WO2023274028A1 - Procédé et appareil de commande de verrou à empreinte vocale, et dispositif électronique - Google Patents

Procédé et appareil de commande de verrou à empreinte vocale, et dispositif électronique Download PDF

Info

Publication number
WO2023274028A1
WO2023274028A1 PCT/CN2022/100787 CN2022100787W WO2023274028A1 WO 2023274028 A1 WO2023274028 A1 WO 2023274028A1 CN 2022100787 W CN2022100787 W CN 2022100787W WO 2023274028 A1 WO2023274028 A1 WO 2023274028A1
Authority
WO
WIPO (PCT)
Prior art keywords
voiceprint
information
registered
matching
voiceprint information
Prior art date
Application number
PCT/CN2022/100787
Other languages
English (en)
Chinese (zh)
Inventor
侯涛
任川霞
李玲
Original Assignee
汉王科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 汉王科技股份有限公司 filed Critical 汉王科技股份有限公司
Publication of WO2023274028A1 publication Critical patent/WO2023274028A1/fr

Links

Images

Classifications

    • 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
    • G07C9/00174—Electronically operated locks; Circuits therefor; Nonmechanical keys therefor, e.g. passive or active electrical keys or other data carriers without mechanical keys
    • G07C9/00563—Electronically operated locks; Circuits therefor; Nonmechanical keys therefor, e.g. passive or active electrical keys or other data carriers without mechanical keys using personal physical data of the operator, e.g. finger prints, retinal images, voicepatterns
    • 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
    • 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 embodiments of the present application relate to the security field, and in particular, to a voiceprint lock control method and device, a voiceprint lock, and a computer-readable storage medium.
  • voiceprint recognition technology As a branch of biometric identification technology, voiceprint recognition technology is developing rapidly. From time attendance to smart access control, to smart payment, voiceprint recognition technology has been more and more widely used. In the application of voiceprint recognition in some cases, it is first necessary to establish the user's voiceprint feature library, and then, in the voiceprint recognition stage, by collecting the voice input by the user and extracting the voiceprint features, after that, the extracted voiceprint The feature is compared with the voiceprint feature in the pre-established voiceprint feature library to determine whether the identification is passed. The establishment of the voiceprint feature library directly affects the efficiency of voiceprint recognition.
  • voiceprint lock As a weakly interactive voiceprint recognition application, often only collects a limited amount of voiceprint information to build a voiceprint feature library in order to improve user convenience and user experience. Security risks.
  • An embodiment of the present application provides a method for controlling a voiceprint lock, which helps to improve the security of the voiceprint lock.
  • the embodiment of the present application provides a voiceprint lock control method, including: outputting to the user the perception information of the voiceprint matching object stored in the voiceprint lock; Feedback the voice, determine the voiceprint information to be recognized in the feedback voice; perform voiceprint matching on the voiceprint information to be recognized and the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock, and determine The matching result obtained by performing the voiceprint matching; in response to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, performing a process based on the voiceprint to be recognized information to update the registration voiceprint information database, and/or execute the unlocking control operation of the voiceprint lock.
  • the embodiment of the present application provides a voiceprint lock control device, including: a collection information guidance module, configured to output to the user the perception information of the voiceprint matching object stored in the voiceprint lock; a voiceprint information collection module, It is used to determine the voiceprint information to be recognized in the feedback voice in response to the feedback voice for the perceptual information collected by the voiceprint lock; the voiceprint matching module is used to combine the voiceprint information to be recognized with the voiceprint information
  • the registered voiceprint information in the registered voiceprint information database stored in the fingerprint lock performs voiceprint matching, and determines the matching result obtained by performing the voiceprint matching;
  • the registration voiceprint information database update module is used to indicate in response to the matching result
  • the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, and an update operation is performed on the registered voiceprint information database based on the voiceprint information to be recognized; and, the voiceprint lock A control module, configured to perform an unlocking control operation of the voiceprint lock in response to the matching result indicating that the voice
  • the embodiment of the present application provides a voiceprint lock, including: a main control device, a voice collection device, a memory, a lock actuator, and a sensory information output device respectively connected to the main control device in communication, wherein the The memory is used to store a dictionary database and a registered voiceprint information database, wherein the registered voiceprint information database stores: voiceprint matching objects, and registration voiceprint information corresponding to the corresponding voiceprint matching;
  • the information output device is configured to output the perceptual information presenting the specified voiceprint matching object according to the control of the main control device, wherein the output presenting the perceptual information of the specified voiceprint matching object includes: displaying the information of the specified voiceprint matching object text; and/or, play the reading voice of the specified voiceprint matching object;
  • the voice collection device is used to collect the user's feedback voice for the perceptual information according to the control of the main control device, and collect the feedback
  • the voice is sent to the main control device; the main control device is also used to determine the voiceprint information to be recognized in the feedback
  • Perform voiceprint matching on the registered voiceprint information in the voiceprint information database determine the matching result obtained by performing the voiceprint matching, and indicate the voiceprint information to be recognized and the registered voiceprint information in response to the matching result If the registered voiceprint information in the database is successfully matched, an update operation of the registered voiceprint information database is performed based on the voiceprint information to be recognized; and/or, an unlocking control operation of the voiceprint lock is performed.
  • the embodiment of the present application also discloses an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor.
  • the processor executes the computer program, it realizes The voiceprint lock control method described in the embodiment of this application.
  • the embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and the program is executed by a processor as the steps of the voiceprint lock control method disclosed in the embodiment of the present application.
  • Fig. 1 is one of the flow charts of the voiceprint lock control method according to an embodiment of the present application
  • Fig. 2 is one of the structural schematic diagrams of the voiceprint lock according to an embodiment of the present application.
  • Fig. 3 is the second structural diagram of the voiceprint lock according to an embodiment of the present application.
  • Fig. 4 is the second flow chart of the voiceprint lock control method according to an embodiment of the present application.
  • Fig. 5 is a flowchart of a voiceprint lock control method according to another embodiment of the present application.
  • FIG. 6 is a flow chart of a voiceprint lock control method according to another embodiment of the present application.
  • Fig. 7 is one of the structural schematic diagrams of a voiceprint lock control device according to another embodiment of the present application.
  • Fig. 8 is the second structural diagram of a voiceprint lock control device according to another embodiment of the present application.
  • Fig. 9 is a schematic structural diagram of a voiceprint lock control device according to another embodiment of the present application.
  • Fig. 10 is a schematic structural diagram of a voiceprint lock control device according to another embodiment of the present application.
  • An embodiment of the present application discloses a voiceprint lock control method, as shown in FIG. 1 , the method includes: step 110 to step 140 .
  • Step 110 output the perception information of the voiceprint matching object stored in the voiceprint lock to the user.
  • the voiceprint lock control method described in the embodiment of the present application is applied to the voiceprint lock shown in FIG.
  • a voice collection device 220 , a memory 230 , a lock actuator 240 and a sensory information output device 250 are connected in communication.
  • the main control device 210 may be a processor with storage space.
  • the memory 230 is built in the main control device 210 .
  • the voiceprint matching objects described in the embodiments of the present application may be characters, words or sentences.
  • a dictionary library is pre-stored in the voiceprint lock, and the dictionary library is stored in the memory 230.
  • the dictionary library may include one of characters, words, sentences, and music clips. or multiple forms of voiceprint matching objects.
  • the voiceprint matching objects (for example, characters, words, sentences, music fragments) stored in the dictionary database can be used to output to the user as voiceprint matching objects that need to register voiceprint information during the voiceprint registration stage, or can be used to In the voiceprint authentication stage, it is output to the user as a voiceprint matching object that prompts the user to perform voiceprint authentication.
  • Step 100 Before outputting the perception information of the voiceprint matching object stored in the voiceprint lock to the user, it further includes: Step 100 .
  • Step 100 register voiceprint information.
  • One or more pieces of registration voiceprint information generated when the user performs voiceprint registration will be stored in the registration voiceprint information database, and the registration voiceprint information database is stored in the voiceprint lock, for example, the registration voiceprint information database stores in the memory 230 .
  • the main control device 210 randomly selects a voiceprint matching object from the dictionary library, and outputs the perception information through the perception information output device 250.
  • the output of the perception information includes at least three ways, including The output device 250 displays the text content of the selected voiceprint matching object, or plays the voice of the selected voiceprint matching object, or displays the text content of the selected voiceprint matching object and plays the corresponding voice, and prompts the user for the selected voiceprint matching object.
