WO2019127929A1 - 一种电子设备声纹支付方法及装置 - Google Patents

一种电子设备声纹支付方法及装置 Download PDF

Info

Publication number
WO2019127929A1
WO2019127929A1 PCT/CN2018/079533 CN2018079533W WO2019127929A1 WO 2019127929 A1 WO2019127929 A1 WO 2019127929A1 CN 2018079533 W CN2018079533 W CN 2018079533W WO 2019127929 A1 WO2019127929 A1 WO 2019127929A1
Authority
WO
WIPO (PCT)
Prior art keywords
audio data
real
time
payment
acoustic features
Prior art date
Application number
PCT/CN2018/079533
Other languages
English (en)
French (fr)
Inventor
郭一明
Original Assignee
广州势必可赢网络科技有限公司
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 广州势必可赢网络科技有限公司 filed Critical 广州势必可赢网络科技有限公司
Publication of WO2019127929A1 publication Critical patent/WO2019127929A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q20/00—Payment architectures, schemes or protocols
    • G06Q20/30—Payment architectures, schemes or protocols characterised by the use of specific devices or networks
    • G06Q20/32—Payment architectures, schemes or protocols characterised by the use of specific devices or networks using wireless devices
    • G06Q20/327—Short range or proximity payments by means of M-devices
    • G06Q20/3272—Short range or proximity payments by means of M-devices using an audio code
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q20/00—Payment architectures, schemes or protocols
    • G06Q20/38—Payment protocols; Details thereof
    • G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
    • G06Q20/401—Transaction verification
    • G06Q20/4014—Identity check for transactions
    • G06Q20/40145—Biometric identity checks

