WO2018233137A1 - 电子装置、进线身份识别方法和计算机可读存储介质 - Google Patents

电子装置、进线身份识别方法和计算机可读存储介质 Download PDF

Info

Publication number
WO2018233137A1
WO2018233137A1 PCT/CN2017/104933 CN2017104933W WO2018233137A1 WO 2018233137 A1 WO2018233137 A1 WO 2018233137A1 CN 2017104933 W CN2017104933 W CN 2017104933W WO 2018233137 A1 WO2018233137 A1 WO 2018233137A1
Authority
WO
WIPO (PCT)
Prior art keywords
customer
voiceprint feature
voice data
incoming
voiceprint
Prior art date
Application number
PCT/CN2017/104933
Other languages
English (en)
French (fr)
Inventor
晏湘涛
Original Assignee
平安科技（深圳）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 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2018233137A1 publication Critical patent/WO2018233137A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/22—Arrangements for supervision, monitoring or testing
    • H04M3/229—Wire identification arrangements; Number assignment determination
    • 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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/22—Arrangements for supervision, monitoring or testing
    • H04M3/2281—Call monitoring, e.g. for law enforcement purposes; Call tracing; Detection or prevention of malicious calls
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M2203/00—Aspects of automatic or semi-automatic exchanges
    • H04M2203/60—Aspects of automatic or semi-automatic exchanges related to security aspects in telephonic communication systems
    • H04M2203/6045—Identity confirmation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M2203/00—Aspects of automatic or semi-automatic exchanges
    • H04M2203/60—Aspects of automatic or semi-automatic exchanges related to security aspects in telephonic communication systems
    • H04M2203/6054—Biometric subscriber identification