  • the main control device 210 controls the voice collection device 220 to collect the user's feedback voice for the perceptual information, and extracts the voiceprint features in the feedback voice, as the selected voice print Then, correspondingly store the selected voiceprint matching object and the extracted registered voiceprint information in the registered voiceprint information database.
  • each piece of registered voiceprint information stored in the registered voiceprint information database corresponds to a different voiceprint matching object, and the voiceprint matching object can be a single character, a single word, a complete sentence or music fragment.
  • each piece of registered voiceprint information stored in the registered voiceprint information database corresponds to a different character.
  • each piece of registered voiceprint information stored in the registered voiceprint information database corresponds to a different word, or corresponds to a different sentence, or corresponds to a piece of music (pure melody music or music with music with lyrics).
  • some of the registered voiceprint information stored in the registered voiceprint information database correspond to characters, some correspond to words, and some correspond to music clips.
  • the voiceprint information corresponding to different registered voiceprint information Match objects vary.
  • Each piece of registered voiceprint information in the registered voiceprint information database stored by the voiceprint lock is stored in association with its corresponding voiceprint matching object.
  • the voiceprint matching object stored in the registered voiceprint information database is used to output to the user as a voiceprint matching object that prompts the user to perform voiceprint authentication during the voiceprint authentication stage.
  • the voiceprint matching object described in the embodiment of the present application can be stored in the dictionary database, or can be stored in the registered voiceprint information database, or can be the voiceprint matching object stored in the dictionary database and the registered voiceprint information
  • the voiceprint matching object formed by combining the voiceprint matching objects in the library can be a word stored in a dictionary, or a character or a word stored in a registered voiceprint information database, or a word and a word stored in a registered voiceprint information database. Words or sentences formed by combining the words in .
  • the voiceprint lock first outputs to the user the perception information of the voiceprint matching object that requires voiceprint authentication, and then further collects the voiceprint information to be recognized input by the user for the voiceprint matching object.
  • the perceptual information output method of the voiceprint matching object includes displaying text, reading voice, and/or playing music clips.
  • the perceptual information output device 250 is used to output the perceptual information of the voiceprint matching object
  • the voice collection device 220 is used to collect the feedback voice input by the user for the voiceprint matching object, so as to facilitate The voiceprint information to be recognized is extracted from the collected feedback voice.
  • the main control device 210 is used to control the perception information output device 250 to output the perception information of the voiceprint matching object stored in the voiceprint lock to the user.
  • outputting the perception information of the voiceprint matching object stored in the voiceprint lock to the user includes: showing the user the text of the voiceprint matching object stored in the voiceprint lock, playing the text of the voiceprint matching object stored in the voiceprint lock to the user.
  • the voiceprint lock reads any voiceprint matching object from the dictionary library, and displays the text of the voiceprint matching object to the user, and at the same time, plays the reading voice of the voiceprint matching object to prompt the user to input
  • the voiceprint lock arbitrarily selects a voiceprint matching object from the registered voiceprint information database, and displays the text of the voiceprint matching object, so as to prompt the user to input the waiting time corresponding to the voiceprint matching object by reading the text content aloud. Identify voiceprint information.
  • the reading voice of the voiceprint matching object may be the reading voice of the voiceprint matching object pre-stored in the dictionary database, or it may be the text-to-speech module built in the voiceprint lock according to the voiceprint matching object's reading voice.
  • the reading voice obtained by text conversion.
  • Step 120 in response to the voiceprint lock collecting the feedback voice for the perceptual information, determine voiceprint information to be recognized in the feedback voice.
  • the main control device 210 is further configured to determine the voice to be recognized in the feedback voice in response to the feedback voice for the perception information collected by the voiceprint lock. pattern information.
  • the main control device 210 controls the voice collection device 220 to collect the user's feedback voice for the voiceprint matching object "Zesame".
  • the voice collection device 220 will collect the audio data of "zhima” (that is, the feedback voice) and send it to the main control device 210 .
  • the main control device 210 receives the audio data of "zhima", it determines the voiceprint information to be recognized in the feedback voice through a voiceprint extraction algorithm.
  • the main control device 210 can also use other technical means in the prior art to determine the voiceprint information to be recognized in the feedback voice.
  • the specific implementation manner of recognizing the voiceprint information is not limited.
  • Step 130 perform voiceprint matching on the voiceprint information to be recognized and the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock, and determine the matching result obtained by the voiceprint matching.
  • the main control device 210 is further configured to combine the voiceprint information to be recognized with the registered voiceprint information stored in the voiceprint lock Perform voiceprint matching on the registered voiceprint information in the library, and determine the matching result obtained from the voiceprint matching.
  • multiple voiceprint matching means can be used to perform voiceprint matching between the voiceprint information to be recognized and the registered voiceprint information. Examples are given below.
  • the first solution is to perform overall matching on the voiceprint information to be recognized.
  • the obtained matching result includes: in response to the registered voiceprint information corresponding to the voiceprint matching object in the registered voiceprint information database stored in the voiceprint lock, combining the voiceprint information to be identified with the registered voiceprint information performing voiceprint matching on registered voiceprint information corresponding to the voiceprint matching object in the voiceprint information database, and determining a matching result obtained by performing voiceprint matching.
  • the main control device 210 controls the voice collection device 220 to collect the user's feedback voice for the voiceprint matching object "Sesame", and extract the voiceprint in the feedback voice Information, as the voiceprint information to be recognized.
  • the main control device 210 judges whether the voiceprint matching object "Sesame" has registered voiceprint information.
  • the voiceprint lock can mark the voiceprint matching object that has registered registered voiceprint information.
  • the voiceprint matching object "Zesame” After it is determined that the voiceprint matching object "Zesame” has registered voiceprint information, further obtain the registered voiceprint information correspondingly stored in the registered voiceprint information database of the voiceprint matching object "Zesame", and store the acquired voiceprint matching object " Sesame” performs voiceprint matching between the registered voiceprint information stored in the registered voiceprint information database and the currently collected voiceprint information to be recognized, and determines the voiceprint matching result.
  • the obtained matching result includes: in response to the fact that there is no registered voiceprint information corresponding to the voiceprint matching object in the registered voiceprint information database stored in the voiceprint lock, combining the voiceprint information to be identified with the voiceprint information Perform voiceprint matching on any one or more pieces of registered voiceprint information in the registered voiceprint information database stored in the fingerprint lock, and determine the matching result obtained by performing the voiceprint matching.
  • the main control device 210 controls the voice collection device 220 to collect the user's feedback voice for the voiceprint matching object "Apple”, and extract the voiceprint in the feedback voice Information, as the voiceprint information to be recognized.
  • the main control device 210 judges whether the voiceprint matching object "Apple” has registered voiceprint information. In some embodiments of the present application, it may be determined by comparing the voiceprint matching object "apple” with the voiceprint matching object corresponding to the registered voiceprint information stored in the registered voiceprint information database of the voiceprint lock.
  • the voiceprint lock can mark the voiceprint matching object that has registered registered voiceprint information.
  • the voiceprint matching phase the Whether the voiceprint matching object has registered voiceprint information.
  • the voiceprint matching result may be determined by calculating the similarity between the registered voiceprint information and the voiceprint information to be recognized. For example, if the similarity between the registered voiceprint information and the voiceprint information to be identified is greater than the preset similarity threshold, it can be considered that the registered voiceprint information and the voiceprint information to be identified are successfully matched; otherwise, the registered voiceprint information and the voiceprint information to be identified are considered to be Pattern information matching failed.
  • different similarity thresholds can be set for different situations of whether the voiceprint matching object has registered voiceprint information, so as to reduce the influence of the text content corresponding to the voiceprint information on the voiceprint information similarity calculation. For example, when the voiceprint matching object has registered voiceprint information, and the voiceprint information to be recognized is compared with the registered voiceprint information corresponding to the voiceprint matching object, the preset similarity threshold can be set to the first Threshold; when the voiceprint matching object does not have registered voiceprint information, the voiceprint information to be identified is compared with any piece of registered voiceprint information, and the preset similarity threshold can be set to the second threshold, where the first A threshold is greater than the second threshold.