Definitions

  • the embodiments of the present invention relate to the field of electronic payment, and in particular, to a method and device for paying voiceprint of an electronic device.
  • verification is often performed by a verification code such as a verification code and a conventional input payment password, and a password or a verification code is required to be manually input.
  • a verification code such as a verification code and a conventional input payment password
  • a password or a verification code is required to be manually input.
  • the process is cumbersome, but if the secret payment is directly used, the user is Account security has a greater risk.
  • the present application provides an electronic device voiceprint payment method and device, which solves the technical problem that the current electronic device payment method is difficult to satisfy both convenience and security.
  • the application provides a voiceprint payment method for an electronic device, including:
  • S1 receiving a payment instruction, and acquiring real-time audio data of real-time monitoring in a preset time period in the buffer area according to the payment instruction;
  • S2 extracting real-time acoustic features in the real-time audio data, matching the real-time acoustic features with the preset acoustic features in the voiceprint database to obtain a matching score, determining whether the matching score is greater than a preset threshold, and if so, performing a payment operation.
  • step S2 further comprising: step S3;
  • the S2 specifically includes: extracting real-time acoustic features in the real-time audio data, matching the real-time acoustic features with the preset acoustic features in the voiceprint database to obtain a matching score, determining whether the matching score is greater than a preset threshold, and if so, performing a payment operation, If not, proceed to step S3;
  • step S2 specifically includes:
  • S21 performing endpoint detection on real-time audio data, and acquiring speaker audio data in real-time audio data;
  • the method further includes:
  • S02 Extract the registered acoustic features in the registered audio data, and store the registered acoustic features in the voiceprint database as preset acoustic features.
  • step S01 before step S1, further comprising: step S03;
  • the S01 specifically includes: receiving the registration instruction and registering the audio data, determining whether the registered audio data meets the requirements of the preset voice quality standard according to the registration instruction, if yes, executing step S02; if not, executing step S03;
  • the application provides an electronic device voiceprint payment device, comprising:
  • a data obtaining unit configured to receive a payment instruction, and obtain real-time audio data of real-time monitoring in a preset time period in the buffer area according to the payment instruction;
  • the feature matching unit is configured to extract real-time acoustic features in the real-time audio data, match the real-time acoustic features with the preset acoustic features in the voiceprint database to obtain a matching score, determine whether the matching score is greater than a preset threshold, and if so, perform payment operating.
  • the method further includes: a failure reminding unit;
  • the feature matching unit is specifically configured to extract real-time acoustic features in the real-time audio data, and match the real-time acoustic features with the preset acoustic features in the voiceprint database to obtain a matching score, and determine whether the matching score is greater than a preset threshold, and if so, execute Payment operation, if not, triggering a failure reminder unit;
  • the failure reminding unit is used for voiceprint verification failure, prompting the operation user to input a payment password.
  • the feature matching unit specifically includes:
  • a detecting subunit configured to perform endpoint detection on real-time audio data, and obtain speaker audio data in real-time audio data
  • a feature sub-unit for denoising the speaker audio data, extracting features of the denominated speaker audio data, and acquiring real-time acoustic features in the speaker audio data;
  • a matching subunit configured to match the real-time acoustic feature with the preset acoustic feature in the voiceprint database to obtain a matching score, determine whether the matching score is greater than a preset threshold, and if so, perform a payment operation, and if not, trigger a failure reminding unit .
  • the method further comprises:
  • a voice registration unit configured to receive a registration instruction and register audio data, and determine, according to the registration instruction, whether the registered audio data meets the requirements of the preset voice quality standard, and if yes, trigger the extraction of the archive unit;
  • An archive unit is extracted for extracting registered acoustic features in the registered audio data, and the registered acoustic features are stored in the voiceprint database as preset acoustic features.
  • the method further includes: a failure return unit;
  • the voice registration unit is specifically configured to receive the registration instruction and the registered audio data, and determine, according to the registration instruction, whether the registered audio data meets the requirement of the preset voice quality standard, and if yes, triggers the extraction of the archive unit, and if not, triggers the failure return unit;
  • the failure return unit is used for voiceprint registration failure, prompting the user to re-enter the registered audio data and trigger the voice registration unit.
  • the present application has the following advantages:
  • the present application provides an electronic device voiceprint payment method, including: S1: receiving a payment instruction, acquiring real-time audio data of real-time monitoring in a preset time period in a buffer area according to a payment instruction; S2: extracting real-time data in real-time audio data The acoustic feature matches the real-time acoustic features with the preset acoustic features in the voiceprint database to obtain a matching score, determines whether the matching score is greater than a preset threshold, and if so, performs a payment operation.
  • the electronic device when the user selects the payment, the electronic device receives the payment instruction, and the real-time audio data in the real-time monitoring period in the preset time period is called from the buffer area, and the real-time acoustic features in the real-time audio data are extracted and the real-time acoustic characteristics are extracted.