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to an electronic device, an incoming line identification method, and a computer readable storage medium.
  • the company providing the service usually uses the incoming number for identification.
  • the telephone answering service personnel need to use the voice mode on the telephone. Carry out relevant information verification with the customer to confirm the identity of the customer, and provide relevant services to the customer after the identity is confirmed; this method is cumbersome and inconvenient, and the experience for the customer is not good; 2.
  • Careful storage of all customer recordings is required in case of disputes. The cost of saving a large number of recordings is high; 3. When the phone is stolen, it cannot be effectively identified.
  • the main object of the present invention is to provide an incoming line identification method, which aims to make the identification of incoming customers more convenient, accurate and efficient, and to reduce recording storage and reduce costs.
  • a first aspect of the present invention provides an electronic device including a memory and a processor, the memory having an incoming line identification system operable on the processor, the incoming line identification system being The following steps are implemented during execution:
  • the extracted voiceprint feature is successfully matched with the determined customer voiceprint feature data, it is confirmed that the current incoming customer is the customer corresponding to the incoming number.
  • a second aspect of the present invention provides a method for identifying an incoming line, the method comprising:
  • the extracted voiceprint feature is successfully matched with the determined customer voiceprint feature data, it is confirmed that the current incoming customer is the customer corresponding to the incoming number.
  • a third aspect of the present invention provides a computer readable storage medium storing an incoming line identification system, the incoming line identification system being executable by at least one processor to cause the at least one The processor performs the following operations:
  • the extracted voiceprint feature is successfully matched with the determined customer voiceprint feature data, it is confirmed that the current incoming customer is the customer corresponding to the incoming number.
  • the technical solution of the present invention first obtains the corresponding customer identity information by the incoming line number, and determines the corresponding pre-stored voiceprint feature data from the voiceprint feature database according to the customer identity information, and extracts the voice data from the current incoming client voice data.
  • the voiceprint feature matches the determined voiceprint feature data, and according to the matching result, it is verified whether the current incoming customer is the customer of the incoming number, and the matching success indicates that the current incoming customer is the customer of the incoming number, and vice versa. No.
  • the technical scheme of the present invention adopts the method of automatically verifying the identity of the customer by using voiceprint, and does not need to answer the service personnel to verify the identity of the customer through the telephone, which is more convenient, accurate and efficient, and can also prevent others from stealing the customer's telephone number.
  • the voiceprint verification method is accurate, safe and reliable, and can eliminate the need to record the call, reduce the storage of the recorded data, and reduce the storage cost.
  • FIG. 1 is a flow chart of an embodiment of a method for identifying an incoming line of the present invention
  • FIG. 2 is a flowchart of an embodiment of a method for identifying an incoming line identity according to the present invention
  • FIG. 3 is a flowchart of a third embodiment of an incoming line identification method according to the present invention.
  • FIG. 4 is a schematic diagram of an operating environment of a preferred embodiment of the incoming identification system of the present invention.
  • FIG. 5 is a block diagram showing a program module of an embodiment of the incoming line identification system of the present invention.
  • FIG. 6 is a program block diagram of a second embodiment of the incoming line identification system of the present invention.
  • FIG. 7 is a schematic structural diagram of an embodiment of an acquisition and extraction module in an incoming line identification system of the present invention.
  • FIG. 8 is a schematic structural diagram of another embodiment of an acquisition and extraction module in the incoming line identification system of the present invention.
  • the invention provides an incoming line identification method.
  • FIG. 1 is a flowchart of an embodiment of an incoming line identification method according to the present invention.
  • Step S10 After receiving the customer call, obtain the customer identity information corresponding to the current incoming line number according to the mapping relationship between the predetermined phone number and the customer identity information;
  • the database of the system has a mapping table of phone numbers and customer identity information (eg, may include a username, a certificate number, etc.) including all phone numbers already stored in the system and customer identities corresponding to those phone numbers Information; after receiving the incoming call from the customer, according to the number of the incoming customer, from the mapping table of the telephone number and the customer identity information, to find the customer identity information corresponding to the current incoming number;
  • customer identity information eg, may include a username, a certificate number, etc.
  • Step S20 after successfully obtaining the customer identity information, determining the customer voiceprint feature data corresponding to the acquired customer identity information according to the mapping relationship between the predetermined customer identity information and the voiceprint feature data;
  • a voiceprint feature database is established for the stock customer.
  • the voiceprint feature database pre-records the voiceprint feature data of each stock customer, and the mapping relationship between the voiceprint feature data and the customer identity information is preset in the system. After successfully obtaining the customer identity information of the current incoming number, the corresponding customer voiceprint feature data is found according to the obtained customer identity information.
  • Step S30 collecting voice data of the incoming customer, and extracting voiceprint features in the collected voice data
  • the voice data of the incoming call customer during the call is collected, and the collected voice data is analyzed to extract the voiceprint feature of the current incoming customer.
  • Step S40 matching the extracted voiceprint feature with the determined customer voiceprint feature data
  • step S50 if the extracted voiceprint feature is successfully matched with the determined customer voiceprint feature data, it is confirmed that the current incoming customer is the customer corresponding to the incoming line number.
  • the voiceprint feature extracted according to the collected voice data is successfully matched with the voiceprint feature data determined according to the customer identity information, it can be determined that the current incoming customer is the customer of the incoming number, and can be provided according to customer requirements. Related services; if the match fails, the current incoming customer is not the customer of the incoming number, and may be stolen or otherwise; in the case of a failed match, the system will not provide the relevant service to the incoming customer.
  • the solution of the embodiment first obtains the corresponding customer identity information by using the incoming line number, and determines the corresponding pre-stored voiceprint feature data from the voiceprint feature database according to the customer identity information, and extracts the voice data of the current incoming client.
  • the voiceprint feature matches the determined voiceprint feature data, and according to the matching result, it is verified whether the current incoming customer is the customer of the incoming number, and the matching success indicates that the current incoming customer is the customer of the incoming number, and vice versa. No.
  • the technical solution of the embodiment adopts the method of automatically verifying the identity of the customer by voiceprint, and does not need to answer the service personnel to verify the identity of the customer through the telephone, which is more convenient, accurate and efficient, and can also prevent others from stealing the customer.
  • the phone number is dialed; in addition, the voiceprint verification method is accurate, safe and reliable, and can save the recording data and reduce the storage cost without recording the call.
  • FIG. 2 is a flowchart of a second embodiment of the incoming identity identification method of the present invention.
  • the method for identifying the incoming line further includes:
  • Step S60 after failing to obtain the customer identity information, collecting the voice data of the incoming client, and extracting the voiceprint feature in the collected voice data;
  • the customer identity information corresponding to the current incoming number may not be found according to the mapping relationship between the telephone number and the customer identity information, that is, the failure to obtain the customer identity information fails.
  • the system collects the voice data of the incoming customer, and The voiceprint feature is extracted and extracted from the voice data of the incoming customer, so as to perform subsequent verification operations according to the voiceprint feature.
  • Step S70 matching the extracted voiceprint features with the customer voiceprint feature data in the voiceprint database one by one;
  • the voiceprint features extracted from the collected incoming voice data of the incoming line are matched with all the customer voiceprint feature data in the voiceprint feature database one by one to confirm whether the voiceprint feature data exists in the voiceprint feature database.
  • the voice characteristics of the current incoming customers are matched to determine whether the current incoming customer is the company's stock customer.
  • Step S80 if the extracted voiceprint feature is successfully matched with the customer voiceprint feature data in the voiceprint database, the voiceprint feature matching is confirmed according to the mapping relationship between the predetermined customer identity information and the customer voiceprint feature data. Customer identity information corresponding to the customer's voiceprint feature data.
  • the voiceprint feature database When there is a customer voiceprint feature data in the voiceprint feature database that matches the voiceprint feature of the current incoming customer, it indicates that the current incoming customer is the company's stock customer entering the line through the phone number that the company has not registered.
  • the system directly finds the customer identity information corresponding to the matched customer voiceprint feature data according to the mapping relationship between the customer identity information and the customer voiceprint feature data, that is, the identity information of the current incoming customer.
  • the manual answering service can be processed, or the operation can be performed according to other preset methods.
  • the customer can automatically identify the identity of the customer through the voiceprint when the customer does not use the phone number saved by the company providing the service, thereby avoiding the customer having to use the company-saved number to call.
  • the limitations of the situation in which the identity can be identified enable the customer to obtain a better service effect.
  • FIG. 3 is a flowchart of a third embodiment of the incoming identity identification method of the present invention.
  • the step of collecting the voice data of the incoming client and extracting the voiceprint feature in the collected voice data in the incoming line identification method includes:
  • Step S101 randomly generate a character string of a preset number of characters, broadcast the character string, and prompt the user to repeat the prompt of the character string;
  • the system generates a string of a preset number of characters in a random manner (for example, randomly generates an 8-bit character, a letter, a number, or a string formed by any of the above), and broadcasts the string and the broadcast prompts the user to repeat the character string. Prompt (for example, "Please repeat the following statement:********").
  • Step S102 the receiving client repeats the voice data corresponding to the character string, extracts the content in the received voice data, and compares the extracted content with the content of the string broadcast by the system;
  • the system receives the voice data when the customer repeatedly reproduces the randomly generated character string, performs content analysis on the voice data to extract the content in the voice data, and compares the extracted content with the content of the string broadcasted by the system. To confirm if the content is the same.
  • Step S103 after the comparison is successful, extract the voiceprint feature in the received voice data.
  • the comparison is successful, that is, the content extracted by the system according to the voice data repeated by the customer is the same as the content of the string broadcasted by the system, then the voiceprint feature in the voice data of the customer's retelling is extracted to perform the next voiceprint verification. .
  • the comparison is unsuccessful, that is, the content extracted by the system according to the voice data repeated by the customer and the content of the string broadcast by the system are not In the same way, this situation may be caused by the customer's spoofing, the customer's local accent, etc. At this time, the customer may be prompted to repeat the content, and jump to the execution step S101, and provide the character string again for the customer to repeat the comparison.
  • the system may prompt the client to verify that the number of errors is too large, transfer the customer to the telephone answering service personnel for manual service, or may take the interrupt call processing. .
  • the solution of the embodiment is generated by the system in a random dynamic manner for the customer to repeat, and the voice data of the customer retelling is used as the data for the voiceprint feature extraction, and the comparison of the client retelling content is performed before the voiceprint feature is extracted to prevent others from being compared. Using the recorded data of the customer's voice to verify the security of the voiceprint verification through the system's voiceprint verification.
  • the step of collecting the voice data of the incoming customer and extracting the voiceprint feature in the collected voice data in the incoming line identification method further includes:
  • the error count is incremented by one
  • the extracted voiceprint features may be affected by the matching of the determined customer voiceprint features, so the system will set multiple voiceprint verification opportunities.
  • an error count is performed, and according to the value of the error count value, it is determined whether the random generated character string is provided again for the customer to repeat for voiceprint verification;
  • step S101 is performed
  • step S101 If the error count is less than or equal to the preset error count threshold, that is, it is determined that the number of voiceprint matching errors has not exceeded the error permission number, then the process jumps to step S101, and the character string is again provided for the customer to repeat the verification;
  • the emergency security process may be: 1. Sending an abnormal reminder message to the contact mailbox or emergency contact number reserved for the current incoming number; 2. Interrupting the current call, locking the current incoming line number to an abnormal state, and making the current incoming line The number cannot be related to business operations, and so on.
  • the solution of the embodiment provides the customer with multiple opportunities for voiceprint verification, which is more in line with the actual use requirement, and after the number of voiceprint verification errors exceeds a certain number of times, emergency security processing is performed to protect the customer's rights and interests.