  • the registered voiceprint information in the registered voiceprint information database includes the registered voiceprint information stored in the original registration stage, and may also include the registered voiceprint information stored in the voiceprint authentication stage when the registered voiceprint information database is updated and maintained. Register voiceprint information.
  • the second solution is to perform segmentation matching on the voiceprint information to be recognized.
  • the voiceprint matching object output by the voiceprint lock can also be a voiceprint matching object composed of multiple voiceprint matching objects with unregistered voiceprint information in the dictionary database (For example, for the voiceprint matching objects "Qing" and “Lu” that have not registered voiceprint information in the dictionary database, they can be combined into the voiceprint matching object "Qinglu"), or unregistered voiceprint information in the dictionary database
  • a voiceprint matching object formed by combining one or more voiceprint matching objects in the registered voiceprint information database with one or more voiceprint matching objects stored in the registered voiceprint information database for example, for a voiceprint that has no registered voiceprint information in the dictionary database
  • the matching object "Qing" and the voiceprint matching object "Sesame” stored in the registered voiceprint information database can be combined into the voiceprint matching object "Qingzhima"), or one or more voiceprint matching objects stored in the registered voiceprint information database.
  • Voiceprint matching objects formed by combining voiceprint matching objects for example, for the voiceprint
  • segmentation matching can be performed on the voiceprint information to be recognized.
  • Combine at least two voiceprint matching objects to generate a composite voiceprint matching object which is used for voiceprint authentication in the voiceprint authentication stage, and randomly combine to generate voiceprint matching objects, which can improve the security of the voiceprint lock.
  • select the voiceprint matching object of unregistered voiceprint information from the dictionary library and the voiceprint matching object selected from the registered voiceprint information database to generate a composite voiceprint matching object and can also realize the silent update of the registered voiceprint information database , thereby further enhancing the security of the voiceprint lock.
  • the following is an example of voiceprint authentication by selecting a voiceprint matching object with unregistered voiceprint information from the dictionary database and a registered voiceprint matching object from the registered voiceprint information database to generate a composite voiceprint matching object.
  • the voiceprint matching object of the registered voiceprint information in the registered voiceprint information database is called "the first voiceprint matching object”
  • the unregistered voiceprint information in the dictionary database is referred to as "first voiceprint matching object”.
  • the voiceprint matching object is called the "second voiceprint matching object”.
  • the first voiceprint matching object and the second voiceprint matching object are basic voiceprint matching units, for example, the first voiceprint matching object and the second voiceprint matching object are a single character or single words.
  • the voiceprint matching object is based on the first voiceprint matching object of the registered voiceprint information stored in the registered voiceprint information database, and the dictionary library stored in the voiceprint lock
  • the second voiceprint matching object in the combination is generated, and the voiceprint matching is performed on the voiceprint information to be recognized and the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock, and the determined
  • the matching result obtained by voiceprint matching includes: determining the pieces of voiceprint information to be recognized corresponding to each of the first voiceprint matching objects and each of the second voiceprint matching objects in the voiceprint information to be recognized; performing voiceprint matching on the voiceprint information segments to be identified corresponding to each of the first voiceprint matching objects and the corresponding registered voiceprint information; and matching the voiceprint information segments to be identified corresponding to each of the second voiceprint matching objects Perform voiceprint matching on the registered voiceprint information selected in the registered voiceprint information database, and determine the voiceprint information to be identified and the voiceprint matching object (i.e.
  • the selected registered voiceprint information includes: a randomly selected piece of registered voiceprint information corresponding to the first voiceprint matching object whose character distance is closest to the second voiceprint matching object registered voiceprint information.
  • the main control device 210 may simultaneously record the first voiceprint matching object and the second voiceprint matching object that make up the composite voiceprint matching object. Afterwards, when extracting the voiceprint information to be recognized, speech recognition can be performed on the feedback voice used to extract the voiceprint information to be recognized at the same time, to obtain a character sequence, and further match the object and the second voiceprint with the first voiceprint according to the character sequence The character matching result of the matching object is divided into the voiceprint information to be recognized, and the voiceprint information segments to be recognized corresponding to each first voiceprint matching object and each second voiceprint matching object included in the composite voiceprint matching object are determined .
  • the voiceprint matching object is composed of the second voiceprint matching object "Qing" that has no voiceprint information registered in the dictionary database and the first voiceprint matching object "Zesame" stored in the registered voiceprint information database.
  • the voiceprint information will be divided into a piece of voiceprint information to be recognized corresponding to the second voiceprint matching object "Qing" (hereinafter denoted as F1), and a piece of waiting-to-be-recognized voiceprint information corresponding to the first voiceprint matching object "Sesame". Identify the piece of voiceprint information (hereinafter denoted as F2).
  • the voiceprint to be recognized is the voiceprint to be recognized.
  • the voiceprint information segment (such as F2) is matched with the registered voiceprint information correspondingly stored in the registered voiceprint information database corresponding to the first voiceprint matching object (such as "Sesame"), and the voiceprint information segment corresponding to the voiceprint information segment to be identified is determined. Voiceprint matching result.
  • the to-be-identified voiceprint information segment can be The voiceprint information segment (such as F1) performs voiceprint matching with any piece of registered voiceprint information stored in the registered voiceprint information database, and determines the voiceprint matching result corresponding to the voiceprint information segment to be recognized.
  • voiceprint matching can be performed on the piece of voiceprint information to be recognized (such as F1) and the registered voiceprint information correspondingly stored in the registered voiceprint information database of the first voiceprint matching object (such as "Sesame"), and determine the voiceprint information to be identified. Identify the voiceprint matching result corresponding to the voiceprint information segment.
  • the pronunciation of the characters in the voice is continuous when the user inputs a complete voice, therefore, for the piece of voiceprint information to be recognized corresponding to the second voiceprint matching object, the character with the closest distance to the second voiceprint matching object is used.
  • the registration voiceprint information corresponding to the first voiceprint matching object performs voiceprint matching, which can improve the matching accuracy of voiceprint authentication.
  • the information of the first voiceprint matching object and the information of the second voiceprint matching object that make up the compound voiceprint matching object can also be output separately, so that the user can input the information for the first voiceprint matching object respectively.
  • a complex voiceprint matching object is generated by selecting a combination of voiceprint matching objects with unregistered voiceprint information from the dictionary library for voiceprint authentication, or a complex voiceprint matching object is generated by selecting a combination of voiceprint matching objects from the registered voiceprint information database
  • voiceprint authentication only segments of the voiceprint information to be identified corresponding to each first voiceprint matching object are segmented out of the voiceprint information to be identified, or only pieces of voiceprint information corresponding to each second voiceprint information are segmented out.
  • the piece of voiceprint information to be recognized corresponding to the fingerprint matching object For the specific implementation manner of performing voiceprint matching on each piece of voiceprint information to be recognized, refer to the foregoing description, which will not be repeated here.
  • the first voiceprint is considered to be The voiceprint matching of the segment of voiceprint information to be identified corresponding to the matching object is successful; the matching degree determined after voiceprint matching between the registration voiceprint information corresponding to the second voiceprint matching object and the corresponding segment of voiceprint information to be identified is greater than that of the second matching degree threshold, it is considered that the voiceprint of the to-be-recognized voiceprint information segment corresponding to the second voiceprint matching object is successfully matched, wherein the first matching degree threshold is greater than the second matching degree threshold.
  • the voiceprint to be recognized is considered to be recognized only when the pieces of voiceprint information to be recognized corresponding to each of the first voiceprint matching object and the second voiceprint matching object that are combined to generate a composite voiceprint matching object are successfully matched.
  • Information matching is successful. If the segment of voiceprint information to be recognized corresponding to a certain first voiceprint matching object or a certain second voiceprint matching object combined to generate a composite voiceprint matching object fails to match, it is considered that the matching of the entire voiceprint information to be recognized is unsuccessful.
  • Step 140 in response to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, perform the processing of the registered voiceprint information based on the voiceprint information to be recognized.
