,其中,所述计算机程序被处理器执行时还实现以下步骤:当查找到的所述声纹模板大于或等于2个时,分别计算每个声纹模板与所述声纹信息间的相似度;从所有声纹模板中选择与所述声纹信息间的所述相似度最高的所述声纹模板作为参考声纹模板;判断所述参考声纹模板与所述声纹信息间的相似度是否大于第一预设阈值;当所述参考声纹模板与所述声纹信息间的相似度大于所述第一预设阈值时,确定所述用户与该参考声纹模板的归属者一致。
- 根据权利要求17所述的计算机可读存储介质,其中,所述计算机程序被处理器执行时还实现以下步骤:从所述声纹信息中截取所述映射随机码对应的第一声纹信息以及所述映射固定码对应的第二声纹信息;计算所述第一声纹信息与所述第二声纹信息间的相似度;判断所述第一声纹信息与所述第二声纹信息间的相似度是否大于第二预设阈值;当所述第一声纹信息与所述第二声纹信息间的相似度大于第二预设阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述第一声纹信息与所述第二声纹信息间的相似度小于或等于第二预设阈值时,确定所述门禁系统识别错误;或者,根据所述声纹信息,确定所述用户发出所述映射随机码对应的第一声纹信息至发出所述映射固定码对应的第二声纹信息间的间隔时间;判断所述间隔时间是否小于预设间隔阈值;当所述间隔时间小于预设间隔阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述间隔时间大于或等于预设间隔阈值时,确定所述门禁系统识别错误。
- 根据权利要求17所述的计算机可读存储介质,其中,所述计算机程序被处理器执行时还实现以下步骤:将所述声纹信息与本地犯罪人员声纹库中存储的特殊声纹模板进行比对;当所述声纹信息与所述本地犯罪人员声纹库中任一个特殊声纹模板间的相似度大于第一预设阈值时,发出预警信息至管理者。
- 根据权利要求20所述的计算机可读存储介质,其中,所述计算机程序被处理器执行时还实现以下步骤:从所述声纹信息中截取所述映射随机码对应的第一声纹信息以及所述映射固定码对应的第二声纹信息;计算所述第一声纹信息与所述第二声纹信息间的相似度;判断所述第一声纹信息与所述第二声纹信息间的相似度是否大于第二预设阈值;当所述第一声纹信息与所述第二声纹信息间的相似度大于第二预设阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述第一声纹信息与所述第二声纹信息间的相似度小于或等于第二预设阈值时,确定所述门禁系统识别错误;或者,根据所述声纹信息,确定所述用户发出所述映射随机码对应的第一声纹信息至发出所述映射固定码对应的第二声纹信息间的间隔时间;判断所述间隔时间是否小于预设间隔阈值;当所述间隔时间小于预设间隔阈值时,执行判断所述映射随机码与所述门禁系统显示的随机码是否一致的步骤;当所述间隔时间大于或等于预设间隔阈值时,确定所述门禁系统识别错误。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810737583.4 | 2018-07-06 | ||
| CN201810737583.4A CN109727342A (zh) | 2018-07-06 | 2018-07-06 | 门禁系统的识别方法、装置、门禁系统及存储介质 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020006886A1 true WO2020006886A1 (zh) | 2020-01-09 |
Family
ID=66293885
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2018/106772 Ceased WO2020006886A1 (zh) | 2018-07-06 | 2018-09-20 | 门禁系统的识别方法、装置、门禁系统及存储介质 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109727342A (zh) |
| WO (1) | WO2020006886A1 (zh) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110232760A (zh) * | 2019-06-13 | 2019-09-13 | 珠海格力电器股份有限公司 | 智能音箱、智能门锁、服务器、开锁系统及其开锁方法 |
| CN112309060A (zh) * | 2019-08-02 | 2021-02-02 | 广东美的制冷设备有限公司 | 安防设备及其室内监控方法、控制装置和可读存储介质 |
| CN111597531A (zh) * | 2020-04-07 | 2020-08-28 | 北京捷通华声科技股份有限公司 | 一种身份认证方法、装置、电子设备及可读存储介质 |
| CN112863025A (zh) * | 2021-01-11 | 2021-05-28 | 宁波大学 | 一种基于音频密钥识别的门禁控制方法及系统 |
| CN114582346A (zh) * | 2022-03-07 | 2022-06-03 | 云知声智能科技股份有限公司 | 一种识别的方法、装置、终端及存储介质 |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104202486A (zh) * | 2014-09-26 | 2014-12-10 | 上海华勤通讯技术有限公司 | 移动终端及其屏幕解锁方法 |
| CN104765996A (zh) * | 2014-01-06 | 2015-07-08 | 讯飞智元信息科技有限公司 | 声纹密码认证方法及系统 |
| CN106340298A (zh) * | 2015-07-06 | 2017-01-18 | 南京理工大学 | 融合内容识别和话者识别的声纹解锁方法 |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102869008A (zh) * | 2011-07-07 | 2013-01-09 | 上海先先信息科技有限公司 | 一种应用声纹验证作为移动应用安全密码的技术 |
| JP5704013B2 (ja) * | 2011-08-02 | 2015-04-22 | ソニー株式会社 | ユーザ認証方法、ユーザ認証装置、およびプログラム |
| CN103390305B (zh) * | 2013-07-15 | 2015-09-02 | 厦门大学 | 文本提示型声纹门禁系统 |
| CN104168270B (zh) * | 2014-07-31 | 2016-01-13 | 腾讯科技(深圳)有限公司 | 身份验证方法、服务器、客户端及系统 |