  • the matching result of the voiceprint feature may be determined by comparing the matching degree of the voiceprint feature with a preset matching threshold, and confirming the matching when the matching threshold is greater than the preset matching threshold. Success, if it is less than or equal to the preset matching threshold, it is confirmed as a match failure.
  • the invention also proposes an incoming line identification system.
  • FIG. 4 is a schematic diagram of an operating environment of a preferred embodiment of the incoming identification system 10 of the present invention.
  • the incoming line identification system 10 is installed and operates in the electronic device 1.
  • the electronic device 1 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a server.
  • the electronic device 1 may include, but is not limited to, a memory 11, a processor 12, and a display 13.
  • Figure 3 shows only the electronic device 1 with components 11-13, but it should be understood that not all illustrated components may be implemented, and more or fewer components may be implemented instead.
  • the memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, such as a hard disk or memory of the electronic device 1.
  • the memory 11 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in hard disk equipped on the electronic device 1, a smart memory card (SMC), and a secure digital (SD). Card, flash card, etc.
  • the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device.
  • the memory 11 is used to store application software and various types of data, such as program codes of the incoming line identification system 10, installed in the electronic device 1.
  • the memory 11 can also be used to temporarily store data that has been output or is about to be output.
  • the processor 12 in some embodiments, may be a Central Processing Unit (CPU), microprocessor or other data processing chip for running program code or processing data stored in the memory 11, such as performing incoming identity Identify system 10 and the like.
  • CPU Central Processing Unit
  • microprocessor or other data processing chip for running program code or processing data stored in the memory 11, such as performing incoming identity Identify system 10 and the like.
  • the display 13 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch sensor, or the like in some embodiments.
  • the display 13 is for displaying information processed in the electronic device 1 and a user interface for displaying visualization, such as a business customization interface or the like.
  • the components 11-13 of the electronic device 1 communicate with one another via a system bus.
  • FIG. 5 is a program block diagram of an embodiment of the incoming line identification system 10 of the present invention.
  • the incoming line identification system 10 can be divided into one or more modules, one or more modules being stored in the memory 11 and being processed by one or more processors (this embodiment is the processor 12) ) is performed to complete the present invention.
  • the incoming line identification system 10 can be divided into an acquisition module 101, a determination module 102, an acquisition extraction module 103, a first matching module 104, and a first identity validation module 105.
  • module refers to a series of computer program instruction segments capable of performing a specific function, which is more suitable than the program for describing the execution process of the incoming line identification system 10 in the electronic device 1, wherein:
  • the obtaining module 101 is configured to obtain, according to a mapping relationship between the predetermined phone number and the customer identity information, the customer identity information corresponding to the current incoming line number after receiving the customer call;
  • the database of the system has a mapping table of phone numbers and customer identity information (eg, may include a username, a certificate number, etc.) including all phone numbers already stored in the system and customer identities corresponding to those phone numbers
  • the obtaining module 101 searches from the mapping table of the phone number and the customer identity information according to the number of the incoming customer to obtain the customer identity information corresponding to the current incoming number;
  • the determining module 102 is configured to determine the customer voiceprint feature data corresponding to the acquired customer identity information according to the mapping relationship between the predetermined customer identity information and the voiceprint feature data after successfully acquiring the client identity information;
  • a voiceprint feature database is established for the stock customer.
  • the voiceprint feature database pre-records the voiceprint feature data of each stock customer, and the mapping relationship between the voiceprint feature data and the customer identity information is preset in the system.
  • the determining module 102 finds the corresponding customer voiceprint feature data according to the acquired customer identity information.
  • the collection and extraction module 103 is configured to collect voice data of the incoming customer, and extract voiceprint features in the collected voice data;
  • the voice data of the incoming call customer during the call is collected, and the collected voice data is analyzed to extract the voiceprint feature of the current incoming customer.
  • a first matching module 104 configured to match the extracted voiceprint feature with the determined customer voiceprint feature data
  • the first matching module 104 matches the voiceprint feature extracted by the acquisition extraction module 103 with the customer voiceprint feature data determined according to the determining module 102 to confirm whether the two belong to the same person's voiceprint;
  • the first identity confirmation module 105 is configured to confirm that the current incoming customer is the customer corresponding to the incoming number after the extracted voiceprint feature is successfully matched with the determined customer voiceprint feature data.
  • the first identity verification module 105 determines that the current incoming customer is the customer of the incoming number, and may The customer needs to provide related services; if the matching fails, the first identity confirmation module 105 confirms that the current incoming customer is not the customer of the incoming number, and may be stolen or otherwise; in the case of a matching failure, the system does not Provide related services to incoming customers.
  • the solution of the embodiment first obtains the corresponding customer identity information by using the incoming line number, and determines the corresponding pre-stored voiceprint feature data from the voiceprint feature database according to the customer identity information, and extracts the voice data of the current incoming client.
  • the voiceprint feature matches the determined voiceprint feature data, and according to the matching result, it is verified whether the current incoming customer is the incoming number
  • the customer of the code, the matching success indicates that the current incoming customer is the customer of the incoming number, and vice versa.
  • the technical solution of the embodiment adopts the method of automatically verifying the identity of the customer by voiceprint, and does not need to answer the service personnel to verify the identity of the customer through the telephone, which is more convenient, accurate and efficient, and can also prevent others from stealing the customer.
  • the phone number is dialed; in addition, the voiceprint verification method is accurate, safe and reliable, and can save the recording data and reduce the storage cost without recording the call.
  • FIG. 6 is a program module diagram of a second embodiment of the incoming line identification system 10 of the present invention.
  • the incoming identity identification system further includes a second matching module 106 and a second identity confirmation module 107, where:
  • the collection and extraction module 103 is further configured to collect voice data of the incoming customer after the failure to acquire the customer identity information, and extract the voiceprint feature in the collected voice data; where the current incoming number is a telephone number in which no record is registered in the system.
  • the customer identity information corresponding to the current incoming number is not found, that is, the failure to obtain the customer identity information fails, and in this case, it is not determined that the current incoming customer is not a company.
  • the stock customer is likely to be entered by the company's stock customer through the unregistered phone; therefore, the acquisition and extraction module 103 collects the incoming voice data of the incoming customer and the voice data of the incoming customer is collected.
  • the voiceprint features are extracted and analyzed to perform subsequent verification operations based on the voiceprint features.
  • the second matching module 106 is configured to match the extracted voiceprint features with the customer voiceprint feature data in the voiceprint database one by one;
  • the second matching module 106 matches each of the customer voiceprint feature data in the voiceprint feature database one by one by using the voiceprint feature extracted from the collected incoming client voice data to confirm whether there is a client in the voiceprint feature database.
  • the voiceprint feature data is matched with the voiceprint characteristics of the current incoming customer to determine whether the current incoming customer is a stock customer of the company.
  • the second identity confirmation module 107 is configured to confirm the extraction according to the mapping relationship between the predetermined customer identity information and the customer voiceprint feature data after the extracted voiceprint feature is successfully matched with the customer voiceprint feature data in the voiceprint database.
  • the customer's identity information corresponding to the customer's voiceprint feature data is matched by the voiceprint feature.
  • the second identity confirmation module 107 When there is a customer voiceprint feature data in the voiceprint feature database that matches the voiceprint feature of the current incoming customer, it indicates that the current incoming customer is the company's stock customer entering the line through the phone number that the company has not registered.
  • the second identity confirmation module 107 directly finds the customer identity information corresponding to the matched customer voiceprint feature data according to the mapping relationship between the client identity information and the customer voiceprint feature data, that is, the identity information of the current incoming client. Of course, if all the matching fails, the incoming customer is not the existing customer. In this case, the manual answering service can be processed, or the operation can be performed according to other preset methods.
  • the customer can automatically identify the identity of the customer through the voiceprint when the customer does not use the phone number saved by the company providing the service, thereby avoiding the customer having to use the company-saved number to call.
  • the limitations of the situation in which the identity can be identified enable the customer to obtain a better service effect.
  • the collection and extraction module 103 of the incoming line identification system 10 includes:
  • the generating sub-module 1031 is configured to randomly generate a character string of a preset number of characters, broadcast the character string, and prompt the user to repeat the prompt of the character string;
  • the system generates a string of a preset number of characters in a random manner (for example, randomly generates an 8-bit character, a letter, a number, or a string formed by any of the above), and broadcasts the string and the broadcast prompts the user to repeat the character string. Prompt (for example, "Please repeat the following statement:********").
  • the comparison sub-module 1032 is configured to receive, by the client, the voice data corresponding to the character string, extract the content in the received voice data, and compare the extracted content with the content of the string broadcast by the system;
  • the comparison sub-module 1032 receives the voice data when the character string randomly generated by the client paraphrase system, performs content analysis on the voice data to extract the content in the voice data, and extracts the extracted content and the content of the string broadcasted by the system. Make a comparison to confirm if the content is the same.
  • the extraction sub-module 1033 extracts the voiceprint features in the received voice data after the comparison is successful.
  • the extraction sub-module 1033 extracts the voiceprint features in the voice data of the customer-repeated voice for the next voiceprint verification. If the comparison sub-module 1032 confirms that the comparison is unsuccessful, since the situation may be caused by the customer's account error, the customer's local accent, etc., at this time, the customer may be prompted to repeat the content and jump to the execution generation sub-module. 1031, again provide a string for the customer to repeat the comparison verification; in addition, if the number of unsuccessful comparisons exceeds the preset error threshold, the system may prompt the client to verify the number of errors too much, and transfer the customer to the telephone answering service personnel Perform manual services, or you can take interrupted calls and other processing methods.
  • the solution of the embodiment is generated by the system in a random dynamic manner for the customer to repeat, and the voice data of the customer retelling is used as the data for the voiceprint feature extraction, and the comparison of the client retelling content is performed before the voiceprint feature is extracted to prevent others from being compared. Using the recorded data of the customer's voice to verify the security of the voiceprint verification through the system's voiceprint verification.
  • the collection and extraction module 103 of the incoming line identification system 10 further includes:
  • Counting sub-module 1034 for extracting voiceprint features and determined customer voiceprints After the feature data matching fails, the error count is incremented by one;
  • the extracted voiceprint features may be affected by the matching of the determined customer voiceprint features, so the system will set multiple voiceprint verification opportunities.
  • an error count is performed, and according to the value of the error count value, it is determined whether the random generated character string is provided again for the customer to repeat for voiceprint verification;
  • the jump sub-module 1035 is configured to randomly generate a character string of a preset number of characters when the error count is less than a preset error count threshold, broadcast the character string, and prompt the user to repeat the prompt of the character string;
  • the jump sub-module 1035 jumps to the processing sub-module 1031 to execute again. Provide a string for the customer to repeat to verify again;
  • the alarm sub-module 1036 is configured to perform emergency security processing after the error count exceeds a preset error count threshold.
  • the alert sub-module 1036 When the error count exceeds a preset error count threshold, the alert sub-module 1036 performs emergency security processing.
  • the emergency security process may be: 1. Sending an abnormal reminder message to the contact mailbox or emergency contact number reserved for the current incoming number; 2. Interrupting the current call, locking the current incoming line number to an abnormal state, and making the current incoming line The number cannot be related to business operations, and so on.
  • the solution of the embodiment provides the customer with multiple opportunities for voiceprint verification, which is more in line with the actual use requirement, and after the number of voiceprint verification errors exceeds a certain number of times, emergency security processing is performed to protect the customer's rights and interests.
  • the matching result of the voiceprint feature may be determined by comparing the matching degree of the voiceprint feature with a preset matching threshold, and confirming that the matching is successful when the matching threshold is greater than the preset matching threshold, and less than or equal to the preset matching threshold. Confirmed that the match failed.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Technology Law (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Telephonic Communication Services (AREA)