  • the main control device 210 After performing voiceprint matching on the user's voiceprint information to be recognized through the aforementioned steps, if the matching is successful, the main control device 210 will control the lock actuator 240 to open the door lock, and if the matching fails, the door lock will not be opened.
  • the main control device 210 will also execute the registration of the registered voiceprint information database based on the voiceprint information to be recognized. update operation.
  • performing a process based on the voiceprint information to be recognized includes: responding to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, updating the voiceprint information to be recognized.
  • the voiceprint information is stored in the registered voiceprint information database as the registered voiceprint information corresponding to the voiceprint matching object, or the registered voiceprint information in the registered voiceprint information database is updated and maintained.
  • the main control device 210 first determines whether there is corresponding registered voiceprint information in the voiceprint matching object output to the user. , if the voiceprint matching object does not store corresponding registered voiceprint information in the registered voiceprint information database, then update the registered voiceprint information database with the current voiceprint information to be identified; if the voiceprint matching object is registered If the corresponding registered voiceprint information is stored in the voiceprint information database, the registered voiceprint information database is not updated with the current voiceprint information to be recognized, that is, there is no need to operate the registered voiceprint information database.
  • the registered voiceprint information database when updating the registered voiceprint information database with the current voiceprint information to be identified, if the number of registered voiceprint information stored in the registered voiceprint information database stored in the voiceprint lock is greater than or is equal to the preset number threshold, the registered voiceprint information database is updated and maintained through the current voiceprint information to be recognized. For example, delete a set of correspondence data between voiceprint matching objects and registered voiceprint information according to preset rules, and then use the voiceprint information to be recognized that is currently successfully matched with the voiceprint matching object as the registered voiceprint information corresponding to the voiceprint matching object.
  • the voiceprint information is stored in the registered voiceprint information database corresponding to the voiceprint matching object.
  • the voiceprint information to be recognized that is currently successfully matched with the voiceprint matching object is directly used as the registered voiceprint information corresponding to the voiceprint matching object, and stored in the registration voiceprint corresponding to the voiceprint matching object in the information base.
  • the preset number threshold is determined according to the storage capacity of the registered voiceprint information database in the voiceprint lock.
  • performing a process based on the voiceprint information to be recognized includes: responding to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, updating each of the second voiceprint information.
  • the piece of voiceprint information to be matched corresponding to the fingerprint matching object is stored in the registered voiceprint information database as the registered voiceprint information corresponding to each of the second voiceprint matching objects or for the registered voiceprint information database.
  • the voiceprint matching object is a composite voiceprint matching object including the first voiceprint matching object and/or the second voiceprint matching object
  • update Register voiceprint database when the voiceprint matching object is a composite voiceprint matching object including the first voiceprint matching object and/or the second voiceprint matching object.
  • the voiceprint information segment F2 to be recognized corresponding to the first voiceprint matching object "Zhima” and the voiceprint to be recognized corresponding to the second voiceprint matching object "Qing” After the information fragments F1 are all successfully matched, or when the voiceprint information fragment F2 corresponding to the first voiceprint matching object "Sesame" is successfully matched, it is determined that the voiceprint information matching for the voiceprint matching object "Qing Zhima" is matched.
  • the main control device 210 uses the voiceprint information segment F1 corresponding to the second voiceprint matching object "Qing" to be recognized as the registered voiceprint information of the second voiceprint matching object "Qing", and sends the second voiceprint matching object "Qing"
  • the pattern matching object "Qing" and the voiceprint information fragment F1 to be identified are directly stored in the registered voiceprint information database in the voiceprint lock, or a group of voiceprint matching objects and registered voiceprint information are deleted according to preset rules After that, the to-be-recognized voiceprint information segment F1 is used as the registered voiceprint information corresponding to the second voiceprint matching object, and stored corresponding to the second voiceprint matching object in the registered voiceprint information database.
  • the voiceprint matching object in order to further improve the security of voiceprint authentication, can be split into independent voiceprint matching units in the registration stage, so as to facilitate the flexible combination and generation of multiple composites in the voiceprint authentication stage.
  • the voiceprint matching object in order to further improve the security of voiceprint authentication, can be split into independent voiceprint matching units in the registration stage, so as to facilitate the flexible combination and generation of multiple composites in the voiceprint authentication stage.
  • the registration stage further includes: outputting to the user the perception information of the voiceprint matching object to be registered stored in the dictionary database of the voiceprint lock; collecting the registration voice of the user for the perception information of the voiceprint matching object to be registered, And determine the registration voiceprint information corresponding to the registered voice; split the voiceprint matching object to be registered, and determine the independent voiceprint matching objects that form the voiceprint matching object to be registered; determine the voiceprint matching objects respectively Registering the registered voiceprint information segments corresponding to each of the independent voiceprint matching objects in the registered voiceprint information, and then storing each of the registered voiceprint information segments as the registered voiceprint information of the corresponding independent voiceprint matching objects in In the registered voiceprint information database of the voiceprint lock.
  • the following takes the registered voiceprint information corresponding to the originally registered voiceprint matching object "Zhima" as an example to illustrate the voiceprint registration process.
  • the main control device 210 randomly selects the voiceprint matching object "Sesame” from the dictionary database as the voiceprint matching object to be registered, and displays the text content of the voiceprint matching object "Sesame” to be registered through the sensory information output device 250, and/ Or, play the reading voice of the voiceprint matching object "Zesame” to be registered, so as to output the perceptual information of the voiceprint matching object "Zesame” to be registered; after that, prompt the user for the voiceprint matching object "Zesame” to be registered Register the voiceprint information; then, the main control device 210 controls the voice collection device 220 to collect the user's feedback voice on the perceptual information as the registration voice.
  • the voiceprint registration process described here does not need to combine the collected voiceprint features with the registered voiceprint information stored in the registered voiceprint information database. Therefore, the voiceprint registration process that does not need to compare the collected voiceprint features with the registered voiceprint information stored in the registered voiceprint information database is called the original registration process.
  • the main control device 210 splits the voiceprint matching object "Sesame” to be registered according to preset rules (such as splitting by single character), and determines the components that make up the voiceprint matching object "Sesame” to be registered.
  • Two independent voiceprint matching objects “zhi” and “ma” after that, further split the registered voiceprint information corresponding to the registered voice, and determine that the registered voiceprint information matches the independent voiceprint obtained by splitting
  • the independent voiceprint matching object "Ma” and the corresponding registered voiceprint information fragment obtained by splitting are correspondingly stored in the registered voiceprint information database, and a total of two pieces of registered voiceprint data are generated.
  • the independent voiceprint matching object may only include a single character, or may include both a single character and a word.
  • the registered voiceprint information extracted from the user's registered voice can be split into three parts: “I", “Come back” and "Are".
  • the registered voiceprint information segments corresponding to each of the independent voiceprint matching objects are stored as registered voiceprint information of three independent voiceprint matching objects in the registered voiceprint information database.
  • the voiceprint lock control method disclosed in the embodiment of the present application outputs the perception information of the voiceprint matching object stored in the voiceprint lock to the user; in response to the feedback voice for the perception information collected by the voiceprint lock, the feedback is determined.
  • the voiceprint information to be recognized in the voice is matched with the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock, and the voiceprint matching is determined to obtain The matching result; in response to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, performing the registration of the registered voiceprint based on the voiceprint information to be recognized
  • the updating operation of the information base and the execution of the unlocking control operation of the voiceprint lock effectively improve the security of the voiceprint lock.
  • the voiceprint lock control method disclosed in the embodiment of the present application enriches the registered voiceprint information database in the voiceprint lock by updating the registered voiceprint information base in the voiceprint lock based on the currently matched voiceprint information to be identified after the voiceprint is successfully matched.
  • the registered voiceprint information, and the dynamic update of the registered voiceprint information in the registered voiceprint information database is realized, which effectively improves the security of the voiceprint lock.