  • Matching with the preset acoustic features determining whether the electronic device owner's own voice exists in the real-time audio data according to the matching score, and if so, indicating that the electronic device is used by the electronic device owner himself, and the electronic device performs the payment operation, in the whole process No need for the user to input the verification code or payment password, the user only needs to release the payment instruction, the electronic device will automatically judge whether to pay according to the real-time audio data, the payment process is convenient and fast, and because the acoustic characteristics corresponding to each person's voice are different, so through the acoustic
  • the feature verification is safe and reliable, and solves the technical problem that the current electronic device payment method is difficult to satisfy both convenience and security.
  • FIG. 1 is a schematic flowchart of an embodiment of a voiceprint payment method for an electronic device according to an embodiment of the present disclosure
  • FIG. 2 is a schematic flowchart diagram of another embodiment of a voiceprint payment method for an electronic device according to an embodiment of the present disclosure
  • FIG. 3 is a schematic structural diagram of an embodiment of an electronic device voiceprint payment device according to an embodiment of the present disclosure.
  • the embodiment of the present application provides a method and a device for paying a voiceprint of an electronic device, which solves the technical problem that the current electronic device payment method is difficult to satisfy both convenience and security.
  • an embodiment of the present application provides an embodiment of a voiceprint payment method for an electronic device, including:
  • Step 101 Receive a payment instruction, and acquire real-time audio data of real-time monitoring in a preset time period in the buffer area according to the payment instruction.
  • the electronic device receives the payment instruction, and obtains the real-time audio data of the real-time monitoring in the preset time period in the buffer according to the payment instruction.
  • Step 102 Extract the real-time acoustic features in the real-time audio data, match the real-time acoustic features with the preset acoustic features in the voiceprint database to obtain a matching score, determine whether the matching score is greater than a preset threshold, and if yes, perform step 103.
  • real-time audio data in real-time monitoring in a preset time period in the buffer area is acquired, real-time acoustic features in the real-time audio data are extracted, and real-time acoustic features are matched with preset acoustic features in the voiceprint database. Matching the score, determining whether the matching score is greater than a preset threshold to decide whether to perform step 103 for payment.
  • Step 103 Perform a payment operation.
  • the electronic device when the user selects the payment in the embodiment, when the electronic device receives the payment instruction, the real-time audio data of the real-time monitoring within the preset time period is called from the buffer area, and the real-time acoustic data in the real-time audio data is extracted.
  • the feature matches the real-time acoustic feature with the preset acoustic feature, and determines whether the electronic device owner's own voice exists in the real-time audio data according to the matching score, and if so, the electronic device is used by the electronic device owner himself, and the electronic device performs Payment operation, the user does not need to input the verification code or payment password in the whole process, the user only needs to release the payment instruction, the electronic device will automatically judge whether to pay according to the real-time audio data, the payment process is convenient and fast, and because of the acoustic characteristics difference corresponding to each person's voice Larger, so the verification by acoustic features is safe and reliable, and solves the technical problem that the current electronic device payment method is difficult to satisfy both convenience and security.
  • an embodiment of the present application provides another embodiment of an electronic device voiceprint payment method, including:
  • Step 201 Receive a registration instruction and register audio data, and determine, according to the registration instruction, whether the registered audio data meets the requirements of the preset voice quality standard. If yes, step 202 is performed, and if not, step 203 is performed.
  • the preset voice quality standard may include, but is not limited to, a preset signal to noise ratio standard, a preset volume standard, and a preset effective duration standard.
  • Step 202 Extract the registered acoustic features in the registered audio data, and store the registered acoustic features in the voiceprint database as preset acoustic features.
  • the preset acoustic features in the voiceprint database may be preset acoustic features directly acquired from other devices or systems, or may be registered acoustic features obtained by registration processing.
  • the registered acoustic features in the registered audio data are extracted, and the registered acoustic features are stored in the voiceprint database as preset acoustic features.
  • Acoustic features may include, but are not limited to, sound spectrum features, prosodic features, and sound quality features.
  • Step 203 Voiceprint registration fails, prompting the user to re-enter the registered audio data and returning to step 201.
  • the voiceprint registration fails, prompting the user to re-enter the registered audio data and returning to step 201 to perform the registration operation again.
  • Step 204 Receive a payment instruction, and acquire real-time audio data of real-time monitoring in a preset time period in the buffer area according to the payment instruction.
  • the real-time audio data of the real-time monitoring in the preset time period in the buffer area is acquired, and the preset time period can be set as needed, such as 30 seconds and 1 minute before the payment instruction is received. And 5 minutes and so on.
  • Step 205 Perform endpoint detection on the real-time audio data to obtain speaker audio data in the real-time audio data.
  • the endpoint detection of the real-time audio data can distinguish the speaker audio data and the non-speaker audio data in the real-time audio data, and only processing the speaker audio data in the subsequent steps can save data processing amount and improve processing efficiency. , reducing processing time.
  • Step 206 Perform denoising processing on the speaker audio data, perform feature extraction on the denominated speaker audio data, and acquire real-time acoustic features in the speaker audio data.