Abstract

一种电子装置、进线身份识别方法和计算机可读存储介质，其中，该方法包括：在接收到客户来电后，根据预先确定的电话号码与客户身份信息的映射关系，获取当前进线号码对应的客户身份信息（S10）；在成功获取客户身份信息后，根据预先确定的客户身份信息与声纹特征数据的映射关系，确定获取到的客户身份信息所对应的客户声纹特征数据（S20）；采集进线客户的语音数据，提取采集到的语音数据中的声纹特征（S30）；将提取出的声纹特征与确定的客户声纹特征数据进行匹配（S40）；若提取出的声纹特征与确定的客户声纹特征数据匹配成功，则确认当前进线客户为该进线号码对应的客户本人（S50）。所述方法使得对进线客户的身份验证更加方便、准确及高效。

Description

电子装置、进线身份识别方法和计算机可读存储介质
本申请基于巴黎公约申明享有2017年6月24日递交的申请号为CN 201710489752.2、名称为“电子装置、进线身份识别方法和计算机可读存储介质”中国专利申请的优先权，该中国专利申请的整体内容以参考的方式结合在本申请中。
技术领域
本发明涉及通信技术领域，特别涉及一种电子装置、进线身份识别方法和计算机可读存储介质。
背景技术
目前，客户电话进线后，提供服务的公司通常都是通过呼入号码进行身份识别，这在需要严格验证客户身份的场景下存在以下问题：1、需要电话接听服务人员在电话中通过语音方式与客户进行相关信息验证以确认客户身份，在身份确认后方能为客户提供相关服务；这种方式过程繁琐不便，给客户的体验不好；2、需要谨慎保存所有客户录音以便出现纠纷时备用，大量录音的保存成本高；3、当电话被盗用时就无法被有效识别客户身份。
因此，急需提出一种解决上述问题的方案。
发明内容
本发明的主要目的是提供一种进线身份识别方法，旨在使对进线客户的身份验证更加的方便、准确、高效，以及减少录音保存、降低成本。
本发明第一方面提供一种电子装置，包括存储器和处理器，所述存储器上存储有可在所述处理器上运行的进线身份识别系统，所述进线身份识别系统被所述处理器执行时实现如下步骤：
在接收到客户来电后，根据预先确定的电话号码与客户身份信息的映射关系，获取当前进线号码对应的客户身份信息；
在成功获取客户身份信息后，根据预先确定的客户身份信息与声纹特征数据的映射关系，确定获取到的客户身份信息所对应的客户声纹特征数据；
采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
将提取出的声纹特征与确定的客户声纹特征数据进行匹配；
若提取出的声纹特征与确定的客户声纹特征数据匹配成功，则确认当前进线客户为该进线号码对应的客户本人。
本发明第二方面提供一种进线身份识别方法，所述方法包括：
在接收到客户来电后，根据预先确定的电话号码与客户身份信息的映射关系，获取当前进线号码对应的客户身份信息；
在成功获取客户身份信息后，根据预先确定的客户身份信息与声纹特征数据的映射关系，确定获取到的客户身份信息所对应的客户声纹特征数据；
采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
将提取出的声纹特征与确定的客户声纹特征数据进行匹配；
若提取出的声纹特征与确定的客户声纹特征数据匹配成功，则确认当前进线客户为该进线号码对应的客户本人。
本发明第三方面提供一种计算机可读存储介质，所述计算机可读存储介质存储有进线身份识别系统，所述进线身份识别系统可被至少一个处理器执行，以使所述至少一个处理器执行如下操作：
在接收到客户来电后，根据预先确定的电话号码与客户身份信息的映射关系，获取当前进线号码对应的客户身份信息；
在成功获取客户身份信息后，根据预先确定的客户身份信息与声纹特征数据的映射关系，确定获取到的客户身份信息所对应的客户声纹特征数据；
采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
将提取出的声纹特征与确定的客户声纹特征数据进行匹配；
若提取出的声纹特征与确定的客户声纹特征数据匹配成功，则确认当前进线客户为该进线号码对应的客户本人。
本发明技术方案先通过进线号码获取对应的客户身份信息，根据客户身份信息从声纹特征数据库中确定其对应的预存的声纹特征数据，将从当前进线客户的语音数据中提取出的声纹特征与该确定的声纹特征数据匹配，根据匹配结果来验证当前进线客户是否是该进线号码的客户本人，匹配成功说明当前进线客户是该进线号码的客户本人，反之则不是。与现有技术相比，本发明技术方案采用声纹自动验证客户身份的方式，不需要通过电话接听服务人员对客户身份进行验证，更加的方便、准确及高效，也可以避免他人盗用客户电话号码进行拨打的情况；另外，通过声纹验证的方式，准确且安全可靠，可以无需对通话进行录音，减少录音数据的保存，降低存储成本。
附图说明
图1为本发明进线身份识别方法一实施例的流程图；
图2为本发明进线身份识别方法二实施例的流程图；
图3为本发明进线身份识别方法三实施例的流程图；
图4为本发明进线身份识别系统较佳实施例的运行环境示意图；
图5为本发明进线身份识别系统一实施例的程序模块图；
图6为本发明进线身份识别系统二实施例的程序模块图；
图7为本发明进线身份识别系统中采集提取模块的一实施方案的结构示意图；
图8为本发明进线身份识别系统中采集提取模块的另一实施方案的结构示意图。
具体实施方式
以下结合附图对本发明的原理和特征进行描述，所举实例只用于解释本发明，并非用于限定本发明的范围。
本发明提出一种进线身份识别方法。
如图1所示，图1为本发明进线身份识别方法一实施例的流程图。
本实施例的进线身份识别方法包括：
步骤S10，在接收到客户来电后，根据预先确定的电话号码与客户身份信息的映射关系，获取当前进线号码对应的客户身份信息；
系统的数据库中具有电话号码与客户身份信息(例如，可包括用户名、证件号等)的映射表，该映射表中包括已经在系统中保存的所有电话号码及与这些电话号码对应的客户身份信息；当接到客户进线来电后，根据进线客户的号码从该电话号码与客户身份信息的映射表中，去查找以获取当前进线号码对应的客户身份信息；
步骤S20，在成功获取客户身份信息后，根据预先确定的客户身份信息与声纹特征数据的映射关系，确定获取到的客户身份信息所对应的客户声纹特征数据；
系统中为存量客户建立了声纹特征数据库，该声纹特征数据库中预先录入保存了每个存量客户的声纹特征数据，且系统中预先设置有声纹特征数据与客户身份信息的映射关系表，在成功获取到当前进线号码的客户身份信息后，根据获取到的客户身份信息找出对应的客户声纹特征数据。
步骤S30，采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
采集进线客户在通话中的语音数据，根据采集的语音数据进行分析以提取出当前进线客户的声纹特征。
步骤S40，将提取出的声纹特征与确定的客户声纹特征数据进行匹配；
将根据采集的语音数据提取出的声纹特征与根据客户身份信息确定的客户声纹特征数据进行匹配，确认两者是否属于同一人的声 纹；
步骤S50，若提取出的声纹特征与确定的客户声纹特征数据匹配成功，则确认当前进线客户为该进线号码对应的客户本人。
当根据采集的语音数据提取出的声纹特征与根据客户身份信息确定的声纹特征数据匹配成功，那么就可以确定当前进线客户为该进线号码的客户本人，此时可根据客户需求提供相关服务；如果匹配失败，则说明当前进线客户并非该进线号码的客户本人，有可能是被盗用或其它情形；在匹配失败的情况下，系统不会提供相关服务给进线客户。
本实施例的方案先通过进线号码获取对应的客户身份信息，根据客户身份信息从声纹特征数据库中确定其对应的预存的声纹特征数据，将从当前进线客户的语音数据中提取出的声纹特征与该确定的声纹特征数据匹配，根据匹配结果来验证当前进线客户是否是该进线号码的客户本人，匹配成功说明当前进线客户是该进线号码的客户本人，反之则不是。与现有技术相比，本实施例的技术方案采用声纹自动验证客户身份的方式，不需要通过电话接听服务人员对客户身份进行验证，更加的方便、准确及高效，也可以避免他人盗用客户电话号码进行拨打的情况；另外，通过声纹验证的方式，准确且安全可靠，可以无需对通话进行录音，减少录音数据的保存，降低存储成本。
如图2所示，图2为本发明进线身份识别方法二实施例的流程图。