  • by updating the registered voiceprint information database in the voiceprint lock only based on the voiceprint information to be recognized of the voiceprint matching object that has not registered voiceprint information it can ensure that the registered voiceprint information in the registered voiceprint information database
  • the diversity of fingerprint information further enhances the security of voiceprint locks.
  • a voiceprint lock control method disclosed in another embodiment of the present application is applied to the voiceprint lock shown in FIG. 2 and FIG. 3 , and the method further includes a voice recognition step. As shown in FIG. 5 , the method includes: Step 510 to Step 580 .
  • Step 510 output the perception information of the voiceprint matching object stored in the voiceprint lock to the user.
  • Step 520 Responsive to the feedback voice for the perception information collected by the voiceprint lock, perform voice recognition on the feedback voice, and determine a voice recognition result corresponding to the feedback voice.
  • the main control device 210 of the voiceprint lock After the main control device 210 of the voiceprint lock receives the feedback voice for the perceptual information collected by the voice collection device 220, it first performs voice recognition on the feedback voice, and determines the voice recognition corresponding to the feedback voice. As a result, the content of the text spoken by the user is determined.
  • Step 530 perform text matching on the speech recognition result corresponding to the feedback speech and the speech recognition result corresponding to the registered voiceprint information of the voiceprint matching object in the registered voiceprint information database stored in the voiceprint lock.
  • the voiceprint lock can also perform voice recognition on the registered voice at the same time, and correspond to the voice recognition result of the registered voice and the registered voice
  • the registered voiceprint information and the voiceprint matching object for the registered voice are correspondingly stored in the registered voiceprint information database.
  • the main control device 210 randomly selects the voiceprint matching object "Sesame” from the dictionary database as the voiceprint matching object to be registered, and displays the voiceprint matching object "Sesame” to be registered through the sensory information output device 250
  • the main control device 210 After collecting the registration voice of "Zesame” displayed by the user for the voiceprint lock, extract the voiceprint features in the registration voice as the registration voiceprint information corresponding to the registration voice, that is, as the registration of the voiceprint matching object "Zesame” At the same time, the main control device 210 performs voice recognition on the registered voice, determines the voice recognition result corresponding to the registered voice—the text "Sesame", and matches the currently registered voiceprint with the object (i.e. the word "Sesame"), The speech recognition result corresponding to the registered voice (that is, the text "sesame") and the registered voiceprint information corresponding to the registered voice are associated and stored in the registered voiceprint information database.
  • the voice recognition result corresponding to the registered voiceprint information of the voiceprint matching object is user-defined registration text for the voiceprint matching object.
  • the voiceprint registration stage when the voiceprint lock outputs the perception information of the voiceprint matching object whose voiceprint information is to be registered, the user can customize the voice recognition result corresponding to the voiceprint matching object to improve the security of the voiceprint lock sex.
  • the perceptual information output device 250 displays the text content of the voiceprint matching object to be registered "Sesame", and/or plays the reading voice of the voiceprint matching object to be registered, so as to output the voiceprint matching object to be registered perception information; after that, the user is prompted to register voiceprint information for the voiceprint matching object "Zhima” to be registered, and the user can read "kaimen", which is the pronunciation of the text "open the door", as the registered voice.
  • the voiceprint lock will store the voiceprint matching object (that is, the word “sesame”), the speech recognition result corresponding to the registered voice (that is, the text "open the door"), and the registered voiceprint information corresponding to the registered voice (that is, the user reads "kaimen The voiceprint features extracted from the voice of ”) are associated and stored in the registered voiceprint information base.
  • the main control device 210 receives the feedback voice (such as the pronunciation of the text "open the door") collected by the voice collection device 220 for the perceptual information (such as the text content of the displayed voiceprint matching object "Sesame”) ), performing speech recognition on the feedback speech, and determining the speech recognition result corresponding to the feedback speech (such as the recognized speech recognition result being the text “open the door”), after that, the master control device 210 converts the feedback speech corresponding to The speech recognition result is text-matched with the speech recognition result corresponding to the registered voiceprint information of the voiceprint matching object "Zesame" in the registered voiceprint information database stored in the voiceprint lock (such as the aforementioned text "open the door”).
  • the feedback voice such as the pronunciation of the text "open the door
  • the voice collection device 220 for the perceptual information (such as the text content of the displayed voiceprint matching object "Sesame")
  • the master control device 210 converts the feedback speech corresponding to
  • the speech recognition result is text-matched with the
  • Step 540 determine whether the speech recognition result corresponding to the feedback speech matches the speech recognition result corresponding to the registered voiceprint information of the voiceprint matching object, if so, perform step 550, otherwise, jump to step 580.
  • the determination is performed in response to the text matching result indicating that the speech recognition result corresponding to the feedback voice matches the speech recognition result corresponding to the registered voiceprint information of the voiceprint matching object.
  • the registration voice entered by the user for the voiceprint matching object "Zhima” is the pronunciation of the custom text "Open the door”
  • the voiceprint authentication stage if the user still reads "Open the door” for the voiceprint matching object "Zesame”. ", the speech recognition is successful. If the user reads "Sesame" for the voiceprint matching object "Sesame", it is considered that the speech recognition fails.
  • Step 550 determine voiceprint information to be recognized in the feedback voice.
  • Step 560 Perform voiceprint matching on the voiceprint information to be recognized and the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock, and determine the matching result obtained by performing the voiceprint matching.
  • Step 570 in response to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, perform the processing of the registered voiceprint information based on the voiceprint information to be recognized.
  • the update operation of the library, and/or, the unlocking control operation of the voiceprint lock is executed.
  • Step 580 end the current voiceprint unlocking process.
  • the user by adding a voice recognition step before voiceprint authentication, the user can customize the voice recognition text corresponding to the voiceprint matching object in the voiceprint registration stage, thereby enhancing the voiceprint lock.
  • Anti-attack ability improve the security of voiceprint lock.
  • FIG. 6 another embodiment of the present application discloses a voiceprint lock control method, which is applied to the voiceprint lock shown in FIG. 2 and FIG. 3 , and includes: Step 610 to Step 670 .
  • Step 610 output the perception information of the voiceprint matching object stored in the voiceprint lock to the user.
  • the voiceprint matching object is selected from the voiceprint matching object corresponding to the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock.
  • the main control device 210 randomly selects a voiceprint matching object from the registered voiceprint information database stored in the voiceprint lock as the current voiceprint matching object, and outputs the current voiceprint matching object through the perceptual information output device 250. Perceptual information of the object.
  • Step 620 In response to the voiceprint lock collecting the feedback voice for the perceptual information, determine voiceprint information to be recognized in the feedback voice.
  • Step 630 Perform voiceprint matching on the voiceprint information to be recognized and the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock, and determine the matching result obtained by performing the voiceprint matching.
  • Step 640 In response to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, output the perception information of unregistered voiceprint matching objects in the dictionary to the user .
  • a further round of voiceprint matching is required. Authentication, that is, two rounds of voiceprint authentication are required in this authentication stage, so that the registered voiceprint information database in the voiceprint lock can be updated and maintained at the same time as the voiceprint authentication, so as to improve the security of the voiceprint lock.
  • the unregistered voiceprint matching object in the dictionary is selected from the dictionary database stored in the voiceprint lock and any voiceprint matching object that has no corresponding registered voiceprint information in the registered voiceprint information database.
  • the main control device 210 After determining the unregistered voiceprint matching object in the dictionary, the main control device 210 outputs the perception information of the unregistered voiceprint matching object in the dictionary through the perception information output device 250 .
  • Step 650 In response to the voiceprint lock collecting the first feedback voice for the perception information of the unregistered voiceprint matching object in the dictionary, determine the first voiceprint information to be recognized in the first feedback voice.
  • the main control device 210 collects the perception information for the unregistered voiceprint matching object in the dictionary through the voice collection device 220, and determines the feedback voice, which is the "first feedback voice” in this embodiment, and further determines the first feedback voice.
  • the voiceprint information in the voice is recorded as "the first voiceprint information to be recognized".
  • Step 660 Perform voiceprint matching on the first to-be-recognized voiceprint information and the currently successfully matched registered voiceprint information, and determine the matching result obtained from the current voiceprint matching.