  • denoising the speaker audio data can remove the background noise and the like, and improve the accuracy of the speaker identification.
  • Step 207 Match the real-time acoustic feature with the preset acoustic feature in the voiceprint database to obtain a matching score, determine whether the matching score is greater than a preset threshold, and if yes, execute step 208, and if no, execute step 209.
  • the real-time acoustic feature is matched with the preset acoustic feature in the voiceprint database to obtain a matching score. If the matching score is greater than the preset threshold, the electronic device is currently performing a payment operation for the voiceprint registered user, and then executing Step 208 performs a payment operation. If the matching score is less than the preset threshold, it indicates that the electronic device is not currently a voiceprint registered user performing the operation, and the voiceprint verification fails.
  • Step 208 Perform a payment operation.
  • Step 209 The voiceprint verification fails, prompting the operation user to input the payment password.
  • the voiceprint verification fails, in order to avoid the situation that the voiceprint verification error or the like causes the payment to be unsuccessful, the user may be prompted to input the payment password, and the payment password is used for the payment verification.
  • the electronic device when the user selects the payment, the electronic device receives the payment instruction, and the real-time audio data in the real-time monitoring period in the preset time period is called from the buffer area, and the real-time acoustic characteristics in the real-time audio data are extracted and the real-time acoustics are extracted.
  • the feature is matched with the preset acoustic feature, and the presence or absence of the voiceprint registration user's own voice in the real-time audio data is determined according to the matching score. If yes, the electronic device is the voiceprint registered user himself, and the electronic device performs the payment operation. The user does not need to input the verification code or the payment password, and the user only needs to release the payment instruction.
  • the electronic device automatically judges whether to pay according to the real-time audio data, and the payment process is convenient and fast, and because each person's voice corresponds to a large difference in acoustic characteristics, so The acoustic features are verified to be safe and reliable, and solve the technical problems that the current electronic device payment method is difficult to satisfy both convenience and security.
  • the above is another embodiment of the electronic device voiceprint payment method provided by the embodiment of the present application.
  • the following is an embodiment of the electronic device voiceprint payment device provided by the embodiment of the present application.
  • an embodiment of the present application provides an embodiment of an electronic device voiceprint payment device, including:
  • the data obtaining unit 303 is configured to receive a payment instruction, and acquire real-time audio data of real-time monitoring in a preset time period in the buffer area according to the payment instruction.
  • the feature matching unit 304 is configured to extract real-time acoustic features in the real-time audio data, and match the real-time acoustic features with the preset acoustic features in the voiceprint database to obtain a matching score, determine whether the matching score is greater than a preset threshold, and if so, execute Payment operation.
  • the method further includes: a failure reminding unit 305.
  • the feature matching unit 304 is specifically configured to extract real-time acoustic features in the real-time audio data, and match the real-time acoustic features with the preset acoustic features in the voiceprint database to obtain a matching score, and determine whether the matching score is greater than a preset threshold, and if so, The payment operation is performed, and if not, the failure reminding unit 305 is triggered.
  • the failure reminding unit 305 is used for voiceprint verification failure, and prompts the operation user to input a payment password.
  • the feature matching unit 304 specifically includes:
  • the detecting subunit 3041 is configured to perform endpoint detection on the real-time audio data, and obtain speaker audio data in the real-time audio data.
  • the feature subunit 3042 is configured to perform denoising processing on the speaker audio data, perform feature extraction on the denominated speaker audio data, and acquire real-time acoustic features in the speaker audio data.
  • the matching sub-unit 3043 is configured to match the real-time acoustic feature with the preset acoustic feature in the voiceprint database to obtain a matching score, determine whether the matching score is greater than a preset threshold, and if yes, perform a payment operation, and if not, trigger a failure reminder Unit 305.
  • the voice registration unit 300 is configured to receive the registration instruction and the registration audio data, and determine, according to the registration instruction, whether the registered audio data meets the requirements of the preset voice quality standard, and if so, trigger the extraction archive unit 301;
  • the extraction archive unit 301 is configured to extract the registered acoustic features in the registered audio data, and store the registered acoustic features in the voiceprint database as preset acoustic features.
  • the method further includes: a failure return unit 302.
  • the voice registration unit 300 is specifically configured to receive the registration instruction and the registered audio data, determine whether the registered audio data meets the requirements of the preset voice quality standard according to the registration instruction, and if yes, trigger the extraction archive unit 301, and if not, trigger the failure return unit. 302.
  • the failure return unit 302 for voiceprint registration failure, prompts the user to re-enter the registered audio data and triggers the voice registration unit 300.
  • the disclosed apparatus and method may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
  • the integrated unit if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium.
  • a computer readable storage medium A number of instructions are included to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present application.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Finance (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Cash Registers Or Receiving Machines (AREA)
  • Telephonic Communication Services (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