本实施例方案基于一实施例，在本实施例中，所述进线身份识别方法还包括：
步骤S60，在获取客户身份信息失败后，采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
在当前进线号码是系统中没有登记记录的电话号码的情况时，根据电话号码与客户身份信息的映射关系表就找不到当前进线号码对应的客户身份信息，即获取客户身份信息失败，此种情况下，也并不能确定当前进线客户不是公司的存量客户，很可能是公司存量客户通过没有登记过的电话打进来的；因此，系统对进线客户的语音数据进线采集，并从采集的进线客户的语音数据中分析提取出声纹特征，以根据声纹特征进行后续的验证操作。
步骤S70，将提取出的声纹特征与声纹数据库中的客户声纹特征数据逐一进行匹配；
通过将从采集的进线客户的语音数据中提取出的声纹特征与声纹特征数据库中的所有客户声纹特征数据逐一进行匹配，以确认声纹特征数据库中是否存在客户声纹特征数据与当前进线客户的声纹特征匹配，从而确定当前进线客户是否为公司的存量客户。
步骤S80，若提取出的声纹特征与声纹数据库中的客户声纹特征数据匹配成功，则根据预先确定的客户身份信息与客户声纹特征数据的映射关系，确认提取出的声纹特征匹配到的客户声纹特征数据所对应的客户身份信息。
当声纹特征数据库中存在有客户声纹特征数据与当前进线客户的声纹特征匹配时，说明当前进线的客户是公司的存量客户通过公司没有登记过的电话号码进线的，此时系统根据客户身份信息与客户声纹特征数据的映射关系，直接找出被匹配到的客户声纹特征数据所对应的客户身份信息，即为当前进线客户的身份信息。当然，如果全部匹配失败，说明进线客户并非存量客户，此时可通过转人工接听服务进行处理，或按预设的其它方式进行操作处理。
本实施例方案，使得在客户不是使用在提供服务的公司那保存的电话号码进行拨打的情况下，也可以自动通过声纹识别出客户的身份，避免了客户必须使用公司保存过的号码进行呼入才能被识别身份的情况的局限性，使客户获得更好的服务效果。
如图3所示，图3为本发明进线身份识别方法三实施例的流程图。
本实施例的方案基于上述一实施例或二实施例。在本实施例中，该进线身份识别方法中的，采集进线客户的语音数据并提取采集到的语音数据中的声纹特征的步骤，包括：
步骤S101，随机生成预设数量字符的字符串，播报该字符串及提示用户复述该字符串的提示语；
系统通过随机的方式生成预设数量字符的字符串(例如，随机生成8位文字、字母、数字或以上任意多种形成的字符串)，并播报该字符串及播报提示用户复述该字符串的提示语(例如，“请您复述以下语句：********”)。
步骤S102，接收客户复述该字符串所对应的语音数据，提取接收到的语音数据中的内容，并将提取出的内容与系统播报的字符串的内容进行比对；
系统接收客户复述系统随机生成的字符串时的语音数据，对该语音数据进行内容分析以提取出该语音数据中的内容，将该提取出的内容与系统播报的字符串的内容进行比对，以确认是否内容相同。
步骤S103，在比对成功后，提取接收到的语音数据中的声纹特征。
若比对成功，即系统根据客户复述的语音数据所提取出的内容与系统播报的字符串的内容相同，那么则提取客户复述的语音数据中的声纹特征，以进行下一步的声纹验证。如果比对不成功，即系统根据客户复述的语音数据所提取出的内容与系统播报的字符串的内容不 相同，这种情况可能是由于客户口误、客户地方口音重等情况造成，此时，则可提示客户复述内容有误，并跳转至执行步骤S101，再次提供字符串供客户复述以再次进行比对验证；另外，如果比对不成功的次数超过预设错误阈值时，系统可提示客户验证错误次数过多，将客户转到电话接听服务人员处进行人工服务，或者也可以采取中断通话等处理方式。
本实施例方案通过系统以随机动态的方式生成字符串供客户复述，将客户复述的语音数据作为供声纹特征提取的数据，并在提取声纹特征前进行客户复述内容的比对验证，防止他人使用对客户声音的录音数据来通过系统的声纹验证，有效的确保了声纹验证的安全性。
进一步地，在本实施例中，该进线身份识别方法中的，采集进线客户的语音数据并提取采集到的语音数据中的声纹特征的步骤，还包括：
若提取出的声纹特征与确定的客户声纹特征数据匹配失败，则使错误计数加一；
在实际情况中，由于环境噪音、客户感冒生病等一些客观因素都可能会提取出的声纹特征与所述确定的客户声纹特征的匹配造成影响，因此系统会设置多次声纹验证机会，在提取出的声纹特征与确定的客户声纹特征数据匹配失败时，进行一次错误计数，根据错误计数的值的大小确定是否再次提供随机生成字符串供客户复述以进行声纹验证；
若所述错误计数小于等预设的错误计数阈值，则执行步骤S101；
如果所述错误计数小于等于预设的错误计数阈值，即确定声纹匹配错误次数还没有超过错误允许次数，此时则跳转至执行步骤S101，再次提供字符串供客户复述以再次进行验证；
若所述错误计数超过预设的错误计数阈值，则进行紧急安全处理。
当所述错误计数超过预设的错误计数阈值后，系统不再提供提供字符串，系统进行紧急安全处理。例如，该紧急安全处理可以是：1、向当前进线号码预留的联系邮箱或紧急联系电话发送异常提醒消息；2、中断当前通话，将当前进线号码锁定为异常状态，使当前进线号码不能进行相关业务操作，等等。
本实施例方案使客户具有多次声纹验证的机会，更符合实际使用时的需求，并且对声纹验证错误次数超过一定次数后，进行紧急安全处理，保障客户的权益。
本案中，声纹特征的匹配结果的判断方式可以是将声纹特征的匹配程度与预设的匹配阈值比较，在大于预设的匹配阈值时确认为匹配 成功，小于等于预设的匹配阈值时确认为匹配失败。
本发明还提出一种进线身份识别系统。
请参阅图4，是本发明进线身份识别系统10较佳实施例的运行环境示意图。
在本实施例中，进线身份识别系统10安装并运行于电子装置1中。电子装置1可以是桌上型计算机、笔记本、掌上电脑及服务器等计算设备。该电子装置1可包括，但不仅限于，存储器11、处理器12及显示器13。图3仅示出了具有组件11-13的电子装置1，但是应理解的是，并不要求实施所有示出的组件，可以替代的实施更多或者更少的组件。
存储器11在一些实施例中可以是电子装置1的内部存储单元，例如该电子装置1的硬盘或内存。存储器11在另一些实施例中也可以是电子装置1的外部存储设备，例如电子装置1上配备的插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)等。进一步地，存储器11还可以既包括电子装置1的内部存储单元也包括外部存储设备。存储器11用于存储安装于电子装置1的应用软件及各类数据，例如进线身份识别系统10的程序代码等。存储器11还可以用于暂时地存储已经输出或者将要输出的数据。
处理器12在一些实施例中可以是一中央处理器(Central Processing Unit,CPU)，微处理器或其他数据处理芯片，用于运行存储器11中存储的程序代码或处理数据，例如执行进线身份识别系统10等。
显示器13在一些实施例中可以是LED显示器、液晶显示器、触控式液晶显示器以及OLED(Organic Light-Emitting Diode，有机发光二极管)触摸器等。显示器13用于显示在电子装置1中处理的信息以及用于显示可视化的用户界面，例如业务定制界面等。电子装置1的部件11-13通过系统总线相互通信。
请参阅图5，是本发明进线身份识别系统10一实施例的程序模块图。在本实施例中，进线身份识别系统10可以被分割成一个或多个模块，一个或者多个模块被存储于存储器11中，并由一个或多个处理器(本实施例为处理器12)所执行，以完成本发明。例如，在图5中，进线身份识别系统10可以被分割成获取模块101、确定模块102、采集提取模块103、第一匹配模块104及第一身份确认模块105。本发明所称的模块是指能够完成特定功能的一系列计算机程序指令段，比程序更适合于描述进线身份识别系统10在电子装置1中的执行过程，其中：
获取模块101，用于在接收到客户来电后，根据预先确定的电话号码与客户身份信息的映射关系，获取当前进线号码对应的客户身份信息；
系统的数据库中具有电话号码与客户身份信息(例如，可包括用户名、证件号等)的映射表，该映射表中包括已经在系统中保存的所有电话号码及与这些电话号码对应的客户身份信息；当接到客户进线来电后，获取模块101根据进线客户的号码从该电话号码与客户身份信息的映射表中，去查找以获取当前进线号码对应的客户身份信息；