  • the registered voiceprint information that is currently successfully matched is: the registered voiceprint information in the registered voiceprint information database that is successfully matched with the voiceprint information to be identified. Afterwards, perform voiceprint matching on the first voiceprint information to be recognized and the registered voiceprint information used for voiceprint matching in step 630, and determine the matching result obtained from the current voiceprint matching.
  • the voiceprint information to be recognized in the first round of voiceprint authentication has been successfully matched with the registered voiceprint information in the registered voiceprint information database (such as steps 630, 640), Perform voiceprint matching on the first to-be-recognized voiceprint information of the unregistered voiceprint matching object in the dictionary in the second round of voiceprint authentication and the successfully matched registered voiceprint information in the first round of voiceprint authentication, and obtain the matching result.
  • Step 670 In response to the matching result of the current voiceprint matching indicating that the first voiceprint information to be recognized is successfully matched with the registered voiceprint information that is currently successfully matched, the first voiceprint information to be recognized is As the registered voiceprint information corresponding to the unregistered voiceprint matching object in the dictionary, it is stored in the registered voiceprint information database or the registered voiceprint information in the registered voiceprint information database is updated and maintained.
  • the registered voiceprint information is updated and maintained according to the result of the second round of voiceprint authentication. That is, in response to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, performing the matching of the registered voiceprint information database based on the voiceprint information to be recognized
  • the update operation includes: in response to the matching result obtained from the current voiceprint matching indicating that the first voiceprint information to be recognized is successfully matched with the registered voiceprint information that is currently matched successfully, updating the first voiceprint information to be recognized
  • the registered voiceprint information is stored in the registered voiceprint information database or updated and maintained in the registered voiceprint information database as registered voiceprint information corresponding to unregistered voiceprint matching objects in the dictionary.
  • the main control device 210 first selects the voiceprint matching object "Sesame” from the registered voiceprint information database, and outputs the perception information of the voiceprint matching object "Zesame” to the user, and starts the first round of voiceprint authentication. After obtaining the user's feedback voice for the voiceprint matching object "Zhima", and determining the voiceprint information to be recognized in the feedback voice (for a clearer description, denoted as "T1"), the voiceprint information to be recognized T1 and registered The registered voiceprint information corresponding to the voiceprint matching object "Zhima" in the voiceprint information database (referred to as "R1" for a clearer description) is used for voiceprint matching.
  • the main control device 210 selects a voiceprint matching object "Apple” from the dictionary database (the voiceprint matching object "Apple” is not stored in the registered voiceprint information database. , that is, the voiceprint matching object "Apple” has not registered voiceprint information), and output the perception information of the voiceprint matching object "Apple” to the user, and start the second round of voiceprint authentication.
  • the voiceprint information to be recognized in the feedback voice for a clearer description, denoted as "T2"
  • T2 the voiceprint information to be recognized T2
  • the registered voiceprint information R1 corresponding to the voiceprint matching object "Zhima" in the registered voiceprint information database performs voiceprint matching, and determines the voiceprint matching result.
  • the voiceprint information T2 to be recognized is successfully matched with the registered voiceprint information R1 corresponding to the voiceprint matching object "Sesame" in the registered voiceprint information database, the voiceprint information T2 to be recognized will be used as the voiceprint information corresponding to the voiceprint matching object "Apple".
  • the registered voiceprint information is directly stored in the registered voiceprint information database. If the number of registered voiceprint information stored in the registered voiceprint information database is greater than the preset quantity threshold, delete an original piece of registered voiceprint information according to a preset policy.
  • the voiceprint lock control method disclosed in the embodiment of the present application sets two rounds of voiceprint authentication, the first round of voiceprint authentication is for the voiceprint matching object whose voiceprint information has been registered, and the second round of voiceprint authentication is for unregistered voiceprint
  • the voiceprint matching object of the information, and the two rounds of voiceprint matching are matched with the same registered voiceprint information in the registered voiceprint information database, so that after the second round of voiceprint authentication is passed, the voiceprint of unregistered voiceprint information can be
  • the fingerprint matching object registers voiceprint information silently, and dynamically updates the registered voiceprint information database stored in the voiceprint lock to improve the security of the voiceprint lock.
  • a voiceprint lock control device the device includes:
  • the collection information guidance module 710 is used to output the perception information of the voiceprint matching object stored in the voiceprint lock to the user;
  • the voiceprint information collection module 720 is used to determine the voiceprint information to be recognized in the feedback voice in response to the voiceprint lock collecting the feedback voice for the perceptual information;
  • the voiceprint matching module 730 is configured to perform voiceprint matching on the voiceprint information to be recognized and the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock, and determine the voiceprint obtained by performing voiceprint matching. matching result;
  • the registered voiceprint information database update module 740 configured to respond to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, and execute the voiceprint information based on the voiceprint to be recognized Information update operation on the registered voiceprint information database; and/or,
  • the voiceprint lock control module 750 is configured to execute the unlocking control operation of the voiceprint lock in response to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database .
  • outputting the perception information of the voiceprint matching object stored in the voiceprint lock to the user includes: showing the user the text of the voiceprint matching object stored in the voiceprint lock, and/or, playing the text of the voiceprint matching object stored in the voiceprint lock to the user.
  • the voiceprint stored in the voiceprint lock matches the reading voice of the object.
  • the voiceprint matching module 730 is further used to:
  • the registered voiceprint information corresponding to the voiceprint matching object in the registered voiceprint information database stored in the voiceprint lock In response to the registered voiceprint information corresponding to the voiceprint matching object in the registered voiceprint information database stored in the voiceprint lock, combining the voiceprint information to be identified with the registered voiceprint information database in the registered voiceprint information database.
  • the registered voiceprint information corresponding to the voiceprint matching object is used to perform voiceprint matching, and the matching result obtained by performing the voiceprint matching is determined.
  • the voiceprint matching module 730 is further used to:
  • combining the voiceprint information to be identified with the registered voiceprint information stored in the voiceprint lock Perform voiceprint matching on any one or more pieces of registered voiceprint information in the fingerprint information database, and determine the matching result obtained by performing the voiceprint matching.
  • the registration voiceprint information database update module 740 is further used to:
  • the voiceprint information is stored in the registered voiceprint information database or updated and maintained for the registered voiceprint information in the registered voiceprint information database.
  • the voiceprint matching object is based on the first voiceprint matching object of the registered voiceprint information stored in the registered voiceprint information database, and the dictionary library stored in the voiceprint lock
  • the inner second voiceprint matching object combination is generated, and the voiceprint matching module 730 is further used to:
  • performing voiceprint matching on the unrecognized voiceprint information segments corresponding to each of the first voiceprint matching objects and corresponding registered voiceprint information, and performing voiceprint matching on the unidentified voiceprint information segments corresponding to each of the second voiceprint matching Perform voiceprint matching with the registered voiceprint information selected in the registered voiceprint information database, and determine a voiceprint matching result between the voiceprint information to be recognized and the voiceprint matching object.
  • the selected registered voiceprint information includes: a randomly selected piece of registered voiceprint information, a character corresponding to the first voiceprint matching object whose character distance is closest to the second voiceprint matching object Register voiceprint information.
  • the registration voiceprint information database updating module 740 is further used for:
  • the pieces of voiceprint information to be matched corresponding to each of the second voiceprint matching objects are used as The registered voiceprint information corresponding to each of the second voiceprint matching objects is stored in the registered voiceprint information database or the registered voiceprint information in the registered voiceprint information database is updated and maintained.
  • the device further includes: a voiceprint registration module 700, further used for:
  • the fingerprint information is stored in the registered voiceprint information database stored in the voiceprint lock.
  • the voiceprint registration module 700 is further used to:
  • the registered voiceprint information is stored in the registered voiceprint information database stored in the voiceprint lock as the registered voiceprint information of the voiceprint matching object to be registered.
  • the voiceprint lock control device disclosed in the embodiment of the present application outputs the perception information of the voiceprint matching object stored in the voiceprint lock to the user; in response to the feedback voice for the perception information collected by the voiceprint lock, the feedback is determined