| CN105701887A (zh) * | 2014-11-26 | 2016-06-22 | 常州峰成科技有限公司 | 一种声纹锁及其开锁方法 |
| CN204496911U (zh) * | 2015-04-14 | 2015-07-22 | 时代亿宝(北京)科技有限公司 | 一种基于动态口令的声纹打卡装置 |
| CN105096432A (zh) * | 2015-09-14 | 2015-11-25 | 山东山大鸥玛软件有限公司 | 一种声纹密码锁控制系统 |
| CN106920303A (zh) * | 2017-01-21 | 2017-07-04 | 云丁网络技术(北京)有限公司 | 一种基于语音识别的开锁方法及其智能门锁系统 |
| CN107977557A (zh) * | 2017-11-30 | 2018-05-01 | 广州势必可赢网络科技有限公司 | 一种基于声纹识别的电子合同签署服务认证协议及服务器 |
| CN108417216B (zh) * | 2018-03-15 | 2021-01-08 | 深圳市声扬科技有限公司 | 语音验证方法、装置、计算机设备和存储介质 |
-
2018
- 2018-07-06 CN CN201810737583.4A patent/CN109727342A/zh active Pending
- 2018-09-20 WO PCT/CN2018/106772 patent/WO2020006886A1/zh not_active Ceased
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104765996A (zh) * | 2014-01-06 | 2015-07-08 | 讯飞智元信息科技有限公司 | 声纹密码认证方法及系统 |
| CN104202486A (zh) * | 2014-09-26 | 2014-12-10 | 上海华勤通讯技术有限公司 | 移动终端及其屏幕解锁方法 |
| CN106340298A (zh) * | 2015-07-06 | 2017-01-18 | 南京理工大学 | 融合内容识别和话者识别的声纹解锁方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109727342A (zh) | 2019-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020006886A1 (zh) | 门禁系统的识别方法、装置、门禁系统及存储介质 | |
| WO2020207035A1 (zh) | 骚扰电话拦截方法、装置、设备及存储介质 | |
| WO2018070780A1 (en) | Electronic device and method for controlling the same | |
| WO2019144738A1 (zh) | 金融业务的验证方法、装置、设备和计算机存储介质 | |
| WO2020206899A1 (zh) | 基于时间戳的身份验证方法、装置、设备及存储介质 | |
| WO2020042463A1 (zh) | 基于生物特征识别的门禁解锁方法、装置、设备和介质 | |
| WO2016129930A1 (en) | Operating method for voice function and electronic device supporting the same | |
| WO2018034552A1 (en) | Language translation device and language translation method | |
| WO2013125910A1 (en) | Method and system for authenticating user of a mobile device via hybrid biometics information | |
| WO2021051558A1 (zh) | 基于知识图谱的问答方法、装置和存储介质 | |
| WO2020258657A1 (zh) | 异常检测方法、装置、计算机设备及存储介质 | |
| WO2020122653A1 (en) | Electronic apparatus and controlling method thereof | |
| WO2020000637A1 (zh) | 基于声音的设备控制方法、装置、设备及存储介质 | |
| WO2019174090A1 (zh) | 截屏文件分享的控制方法、装置、设备和计算机存储介质 | |
| CN111462754B (zh) | 一种电力系统调度控制语音识别模型建立方法 | |
| WO2019112145A1 (ko) | 음성인식 기반의 사진 공유 방법, 장치 및 시스템 | |
| WO2019024485A1 (zh) | 数据共享方法、装置及计算机可读存储介质 | |
| WO2020073494A1 (zh) | 网页后门检测方法、设备、存储介质及装置 | |
| WO2023128342A1 (ko) | 동형 암호화된 음성을 이용한 개인 식별 방법 및 시스템 | |
| WO2018223520A1 (zh) | 面向儿童的学习方法、学习设备及存储介质 | |
| US10997976B2 (en) | Authentication system, authentication method, and, non-transitory computer-readable information recording medium for recording program | |
| WO2021251539A1 (ko) | 인공신경망을 이용한 대화형 메시지 구현 방법 및 그 장치 | |
| CN113947376B (zh) | 基于多重生物特征的c/s打卡方法和装置 | |
| WO2023128345A1 (ko) | 동형 암호화된 이미지를 이용한 개인 식별 방법 및 시스템 | |
| WO2018014593A1 (zh) | 基于大数据的风险预测方法、装置、服务器及存储介质 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18925549 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 16.04.2021) |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 18925549 Country of ref document: EP Kind code of ref document: A1 |