确定模块102，用于在成功获取客户身份信息后，根据预先确定的客户身份信息与声纹特征数据的映射关系，确定获取到的客户身份信息所对应的客户声纹特征数据；
系统中为存量客户建立了声纹特征数据库，该声纹特征数据库中预先录入保存了每个存量客户的声纹特征数据，且系统中预先设置有声纹特征数据与客户身份信息的映射关系表，在获取模块101成功获取到当前进线号码的客户身份信息后，确定模块102根据获取到的客户身份信息找出对应的客户声纹特征数据。
采集提取模块103，用于采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
采集进线客户在通话中的语音数据，根据采集的语音数据进行分析以提取出当前进线客户的声纹特征。
第一匹配模块104，用于将提取出的声纹特征与确定的客户声纹特征数据进行匹配；
第一匹配模块104将采集提取模块103提取出的声纹特征与根据确定模块102确定的客户声纹特征数据进行匹配，以确认两者是否属于同一人的声纹；
第一身份确认模块105，用于在提取出的声纹特征与确定的客户声纹特征数据匹配成功后，确认当前进线客户为该进线号码对应的客户本人。
当采集提取模块103提取出的声纹特征与确定模块102确定的声纹特征数据匹配成功，那么第一身份确认模块105就确定当前进线客户为该进线号码的客户本人，此时可根据客户需求提供相关服务；如果匹配失败，第一身份确认模块105则确认当前进线客户并非该进线号码的客户本人，有可能是被盗用或其它情形；在匹配失败的情况下，系统不会提供相关服务给进线客户。
本实施例的方案先通过进线号码获取对应的客户身份信息，根据客户身份信息从声纹特征数据库中确定其对应的预存的声纹特征数据，将从当前进线客户的语音数据中提取出的声纹特征与该确定的声纹特征数据匹配，根据匹配结果来验证当前进线客户是否是该进线号 码的客户本人，匹配成功说明当前进线客户是该进线号码的客户本人，反之则不是。与现有技术相比，本实施例的技术方案采用声纹自动验证客户身份的方式，不需要通过电话接听服务人员对客户身份进行验证，更加的方便、准确及高效，也可以避免他人盗用客户电话号码进行拨打的情况；另外，通过声纹验证的方式，准确且安全可靠，可以无需对通话进行录音，减少录音数据的保存，降低存储成本。
请参阅图6，是本发明进线身份识别系统10二实施例的程序模块图。
本实施例方案基于一实施例，在本实施例中，所述进线身份识别系统还包括第二匹配模块106和第二身份确认模块107，其中：
采集提取模块103还用于在获取客户身份信息失败后，采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；在当前进线号码是系统中没有登记记录的电话号码的情况时，根据电话号码与客户身份信息的映射关系表就找不到当前进线号码对应的客户身份信息，即获取客户身份信息失败，此种情况下，也并不能确定当前进线客户不是公司的存量客户，很可能是公司存量客户通过没有登记过的电话打进来的；因此，此时采集提取模块103对进线客户的语音数据进线采集，并从采集的进线客户的语音数据中分析提取出声纹特征，以根据声纹特征进行后续的验证操作。
第二匹配模块106，用于将提取出的声纹特征与声纹数据库中的客户声纹特征数据逐一进行匹配；
第二匹配模块106通过将从采集的进线客户的语音数据中提取出的声纹特征与声纹特征数据库中的所有客户声纹特征数据逐一进行匹配，以确认声纹特征数据库中是否存在客户声纹特征数据与当前进线客户的声纹特征匹配，从而确定当前进线客户是否为公司的存量客户。
第二身份确认模块107，用于在提取出的声纹特征与声纹数据库中的客户声纹特征数据匹配成功后，根据预先确定的客户身份信息与客户声纹特征数据的映射关系，确认提取出的声纹特征匹配到的客户声纹特征数据所对应的客户身份信息。
当声纹特征数据库中存在有客户声纹特征数据与当前进线客户的声纹特征匹配时，说明当前进线的客户是公司的存量客户通过公司没有登记过的电话号码进线的，此时第二身份确认模块107根据客户身份信息与客户声纹特征数据的映射关系，直接找出被匹配到的客户声纹特征数据所对应的客户身份信息，即为当前进线客户的身份信息。当然，如果全部匹配失败，说明进线客户并非存量客户，此时可通过转人工接听服务进行处理，或按预设的其它方式进行操作处理。
本实施例方案，使得在客户不是使用在提供服务的公司那保存的电话号码进行拨打的情况下，也可以自动通过声纹识别出客户的身份，避免了客户必须使用公司保存过的号码进行呼入才能被识别身份的情况的局限性，使客户获得更好的服务效果。
如图7所示，本实施例中，该进线身份识别系统10的采集提取模块103包括：
生成子模块1031，用于随机生成预设数量字符的字符串，播报该字符串及提示用户复述该字符串的提示语；
系统通过随机的方式生成预设数量字符的字符串(例如，随机生成8位文字、字母、数字或以上任意多种形成的字符串)，并播报该字符串及播报提示用户复述该字符串的提示语(例如，“请您复述以下语句：********”)。
比对子模块1032，用于接收客户复述该字符串所对应的语音数据，提取接收到的语音数据中的内容，并将提取出的内容与系统播报的字符串的内容进行比对；
比对子模块1032接收客户复述系统随机生成的字符串时的语音数据，对该语音数据进行内容分析以提取出该语音数据中的内容，将该提取出的内容与系统播报的字符串的内容进行比对，以确认是否内容相同。
提取子模块1033，在比对成功后，提取接收到的语音数据中的声纹特征。
若比对子模块1032确认比对成功，那么提取子模块1033则提取客户复述的语音数据中的声纹特征，以进行下一步的声纹验证。如果比对子模块1032确认比对不成功，由于这种情况可能是由于客户口误、客户地方口音重等情况造成，此时，则可提示客户复述内容有误，并跳转至执行生成子模块1031，再次提供字符串供客户复述以再次进行比对验证；另外，如果比对不成功的次数超过预设错误阈值时，系统可提示客户验证错误次数过多，将客户转到电话接听服务人员处进行人工服务，或者也可以采取中断通话等处理方式。
本实施例方案通过系统以随机动态的方式生成字符串供客户复述，将客户复述的语音数据作为供声纹特征提取的数据，并在提取声纹特征前进行客户复述内容的比对验证，防止他人使用对客户声音的录音数据来通过系统的声纹验证，有效的确保了声纹验证的安全性。
如图8所示，本实施例中，该进线身份识别系统10的采集提取模块103还包括：
计数子模块1034，用于在提取出的声纹特征与确定的客户声纹 特征数据匹配失败后，使错误计数加一；
在实际情况中，由于环境噪音、客户感冒生病等一些客观因素都可能会提取出的声纹特征与所述确定的客户声纹特征的匹配造成影响，因此系统会设置多次声纹验证机会，在提取出的声纹特征与确定的客户声纹特征数据匹配失败时，进行一次错误计数，根据错误计数的值的大小确定是否再次提供随机生成字符串供客户复述以进行声纹验证；
跳转子模块1035，用于在所述错误计数小于等预设的错误计数阈值时，随机生成预设数量字符的字符串，播报该字符串及提示用户复述该字符串的提示语；
如果所述错误计数小于等于预设的错误计数阈值，即确定声纹匹配错误次数还没有超过错误允许次数，此时跳转子模块1035则将处理流程跳转至生成子模块1031处执行，再次提供字符串供客户复述以再次进行验证；
预警子模块1036，用于在所述错误计数超过预设的错误计数阈值后，进行紧急安全处理。
当所述错误计数超过预设的错误计数阈值后，预警子模块1036进行紧急安全处理。例如，该紧急安全处理可以是：1、向当前进线号码预留的联系邮箱或紧急联系电话发送异常提醒消息；2、中断当前通话，将当前进线号码锁定为异常状态，使当前进线号码不能进行相关业务操作，等等。
本实施例方案使客户具有多次声纹验证的机会，更符合实际使用时的需求，并且对声纹验证错误次数超过一定次数后，进行紧急安全处理，保障客户的权益。
本案中，声纹特征的匹配结果的判断方式可以是将声纹特征的匹配程度与预设的匹配阈值比较，在大于预设的匹配阈值时确认为匹配成功，小于等于预设的匹配阈值时确认为匹配失败。
以上所述仅为本发明的优选实施例，并非因此限制本发明的专利范围，凡是在本发明的发明构思下，利用本发明说明书及附图内容所作的等效结构变换，或直接/间接运用在其他相关的技术领域均包括在本发明的专利保护范围内。