  • the voiceprint information to be recognized in the voice is matched with the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock, and the voiceprint matching is determined to obtain The matching result; in response to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database, performing the registration of the registered voiceprint based on the voiceprint information to be recognized
  • the update operation of the information base, and/or, the execution of the unlocking control operation of the voiceprint lock effectively improves the security of the voiceprint lock.
  • the voiceprint lock control device disclosed in the embodiment of the present application after the voiceprint matching is successful, updates the registered voiceprint information database in the voiceprint lock based on the currently successfully matched voiceprint information to be recognized, thereby enriching the registered voiceprint information database.
  • Registered voiceprint information, and dynamically updated registered voiceprint information in the registered voiceprint information database effectively improving the security of the voiceprint lock.
  • updating the registered voiceprint information database in the voiceprint lock only based on the voiceprint information to be recognized of the voiceprint matching object that has not registered voiceprint information, it can ensure that the registered voiceprint information in the registered voiceprint information database
  • the diversity of fingerprint information further enhances the security of voiceprint locks.
  • the voiceprint matching object is selected from the voiceprint matching object corresponding to the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock, and some other implementations of the present application
  • a voiceprint lock control device is disclosed, which is different from the voiceprint lock control device described in the above-mentioned embodiments in that the device further includes: a second round of information collection guidance module 760, a second round of voiceprint information A collection module 770, and a second round of voiceprint matching module 780.
  • the second round of information collection guidance module 760 is configured to output a dictionary to the user in response to the matching result indicating that the voiceprint information to be recognized is successfully matched with the registered voiceprint information in the registered voiceprint information database
  • the second-round voiceprint information collection module 770 is configured to determine the first feedback voice in the first feedback voice in response to the voiceprint lock collecting the first feedback voice for the perception information of the unregistered voiceprint matching object in the dictionary.
  • the first voiceprint information to be recognized;
  • the second-round voiceprint matching module 780 is configured to perform voiceprint matching on the first voiceprint information to be recognized and the registered voiceprint information that is currently successfully matched, and determine the matching result obtained by the current voiceprint matching, wherein, the registered voiceprint information that is currently successfully matched is: the registered voiceprint information in the registered voiceprint information database that is successfully matched with the voiceprint information to be identified.
  • the registration voiceprint information database updating module 740 is further used for:
  • the first voiceprint information to be recognized is used as the first voiceprint information to be recognized.
  • the registered voiceprint information corresponding to the unregistered voiceprint matching object in the dictionary is stored in the registered voiceprint information database or the registered voiceprint information in the registered voiceprint information database is updated and maintained.
  • the voiceprint lock control device disclosed in the embodiment of the present application sets two rounds of voiceprint authentication, the first round of voiceprint authentication is for voiceprint matching objects that have registered voiceprint information, and the second round of voiceprint authentication is for unregistered voiceprint
  • the voiceprint matching object of the information, and the two rounds of voiceprint matching are matched with the same registered voiceprint information in the registered voiceprint information database, so that after the second round of voiceprint authentication is passed, the voiceprint of unregistered voiceprint information can be
  • the fingerprint matching object registers voiceprint information silently, and dynamically updates the registered voiceprint information database stored in the voiceprint lock to improve the security of the voiceprint lock.
  • a voiceprint lock control device which is different from the voiceprint lock control device described in the above embodiments in that the device further includes: Speech recognition module 790 .
  • the voice recognition module 790 is configured to perform voice recognition on the feedback voice, determine the voice recognition result corresponding to the feedback voice; and compare the voice recognition result corresponding to the feedback voice with the voice recognition result stored in the voiceprint lock Perform text matching on the speech recognition results corresponding to the registered voiceprint information of the voiceprint matching object described in the registered voiceprint information database;
  • the voice recognition module 790 is further configured to respond to the text matching result indicating that the voice recognition result corresponding to the feedback voice matches the voice recognition result corresponding to the registered voiceprint information of the voiceprint matching object, and jump to To call the voiceprint matching module 730; otherwise, end the current voiceprint unlocking process;
  • the voice recognition result corresponding to the registered voiceprint information of the voiceprint matching object is user-defined registration text for the voiceprint matching object.
  • the voiceprint registration module 700 is further configured to perform voice recognition on the registered voice, and combine the voice recognition result of the registered voice, the registered voiceprint information corresponding to the registered voice, and the voiceprint matching object for the registered voice These kinds of data are correspondingly stored in the registered voiceprint information database.
  • the voiceprint lock control device disclosed in the embodiment of the present application adds a voice recognition step before the voiceprint authentication, so that the user can customize the voice recognition text corresponding to the voiceprint matching object in the voiceprint registration stage, thereby enhancing the voiceprint lock. Anti-attack ability, improve the security of voiceprint lock.
  • the embodiment of the present application discloses a voiceprint lock, the structure of which is shown in Figure 2 and Figure 3, for example, the voiceprint lock includes: a master control device 210, which communicates with the master The voice acquisition device 220, memory 230, lock actuator 240, and sensory information output device 250 connected to the control device, wherein,
  • the memory 230 is used to store a dictionary database and a registered voiceprint information database, wherein the registered voiceprint information database stores: voiceprint matching objects, and registered voiceprint information corresponding to the corresponding voiceprint matching;
  • the perceptual information output device 250 is configured to output the perceptual information presenting the specified voiceprint matching object according to the control of the main control device, wherein the output presenting the perceptual information presenting the specified voiceprint matching object includes: displaying the specified voiceprint matching object the text of the matching object; and/or, play the reading voice of the specified voiceprint matching object;
  • the voice collection device 220 is configured to collect a user's feedback voice for the perceptual information according to the control of the main control device, and send the collected feedback voice to the main control device;
  • the main control device 210 is further configured to determine the voiceprint information to be recognized in the feedback voice, and compare the voiceprint information to be recognized with the registered voiceprint information in the registered voiceprint information database stored in the voiceprint lock. Perform voiceprint matching on the voiceprint information, determine the matching result obtained by performing the voiceprint matching, and, in response to the matching result, indicate that the voiceprint information to be recognized matches the registered voiceprint information in the registered voiceprint information database If successful, perform an update operation of the registered voiceprint information base based on the voiceprint information to be identified; and/or perform an unlock control operation of the voiceprint lock.
  • this application also discloses an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor.
  • the processor executes the computer program, it implements the The voiceprint lock control method described in the above-mentioned embodiment.
  • the electronic device may be a PC, a mobile terminal, a personal digital assistant, a tablet computer, and the like.
  • the present application also discloses a computer-readable storage medium, on which a computer program is stored.
  • the program is executed by a processor, the steps of the voiceprint lock control method described in any one of the above-mentioned embodiments of the present application are realized. .
  • each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same and similar parts of each embodiment can be referred to each other.
  • the description is relatively simple, and for related parts, please refer to the part of the description of the method embodiment.
  • each embodiment can be implemented by means of software plus a necessary general-purpose hardware platform, and of course can also be implemented by hardware.
  • the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic discs, optical discs, etc., including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) execute the methods described in various embodiments or some parts of the embodiments.
  • any reference signs placed between parentheses shall not be construed as limiting the claim.
  • the word “comprising” does not exclude the presence of elements or steps not listed in a claim.
  • the word “a” or “an” preceding an element does not exclude the presence of a plurality of such elements.
  • the application can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means can be embodied by one and the same item of hardware.
  • the use of the words first, second, and third, etc. does not indicate any order. These words can be interpreted as names.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Acoustics & Sound (AREA)
  • Health & Medical Sciences (AREA)
  • Business, Economics & Management (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Game Theory and Decision Science (AREA)
  • General Physics & Mathematics (AREA)
  • Telephone Function (AREA)
  • Lock And Its Accessories (AREA)