本申请实施例提供了一种电子设备声纹支付方法及装置。当用户选择支付时，电子设备就会从缓存区中调用预置时间段内的实时监听的实时音频数据，提取实时音频数据中的实时声学特征并将实时声学特征与预置声学特征进行匹配，根据匹配分数判断实时音频数据中是否存在电子设备拥有者的声音，若是，则说明电子设备是电子设备拥有者本人在使用，电子设备执行支付操作，整个过程中无需用户输入验证码或者支付密码，用户只需要下达支付指令电子设备就会根据实时音频数据自动判断是否支付，支付过程方便快捷，因为每个人声音的声学特征差异较大，所以通过声学特征进行验证安全可靠，解决了当前电子设备支付方法难以同时满足方便和安全的技术问题。

Description

一种电子设备声纹支付方法及装置
本申请要求于2017年12月26日提交中国专利局、申请号为201711435244.2、发明名称为“一种电子设备声纹支付方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请实施例涉及电子支付领域，尤其涉及一种电子设备声纹支付方法及装置。
背景技术
当前在使用电子设备进行付款操作时，往往通过验证码和传统的输入支付密码等验证方式进行验证，需要手动输入密码或者验证码，过程较为繁琐，但是如果直接采用免密支付，则对于用户的账号安全存在较大的风险。
因此，导致了当前电子设备支付方法难以同时满足方便和安全的技术问题。
发明内容
本申请提供了一种电子设备声纹支付方法及装置，解决了当前电子设备支付方法难以同时满足方便和安全的技术问题。
本申请提供了一种电子设备声纹支付方法，包括：
S1：接收支付指令，根据支付指令获取缓存区中预置时间段内的实时监听的实时音频数据；
S2：提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作。
优选地，步骤S2之后还包括：步骤S3；
S2具体包括：提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作，若否，则执行步骤S3；
S3：声纹验证失败，提示操作用户输入支付密码。
优选地，步骤S2具体包括：
S21：对实时音频数据进行端点检测，获取实时音频数据中的说话人音频数据；
S22：对说话人音频数据进行去噪处理，对去噪后的说话人音频数据进行特征提取，获取说话人音频数据中的实时声学特征；
S23：将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作，若否，则执行步骤S3。
优选地，步骤S1之前还包括：
S01：接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则执行步骤S01；
S02：提取注册音频数据中的注册声学特征，将注册声学特征存入声纹数据库中作为预置声学特征。
优选地，步骤S01之后，步骤S1之前还包括：步骤S03；
S01具体包括：接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则执行步骤S02，若否，则执行步骤S03；
S03：声纹注册失败，提示用户重新输入注册音频数据并返回步骤S01。
本申请提供了一种电子设备声纹支付装置，包括：
数据获取单元，用于接收支付指令，根据支付指令获取缓存区中预置时间段内的实时监听的实时音频数据；
特征匹配单元，用于提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作。
优选地，还包括：失败提醒单元；
特征匹配单元，具体用于提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作，若否，则触发失败提醒单元；
失败提醒单元，用于声纹验证失败，提示操作用户输入支付密码。
优选地，特征匹配单元具体包括：
检测子单元，用于对实时音频数据进行端点检测，获取实时音频数据中的说话人音频数据；
特征子单元，用于对说话人音频数据进行去噪处理，对去噪后的说话人音频数据进行特征提取，获取说话人音频数据中的实时声学特征；
匹配子单元，用于将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作， 若否，则触发失败提醒单元。
优选地，还包括：
语音注册单元，用于接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则触发提取存档单元；
提取存档单元，用于提取注册音频数据中的注册声学特征，将注册声学特征存入声纹数据库中作为预置声学特征。
优选地，还包括：失败返回单元；
语音注册单元，具体用于接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则触发提取存档单元，若否，则触发失败返回单元；
失败返回单元，用于声纹注册失败，提示用户重新输入注册音频数据并触发语音注册单元。
从以上技术方案可以看出，本申请具有以下优点：
本申请提供了一种电子设备声纹支付方法，包括：S1：接收支付指令，根据支付指令获取缓存区中预置时间段内的实时监听的实时音频数据；S2：提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作。
本申请中当用户选择支付时，电子设备接收支付指令，就会从缓存区中调用预置时间段内的实时监听的实时音频数据，通过提取实时音频数据中的实时声学特征并将实时声学特征与预置声学特征进行匹配，根据匹配分数判断实时音频数据中是否存在电子设备拥有者自身的声音，若是，则说明电子设备是电 子设备拥有者本人在使用，电子设备执行支付操作，整个过程中无需用户输入验证码或者支付密码，用户只需要下达支付指令电子设备就会根据实时音频数据自动判断是否支付，支付过程方便快捷，同时因为每个人的声音对应的声学特征差异较大，所以通过声学特征进行验证安全可靠，解决了当前电子设备支付方法难以同时满足方便和安全的技术问题。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其它的附图。
图1为本申请实施例提供的一种电子设备声纹支付方法的一个实施例的流程示意图；
图2为本申请实施例提供的一种电子设备声纹支付方法的另一个实施例的流程示意图；
图3为本申请实施例提供的一种电子设备声纹支付装置的一个实施例的结构示意图。
具体实施方式
本申请实施例提供了一种电子设备声纹支付方法及装置，解决了当前电子设备支付方法难以同时满足方便和安全的技术问题。
为使得本申请的发明目的、特征、优点能够更加的明显和易懂，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描 述，显然，下面所描述的实施例仅仅是本申请一部分实施例，而非全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其它实施例，都属于本申请保护的范围。
请参阅图1，本申请实施例提供了一种电子设备声纹支付方法的一个实施例，包括：
步骤101：接收支付指令，根据支付指令获取缓存区中预置时间段内的实时监听的实时音频数据。
需要说明的是，用户需要支付时点击支付选项，电子设备会接收到支付指令，并根据支付指令获取缓存区中预置时间段内的实时监听的实时音频数据。
步骤102：提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行步骤103。
需要说明的是，获取缓存区中预置时间段内的实时监听的实时音频数据之后，提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值以决定是否执行步骤103进行支付。
步骤103：执行支付操作。
需要说明的是，本实施例中当用户选择支付时，电子设备接收支付指令，就会从缓存区中调用预置时间段内的实时监听的实时音频数据，通过提取实时音频数据中的实时声学特征并将实时声学特征与预置声学特征进行匹配，根据匹配分数判断实时音频数据中是否存在电子设备拥有者自身的声音，若是，则说明电子设备是电子设备拥有者本人在使用，电子设备执行支付操作，整个过 程中无需用户输入验证码或者支付密码，用户只需要下达支付指令电子设备就会根据实时音频数据自动判断是否支付，支付过程方便快捷，同时因为每个人的声音对应的声学特征差异较大，所以通过声学特征进行验证安全可靠，解决了当前电子设备支付方法难以同时满足方便和安全的技术问题。
以上为本申请实施例提供的一种电子设备声纹支付方法的一个实施例，以下为本申请实施例提供的一种电子设备声纹支付方法的另一个实施例。
请参阅图2，本申请实施例提供了一种电子设备声纹支付方法的另一个实施例，包括：
步骤201：接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则执行步骤202，若否，则执行步骤203。
需要说明的是，预置语音质量标准可以包括但不限于：预置信噪比标准、预置音量标准和预置有效时长标准。
步骤202：提取注册音频数据中的注册声学特征，将注册声学特征存入声纹数据库中作为预置声学特征。
需要说明的是，声纹数据库中的预置声学特征可以是从其他设备或者系统直接获取的预置声学特征，也可以是通过注册进行处理得到的注册声学特征。
如果注册音频数据符合预置语音质量标准的要求，则提取注册音频数据中的注册声学特征，将注册声学特征存入声纹数据库中作为预置声学特征。
声学特征可以包括但不限于：声谱特征、韵律学特征和音质特征。
步骤203：声纹注册失败，提示用户重新输入注册音频数据并返回步骤201。
需要说明的是，如果注册音频数据不符合预置语音质量标准的要求，则声纹注册失败，提示用户重新输入注册音频数据并返回步骤201重新进行注册操作。
步骤204：接收支付指令，根据支付指令获取缓存区中预置时间段内的实时监听的实时音频数据。
需要说明的是，接收到支付指令之后获取缓存区中预置时间段内的实时监听的实时音频数据，预置时间段可以根据需要进行设定，如接收到支付指令之前的30秒、1分钟和5分钟等。
步骤205：对实时音频数据进行端点检测，获取实时音频数据中的说话人音频数据。
需要说明的是，对实时音频数据进行端点检测可以辨别实时音频数据中的说话人音频数据和非说话人音频数据，后续步骤中只对说话人音频数据进行处理可以节省数据处理量，提高处理效率，减少处理时间。
步骤206：对说话人音频数据进行去噪处理，对去噪后的说话人音频数据进行特征提取，获取说话人音频数据中的实时声学特征。
需要说明的是，对说话人音频数据进行去噪处理可以将背景噪声等去除，提高说话人身份辨识的准确性。
步骤207：将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行步骤208，若否，则执行步骤209。
需要说明的是，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，如果匹配分数大于预置阈值，则说明电子设备当前为声纹注册 用户在进行支付操作，则执行步骤208进行支付操作，如果匹配分数小于预置阈值，则说明电子设备当前并非是声纹注册用户在进行操作，声纹验证失败。
步骤208：执行支付操作。
需要说明的是，如何进行支付操作为本领域技术人员公知技术手段，在此不再赘述。
步骤209：声纹验证失败，提示操作用户输入支付密码。
需要说明的是，声纹验证失败时，为了避免存在声纹验证错误等原因造成无法支付的情况，可以提示用户输入支付密码，通过支付密码进行支付验证。
本实施例中当用户选择支付时，电子设备接收支付指令，就会从缓存区中调用预置时间段内的实时监听的实时音频数据，通过提取实时音频数据中的实时声学特征并将实时声学特征与预置声学特征进行匹配，根据匹配分数判断实时音频数据中是否存在声纹注册用户自身的声音，若是，则说明电子设备是声纹注册用户本人在使用，电子设备执行支付操作，整个过程中无需用户输入验证码或者支付密码，用户只需要下达支付指令电子设备就会根据实时音频数据自动判断是否支付，支付过程方便快捷，同时因为每个人的声音对应的声学特征差异较大，所以通过声学特征进行验证安全可靠，解决了当前电子设备支付方法难以同时满足方便和安全的技术问题。
以上为本申请实施例提供的一种电子设备声纹支付方法的另一个实施例，以下为本申请实施例提供的一种电子设备声纹支付装置的一个实施例。
请参阅图3，本申请实施例提供了一种电子设备声纹支付装置的一个实施例，包括：
数据获取单元303，用于接收支付指令，根据支付指令获取缓存区中预置 时间段内的实时监听的实时音频数据。
特征匹配单元304，用于提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作。
进一步地，还包括：失败提醒单元305。
特征匹配单元304，具体用于提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作，若否，则触发失败提醒单元305。
失败提醒单元305，用于声纹验证失败，提示操作用户输入支付密码。
进一步地，特征匹配单元304具体包括：
检测子单元3041，用于对实时音频数据进行端点检测，获取实时音频数据中的说话人音频数据。
特征子单元3042，用于对说话人音频数据进行去噪处理，对去噪后的说话人音频数据进行特征提取，获取说话人音频数据中的实时声学特征。
匹配子单元3043，用于将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作，若否，则触发失败提醒单元305。
进一步地，还包括：
语音注册单元300，用于接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则触发提取存档单元301；
提取存档单元301，用于提取注册音频数据中的注册声学特征，将注册声学特征存入声纹数据库中作为预置声学特征。
进一步地，还包括：失败返回单元302。
语音注册单元300，具体用于接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则触发提取存档单元301，若否，则触发失败返回单元302。
失败返回单元302，用于声纹注册失败，提示用户重新输入注册音频数据并触发语音注册单元300。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围。

Claims (10)

  1. 一种电子设备声纹支付方法，其特征在于，包括：
    S1：接收支付指令，根据支付指令获取缓存区中预置时间段内的实时监听的实时音频数据；
    S2：提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作。
  2. 根据权利要求1所述的一种电子设备声纹支付方法，其特征在于，步骤S2之后还包括：步骤S3；
    S2具体包括：提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作，若否，则执行步骤S3；
    S3：声纹验证失败，提示操作用户输入支付密码。
  3. 根据权利要求2所述的一种电子设备声纹支付方法，其特征在于，步骤S2具体包括：
    S21：对实时音频数据进行端点检测，获取实时音频数据中的说话人音频数据；
    S22：对说话人音频数据进行去噪处理，对去噪后的说话人音频数据进行特征提取，获取说话人音频数据中的实时声学特征；
    S23：将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作，若否，则执行步骤S3。
  4. 根据权利要求1所述的一种电子设备声纹支付方法，其特征在于，步骤S1之前还包括：
    S01：接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则执行步骤S02；
    S02：提取注册音频数据中的注册声学特征，将注册声学特征存入声纹数据库中作为预置声学特征。
  5. 根据权利要求4所述的一种电子设备声纹支付方法，其特征在于，步骤S01之后，步骤S1之前还包括：步骤S03；
    S01具体包括：接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则执行步骤S02，若否，则执行步骤S03；
    S03：声纹注册失败，提示用户重新输入注册音频数据并返回步骤S01。
  6. 一种电子设备声纹支付装置，其特征在于，包括：
    数据获取单元，用于接收支付指令，根据支付指令获取缓存区中预置时间段内的实时监听的实时音频数据；
    特征匹配单元，用于提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作。
  7. 根据权利要求6所述的一种电子设备声纹支付装置，其特征在于，还包括：失败提醒单元；
    特征匹配单元，具体用于提取实时音频数据中的实时声学特征，将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数 是否大于预置阈值，若是，则执行支付操作，若否，则触发失败提醒单元；
    失败提醒单元，用于声纹验证失败，提示操作用户输入支付密码。
  8. 根据权利要求7所述的一种电子设备声纹支付装置，其特征在于，特征匹配单元具体包括：
    检测子单元，用于对实时音频数据进行端点检测，获取实时音频数据中的说话人音频数据；
    特征子单元，用于对说话人音频数据进行去噪处理，对去噪后的说话人音频数据进行特征提取，获取说话人音频数据中的实时声学特征；
    匹配子单元，用于将实时声学特征与声纹数据库中的预置声学特征进行匹配获取匹配分数，判断匹配分数是否大于预置阈值，若是，则执行支付操作，若否，则触发失败提醒单元。
  9. 根据权利要求6所述的一种电子设备声纹支付装置，其特征在于，还包括：
    语音注册单元，用于接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则触发提取存档单元；
    提取存档单元，用于提取注册音频数据中的注册声学特征，将注册声学特征存入声纹数据库中作为预置声学特征。
  10. 根据权利要求9所述的一种电子设备声纹支付装置，其特征在于，还包括：失败返回单元；
    语音注册单元，具体用于接收注册指令和注册音频数据，根据注册指令判断注册音频数据是否符合预置语音质量标准的要求，如是，则触发提取存档单元，若否，则触发失败返回单元；
    失败返回单元，用于声纹注册失败，提示用户重新输入注册音频数据并触发语音注册单元。
PCT/CN2018/079533 2017-12-26 2018-03-20 一种电子设备声纹支付方法及装置 WO2019127929A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711435244.2 2017-12-26
CN201711435244.2A CN107871236B (zh) 2017-12-26 2017-12-26 一种电子设备声纹支付方法及装置

Publications (1)

Publication Number Publication Date
WO2019127929A1 true WO2019127929A1 (zh) 2019-07-04

Family

ID=61756268

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/079533 WO2019127929A1 (zh) 2017-12-26 2018-03-20 一种电子设备声纹支付方法及装置

Country Status (2)

Country Link
CN (1) CN107871236B (zh)
WO (1) WO2019127929A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108921562A (zh) * 2018-05-24 2018-11-30 佛山市竣智文化传播股份有限公司 一种基于声纹识别的线上支付方法及其装置
CN109598514A (zh) * 2018-07-26 2019-04-09 南京物联传感技术有限公司 一种带有声纹保护的智能音箱
CN109146480A (zh) * 2018-08-23 2019-01-04 交通银行股份有限公司 一种支付方法、装置、电子设备及存储介质
CN109034828A (zh) * 2018-09-02 2018-12-18 珠海横琴现联盛科技发展有限公司 针对录音的声纹识别支付信息防伪方法
CN109118242B (zh) * 2018-09-05 2021-07-09 深圳正品创想科技有限公司 一种账户创建方法及其装置、电子设备
CN110491373A (zh) * 2019-08-19 2019-11-22 Oppo广东移动通信有限公司 模型训练方法、装置、存储介质及电子设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103325037A (zh) * 2013-06-06 2013-09-25 上海讯联数据服务有限公司 一种基于语音识别的移动支付安全验证方法
CN103679451A (zh) * 2012-09-18 2014-03-26 上海语联信息技术有限公司 基于特定人语音的实时在线支付技术及其应用
CN103679452A (zh) * 2013-06-20 2014-03-26 腾讯科技（深圳）有限公司 支付验证方法、装置及系统
CN105096121A (zh) * 2015-06-25 2015-11-25 百度在线网络技术（北京）有限公司 声纹认证方法和装置

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060222210A1 (en) * 2005-03-31 2006-10-05 Hitachi, Ltd. System, method and computer program product for determining whether to accept a subject for enrollment
US9444816B2 (en) * 2011-03-30 2016-09-13 Qualcomm Incorporated Continuous voice authentication for a mobile device
CN103353973B (zh) * 2013-06-17 2016-06-29 上海方付通商务服务有限公司 基于音频验证的银行业务交易认证方法及系统
CN103955825A (zh) * 2014-05-21 2014-07-30 深圳数字电视国家工程实验室股份有限公司 电子支付方法、装置及数字电视
CN104168270B (zh) * 2014-07-31 2016-01-13 腾讯科技（深圳）有限公司 身份验证方法、服务器、客户端及系统
CN105575391B (zh) * 2014-10-10 2020-04-03 阿里巴巴集团控股有限公司 声纹信息管理方法、装置以及身份认证方法、系统
CN105894283A (zh) * 2015-01-26 2016-08-24 中兴通讯股份有限公司 一种基于语音控制的移动支付方法和装置
CN105989267B (zh) * 2015-02-16 2021-02-02 中兴通讯股份有限公司 一种基于声纹识别的安全防护方法和装置
CN107454057A (zh) * 2017-06-28 2017-12-08 深圳市佰仟金融服务有限公司 信息处理方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103679451A (zh) * 2012-09-18 2014-03-26 上海语联信息技术有限公司 基于特定人语音的实时在线支付技术及其应用
CN103325037A (zh) * 2013-06-06 2013-09-25 上海讯联数据服务有限公司 一种基于语音识别的移动支付安全验证方法
CN103679452A (zh) * 2013-06-20 2014-03-26 腾讯科技（深圳）有限公司 支付验证方法、装置及系统
CN105096121A (zh) * 2015-06-25 2015-11-25 百度在线网络技术（北京）有限公司 声纹认证方法和装置

Also Published As

Publication number Publication date
CN107871236B (zh) 2021-05-07
CN107871236A (zh) 2018-04-03

Similar Documents

Publication Publication Date Title
WO2019127929A1 (zh) 一种电子设备声纹支付方法及装置
WO2019127897A1 (zh) 一种自学习声纹识别的更新方法和装置
WO2019090834A1 (zh) 一种基于声纹的快递柜取件方法及装置
CN107977776B (zh) 信息处理方法、装置、服务器及计算机可读存储介质
US10593334B2 (en) Method and apparatus for generating voiceprint information comprised of reference pieces each used for authentication
JP6096333B2 (ja) 決済を検証するための方法、装置、およびシステム
EP3185162A1 (en) Authentication method, terminal and computer storage medium based on voiceprint characteristic
CN105468950B (zh) 身份认证方法、装置、终端及服务器
US9728191B2 (en) Speaker verification methods and apparatus
WO2016015687A1 (zh) 声纹验证方法及装置
TW201525894A (zh) 一種支付帳號管理方法及其行動終端、伺服器與系統
US8903052B2 (en) Voice print tagging of interactive voice response sessions
CN105989267B (zh) 一种基于声纹识别的安全防护方法和装置
WO2020051971A1 (zh) 身份识别方法、装置、电子设备及计算机可读存储介质
WO2017059679A1 (zh) 一种帐号处理方法及装置
CN107533598B (zh) 应用程序的登录密码的输入方法、装置和终端
US20180158462A1 (en) Speaker identification
EP3846052A1 (en) Biometric authentication device, biometric authentication method, and recording medium
JP2007266944A (ja) 電話端末装置および発信者検証方法
CN116013324A (zh) 基于声纹识别的机器人语音控制权限管理方法
US20120330663A1 (en) Identity authentication system and method
CN110322888B (zh) 信用卡解锁方法、装置、设备及计算机可读存储介质
CN106330915A (zh) 语音验证处理方法及装置
WO2020024415A1 (zh) 声纹识别处理方法、装置、电子设备及存储介质
CN109428804B (zh) 一种账号管理方法及装置

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: 18894917

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 05.11.2020)

122 Ep: pct application non-entry in european phase

Ref document number: 18894917

Country of ref document: EP

Kind code of ref document: A1