Claims (20)

  1. 一种电子装置，其特征在于，所述电子装置包括存储器和处理器，所述存储器上存储有可在所述处理器上运行的进线身份识别系统，所述进线身份识别系统被所述处理器执行时实现如下步骤：
    A1、在接收到客户来电后，根据预先确定的电话号码与客户身份信息的映射关系，获取当前进线号码对应的客户身份信息；
    B1、在成功获取客户身份信息后，根据预先确定的客户身份信息与声纹特征数据的映射关系，确定获取到的客户身份信息所对应的客户声纹特征数据；
    C1、采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
    D1、将提取出的声纹特征与确定的客户声纹特征数据进行匹配；
    E1、若提取出的声纹特征与确定的客户声纹特征数据匹配成功，则确认当前进线客户为该进线号码对应的客户本人。
  2. 如权利要求1所述的电子装置，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，包括：
    F1、随机生成预设数量字符的字符串，播报该字符串及提示用户复述该字符串的提示语；
    G1、接收客户复述该字符串所对应的语音数据，提取接收到的语音数据中的内容，并将提取出的内容与系统播报的字符串的内容进行比对；
    H1、在比对成功后，提取接收到的语音数据中的声纹特征。
  3. 如权利要求2所述的电子装置，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，还包括：
    在所述提取出的内容与系统播报的字符串的内容比对失败后，执行步骤F1。
  4. 如权利要求2所述的电子装置，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，还包括：
    若提取出的声纹特征与确定的客户声纹特征数据匹配失败，则使错误计数加一；
    若所述错误计数小于等于预设的错误计数阈值，则执行步骤F1；
    若所述错误计数超过预设的错误计数阈值，则进行紧急安全处 理。
  5. 如权利要求1所述的电子装置，其特征在于，所述进线身份识别系统被所述处理器执行时还实现如下步骤：
    在获取客户身份信息失败后，采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
    将提取出的声纹特征与声纹数据库中的客户声纹特征数据逐一进行匹配；
    若提取出的声纹特征与声纹数据库中的客户声纹特征数据匹配成功，则根据预先确定的客户身份信息与客户声纹特征数据的映射关系，确认提取出的声纹特征匹配到的客户声纹特征数据所对应的客户身份信息。
  6. 如权利要求5所述的电子装置，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，包括：
    F1、随机生成预设数量字符的字符串，播报该字符串及提示用户复述该字符串的提示语；
    G1、接收客户复述该字符串所对应的语音数据，提取接收到的语音数据中的内容，并将提取出的内容与系统播报的字符串的内容进行比对；
    H1、在比对成功后，提取接收到的语音数据中的声纹特征。
  7. 如权利要求6所述的电子装置，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，还包括：
    在所述提取出的内容与系统播报的字符串的内容比对失败后，执行步骤F1。
  8. 如权利要求6所述的电子装置，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，还包括：
    若提取出的声纹特征与确定的客户声纹特征数据匹配失败，则使错误计数加一；
    若所述错误计数小于等于预设的错误计数阈值，则执行步骤F1；
    若所述错误计数超过预设的错误计数阈值，则进行紧急安全处理。
  9. 一种进线身份识别方法，其特征在于，所述方法包括：
    A2、在接收到客户来电后，根据预先确定的电话号码与客户身 份信息的映射关系，获取当前进线号码对应的客户身份信息；
    B2、在成功获取客户身份信息后，根据预先确定的客户身份信息与声纹特征数据的映射关系，确定获取到的客户身份信息所对应的客户声纹特征数据；
    C2、采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
    D2、将提取出的声纹特征与确定的客户声纹特征数据进行匹配；
    E2、若提取出的声纹特征与确定的客户声纹特征数据匹配成功，则确认当前进线客户为该进线号码对应的客户本人。
  10. 如权利要求9所述的进线身份识别方法，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，包括：
    F2、随机生成预设数量字符的字符串，播报该字符串及提示用户复述该字符串的提示语；
    G2、接收客户复述该字符串所对应的语音数据，提取接收到的语音数据中的内容，并将提取出的内容与系统播报的字符串的内容进行比对；
    H2、在比对成功后，提取接收到的语音数据中的声纹特征。
  11. 如权利要求10所述的进线身份识别方法，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，还包括：
    在所述提取出的内容与系统播报的字符串的内容比对失败后，执行步骤F2。
  12. 如权利要求10所述的进线身份识别方法，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，还包括：
    若提取出的声纹特征与确定的客户声纹特征数据匹配失败，则使错误计数加一；
    若所述错误计数小于等于预设的错误计数阈值，则执行步骤F2；
    若所述错误计数超过预设的错误计数阈值，则进行紧急安全处理。
  13. 如权利要求9所述的进线身份识别方法，其特征在于，所述方法还包括：
    在获取客户身份信息失败后，采集进线客户的语音数据，提取采 集到的语音数据中的声纹特征；
    将提取出的声纹特征与声纹数据库中的客户声纹特征数据逐一进行匹配；
    若提取出的声纹特征与声纹数据库中的客户声纹特征数据匹配成功，则根据预先确定的客户身份信息与客户声纹特征数据的映射关系，确认提取出的声纹特征匹配到的客户声纹特征数据所对应的客户身份信息。
  14. 如权利要求13所述的进线身份识别方法，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，包括：
    F2、随机生成预设数量字符的字符串，播报该字符串及提示用户复述该字符串的提示语；
    G2、接收客户复述该字符串所对应的语音数据，提取接收到的语音数据中的内容，并将提取出的内容与系统播报的字符串的内容进行比对；
    H2、在比对成功后，提取接收到的语音数据中的声纹特征。
  15. 如权利要求14所述的进线身份识别方法，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，还包括：
    在所述提取出的内容与系统播报的字符串的内容比对失败后，执行步骤F2。
  16. 如权利要求14所述的进线身份识别方法，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，还包括：
    若提取出的声纹特征与确定的客户声纹特征数据匹配失败，则使错误计数加一；
    若所述错误计数小于等于预设的错误计数阈值，则执行步骤F2；
    若所述错误计数超过预设的错误计数阈值，则进行紧急安全处理。
  17. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有进线身份识别系统，所述进线身份识别系统可被至少一个处理器执行，以使所述至少一个处理器执行如下操作：
    A3、在接收到客户来电后，根据预先确定的电话号码与客户身份信息的映射关系，获取当前进线号码对应的客户身份信息；
    B3、在成功获取客户身份信息后，根据预先确定的客户身份信息与声纹特征数据的映射关系，确定获取到的客户身份信息所对应的客户声纹特征数据；
    C3、采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
    D3、将提取出的声纹特征与确定的客户声纹特征数据进行匹配；
    E3、若提取出的声纹特征与确定的客户声纹特征数据匹配成功，则确认当前进线客户为该进线号码对应的客户本人。
  18. 如权利要求17所述的计算机可读存储介质，其特征在于，所述进线身份识别系统被至少一个处理器执行时，所述至少一个处理器还执行如下操作：
    在获取客户身份信息失败后，采集进线客户的语音数据，提取采集到的语音数据中的声纹特征；
    将提取出的声纹特征与声纹数据库中的客户声纹特征数据逐一进行匹配；
    若提取出的声纹特征与声纹数据库中的客户声纹特征数据匹配成功，则根据预先确定的客户身份信息与客户声纹特征数据的映射关系，确认提取出的声纹特征匹配到的客户声纹特征数据所对应的客户身份信息。
  19. 如权利要求18所述的计算机可读存储介质，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，包括：
    F3、随机生成预设数量字符的字符串，播报该字符串及提示用户复述该字符串的提示语；
    G3、接收客户复述该字符串所对应的语音数据，提取接收到的语音数据中的内容，并将提取出的内容与系统播报的字符串的内容进行比对；
    H3、在比对成功后，提取接收到的语音数据中的声纹特征。
  20. 如权利要求19所述的计算机可读存储介质，其特征在于，所述采集进线客户的语音数据，提取采集到的语音数据中的声纹特征的步骤，还包括：
    若提取出的声纹特征与确定的客户声纹特征数据匹配失败，则使错误计数加一；
    若所述错误计数小于等于预设的错误计数阈值，则执行步骤F3；
    若所述错误计数超过预设的错误计数阈值，则进行紧急安全处 理。
PCT/CN2017/104933 2017-06-24 2017-09-30 电子装置、进线身份识别方法和计算机可读存储介质 WO2018233137A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710489752.2A CN107690036A (zh) 2017-06-24 2017-06-24 电子装置、进线身份识别方法和计算机可读存储介质
CN201710489752.2 2017-06-24

Publications (1)

Publication Number Publication Date
WO2018233137A1 true WO2018233137A1 (zh) 2018-12-27

Family

ID=61152635

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/104933 WO2018233137A1 (zh) 2017-06-24 2017-09-30 电子装置、进线身份识别方法和计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN107690036A (zh)
WO (1) WO2018233137A1 (zh)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108810295A (zh) * 2018-03-28 2018-11-13 平安科技（深圳）有限公司 服务提供方法、电子设备及存储介质
CN108766444B (zh) * 2018-04-09 2020-11-03 平安科技（深圳）有限公司 用户身份验证方法、服务器及存储介质
CN108900326B (zh) * 2018-06-15 2021-08-31 中国联合网络通信集团有限公司 通信信息管理方法及装置
CN109302541A (zh) * 2018-11-12 2019-02-01 平安科技（深圳）有限公司 电子装置、坐席分配方法和计算机可读存储介质
CN110619880A (zh) * 2019-10-14 2019-12-27 百可录(北京)科技有限公司 一种声纹处理系统及用户识别方法
CN111131635B (zh) * 2019-12-20 2020-11-10 中国银行股份有限公司 一种数据请求的语音确认方法及装置
CN111563099B (zh) * 2020-07-02 2021-01-22 浙江百应科技有限公司 一种基于声纹识别的crm客户查重方法及其电子设备
CN111813454B (zh) * 2020-07-07 2024-04-23 深圳微步信息股份有限公司 电子装置、开机画面显示方法和计算机可读存储介质
CN111833068A (zh) * 2020-07-31 2020-10-27 重庆富民银行股份有限公司 基于声纹识别的身份核验系统及方法
CN113707157B (zh) * 2021-08-30 2023-08-01 平安科技（深圳）有限公司 基于声纹识别的身份验证方法、装置、电子设备及介质
CN115021937B (zh) * 2022-06-21 2024-02-09 中国银行股份有限公司 一种用户身份认证方法、系统、电子设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103501296A (zh) * 2013-09-27 2014-01-08 厦门天聪智能软件有限公司 基于声纹识别技术的社保生存认证系统及其认证方法
CN103700371A (zh) * 2013-12-13 2014-04-02 江苏大学 一种基于声纹识别的来电身份识别系统及其识别方法
US9247059B1 (en) * 2014-11-03 2016-01-26 Verizon Patent And Licensing Inc. Priority token-based interactive voice response server
CN105306657A (zh) * 2014-06-20 2016-02-03 中兴通讯股份有限公司 身份识别方法、装置及通讯终端
CN105933493A (zh) * 2016-04-20 2016-09-07 乐视控股（北京）有限公司 一种陌生来电处理方法、装置及移动终端

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1905445B (zh) * 2005-07-27 2012-02-15 国际商业机器公司 使用可移动的语音标识卡的语音认证系统及语音认证方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103501296A (zh) * 2013-09-27 2014-01-08 厦门天聪智能软件有限公司 基于声纹识别技术的社保生存认证系统及其认证方法
CN103700371A (zh) * 2013-12-13 2014-04-02 江苏大学 一种基于声纹识别的来电身份识别系统及其识别方法
CN105306657A (zh) * 2014-06-20 2016-02-03 中兴通讯股份有限公司 身份识别方法、装置及通讯终端
US9247059B1 (en) * 2014-11-03 2016-01-26 Verizon Patent And Licensing Inc. Priority token-based interactive voice response server
CN105933493A (zh) * 2016-04-20 2016-09-07 乐视控股（北京）有限公司 一种陌生来电处理方法、装置及移动终端

Also Published As

Publication number Publication date
CN107690036A (zh) 2018-02-13

Similar Documents

Publication Publication Date Title
WO2018233137A1 (zh) 电子装置、进线身份识别方法和计算机可读存储介质
CN109147797B (zh) 基于声纹识别的客服方法、装置、计算机设备及存储介质
US20170118205A1 (en) User biological feature authentication method and system
CN110489415B (zh) 一种数据更新方法及相关设备
US20160014120A1 (en) Method, server, client and system for verifying verification codes
WO2019196303A1 (zh) 用户身份验证方法、服务器及存储介质
EP2887267A1 (en) Biometric authentication device and reference data verification method
WO2020181809A1 (zh) 基于接口校验的数据处理的方法、系统和计算机设备
CN109089231B (zh) 验证码短信的转发方法、装置、服务器和介质
US20210312023A1 (en) Agent Login Method, Electronic Device And Storage Medium Based On Voiceprint Identification
US9942752B1 (en) Method and system for detecting phishing calls using one-time password
CN109245902A (zh) 即时通信信息验证码的保护方法及装置
CN111343162A (zh) 系统安全登录方法、装置、介质及电子设备
WO2018233314A1 (zh) 电子装置、ivr语音支付方法和计算机可读存储介质
CN105050061A (zh) 一种基于用户声纹信息的消息发送方法及装置
WO2016141697A1 (zh) 临时联系人的存储方法及装置
CN110113346B (zh) 一种网络验证方法、用户终端及服务器
WO2020024415A1 (zh) 声纹识别处理方法、装置、电子设备及存储介质
WO2019140851A1 (zh) 电话销售提示方法、电子装置及可读存储介质
CN113886242A (zh) 数据处理方法、装置、终端及存储介质
WO2017084513A1 (zh) 一种核验信息处理方法及服务器
WO2016188079A1 (zh) 终端设备的数据存储方法及终端设备
CN107896208B (zh) 一种身份认证方法及系统
CN105472162B (zh) 一种信息处理方法及电子设备
CN109327814B (zh) 短消息的处理方法、装置、电子设备及可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17914634

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 12/05/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 17914634

Country of ref document: EP

Kind code of ref document: A1