Abstract

La présente invention appartient au domaine de la sécurité et de la protection, et concerne un procédé et un appareil de commande de verrou à empreinte vocale, et un dispositif électronique, qui aident à améliorer la sécurité des verrous à empreinte vocale. Le procédé consiste à : délivrer des informations de perception d'un objet d'appariement d'empreinte vocale stocké dans un verrou à empreinte vocale à un utilisateur (110) ; en réponse à une voix de rétroaction collectée par le verrou à empreinte vocale relativement aux informations de perception, déterminer des informations d'empreinte vocale à reconnaître dans la voix de rétroaction (120) ; effectuer un appariement d'empreinte vocale sur les informations d'empreinte vocale et des informations d'empreinte vocale d'enregistrement figurant dans une bibliothèque d'informations d'empreinte vocale d'enregistrement stockée dans le verrou à empreinte vocale pour obtenir un résultat d'appariement (130) ; en réponse au fait que le résultat d'appariement indique que l'appariement entre les informations d'empreinte vocale et les informations d'empreinte vocale d'enregistrement figurant dans la bibliothèque d'informations d'empreinte vocale d'enregistrement est réussi, exécuter une opération de mise à jour sur la bibliothèque d'informations d'empreinte vocale d'enregistrement sur la base des informations d'empreinte vocale et/ou exécuter une opération de commande de déverrouillage sur le verrou à empreinte vocale (140). Selon le présent procédé, une fois que l'appariement d'empreinte vocale est réussi, une bibliothèque d'informations d'empreinte vocale d'enregistrement est mise à jour sur la base d'informations d'empreinte vocale actuelles à reconnaître appariées avec succès, et des informations d'empreinte vocale d'enregistrement sont enrichies, améliorant ainsi efficacement la sécurité du verrou à empreinte vocale.
PCT/CN2022/100787 2021-07-01 2022-06-23 Procédé et appareil de commande de verrou à empreinte vocale, et dispositif électronique WO2023274028A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110747765.1 2021-07-01
CN202110747765.1A CN113570754B (zh) 2021-07-01 2021-07-01 声纹锁控制方法、装置、电子设备

Publications (1)

Publication Number Publication Date
WO2023274028A1 true WO2023274028A1 (fr) 2023-01-05

Family

ID=78163421

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/100787 WO2023274028A1 (fr) 2021-07-01 2022-06-23 Procédé et appareil de commande de verrou à empreinte vocale, et dispositif électronique

Country Status (2)

Country Link
CN (1) CN113570754B (fr)
WO (1) WO2023274028A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113570754B (zh) * 2021-07-01 2022-04-29 汉王科技股份有限公司 声纹锁控制方法、装置、电子设备

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106506524A (zh) * 2016-11-30 2017-03-15 百度在线网络技术（北京）有限公司 用于验证用户的方法和装置
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
CN107180632A (zh) * 2017-06-19 2017-09-19 微鲸科技有限公司 语音控制方法、装置及可读存储介质
WO2020007495A1 (fr) * 2018-07-06 2020-01-09 Veridas Digital Authentication Solutions, S.L. Authentification d'un utilisateur
CN111341326A (zh) * 2020-02-18 2020-06-26 RealMe重庆移动通信有限公司 语音处理方法及相关产品
CN111599074A (zh) * 2020-06-09 2020-08-28 苏州思必驰信息科技有限公司 楼宇门禁的注册方法、使用方法及装置
CN113570754A (zh) * 2021-07-01 2021-10-29 汉王科技股份有限公司 声纹锁控制方法、装置、电子设备

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102760434A (zh) * 2012-07-09 2012-10-31 华为终端有限公司 一种声纹特征模型更新方法及终端
CN106098068B (zh) * 2016-06-12 2019-07-16 腾讯科技（深圳）有限公司 一种声纹识别方法和装置
CN106782572B (zh) * 2017-01-22 2020-04-07 清华大学 语音密码的认证方法及系统
CN107610709B (zh) * 2017-08-01 2021-03-19 百度在线网络技术（北京）有限公司 一种训练声纹识别模型的方法及系统
CN112863009B (zh) * 2019-11-08 2022-08-05 北京声智科技有限公司 门锁控制方法、装置、设备及计算机存储介质
CN111091837A (zh) * 2019-12-27 2020-05-01 中国人民解放军陆军工程大学 一种基于在线学习的时变声纹认证方法及系统
CN112309406A (zh) * 2020-09-21 2021-02-02 北京沃东天骏信息技术有限公司 声纹注册方法、装置和计算机可读存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106506524A (zh) * 2016-11-30 2017-03-15 百度在线网络技术（北京）有限公司 用于验证用户的方法和装置
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
CN107180632A (zh) * 2017-06-19 2017-09-19 微鲸科技有限公司 语音控制方法、装置及可读存储介质
WO2020007495A1 (fr) * 2018-07-06 2020-01-09 Veridas Digital Authentication Solutions, S.L. Authentification d'un utilisateur
CN111341326A (zh) * 2020-02-18 2020-06-26 RealMe重庆移动通信有限公司 语音处理方法及相关产品
CN111599074A (zh) * 2020-06-09 2020-08-28 苏州思必驰信息科技有限公司 楼宇门禁的注册方法、使用方法及装置
CN113570754A (zh) * 2021-07-01 2021-10-29 汉王科技股份有限公司 声纹锁控制方法、装置、电子设备

Also Published As

Publication number Publication date
CN113570754B (zh) 2022-04-29
CN113570754A (zh) 2021-10-29

Similar Documents

Publication Publication Date Title
US7447632B2 (en) Voice authentication system
JP6394709B2 (ja) 話者識別装置および話者識別用の登録音声の特徴量登録方法
JP6567040B2 (ja) 人工知能に基づく声紋ログイン方法と装置
KR101757990B1 (ko) 성문 인증 방법 및 장치
WO2017113658A1 (fr) Procédé et dispositif à base d'intelligence artificielle permettant une authentification par empreinte vocale
WO2008100971A1 (fr) Vérification du locuteur en fonction d'un texte
JPH0345417B2 (fr)
JP2004101901A (ja) 音声対話装置及び音声対話プログラム
CN111091837A (zh) 一种基于在线学习的时变声纹认证方法及系统
WO2023274028A1 (fr) Procédé et appareil de commande de verrou à empreinte vocale, et dispositif électronique
JP4318475B2 (ja) 話者認証装置及び話者認証プログラム
JP2001331801A (ja) 個人認証装置および個人認証方法および記録媒体
JP7339116B2 (ja) 音声認証装置、音声認証システム、および音声認証方法
WO2018088534A1 (fr) Dispositif électronique, procédé de commande pour dispositif électronique et programme de commande pour dispositif électronique
JP2003302999A (ja) 音声による個人認証システム
WO2021139182A1 (fr) Procédé et appareil de détection de voix intelligente efficace, dispositif et support de stockage lisible par ordinateur
JP4245948B2 (ja) 音声認証装置、音声認証方法及び音声認証プログラム
JP2001265385A (ja) 話者認識装置
JP2000099090A (ja) 記号列を用いた話者認識方法
JPH11344992A (ja) 音声辞書作成方法、個人認証装置および記録媒体
JP3098157B2 (ja) 話者照合方法及び装置
WO2000058947A1 (fr) Authentification utilisateur destinee a un materiel electronique
JP2004279770A (ja) 話者認証装置及び判別関数設定方法
CN112307181A (zh) 基于特定语料库的提取语料的方法及语料提取器
JPWO2006027844A1 (ja) 話者照合装置

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: 22831844

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE