Voiceprint information management method, device, system, computing device, and storage medium
By collecting multiple voice information from different dimensions in voiceprint recognition and enriching the voiceprint information database, the problem of low accuracy of voiceprint recognition in the prior art is solved, and the recognition accuracy and security are improved.
Patent Information
- Application Number
- CN201910244648.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-03-28
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2039-03-28
AI Technical Summary
The existing products based on voiceprint recognition have shortcomings in recognition accuracy, mainly because the same voiceprint information is used for identification, resulting in a decrease in recognition accuracy.
通过获取用户的多段语音信息,分别对应于不同语音录制策略,采集和存储多个声纹信息,从不同维度(如空间维度、场景维度、语音内容维度、设备类型维度)进行录制,以丰富用户的声纹信息库。
It improves the accuracy and stability of voiceprint recognition, enhances the service security based on voiceprints, and avoids the identification difference problem under the single voiceprint information collection mechanism.
Smart Images

Figure CN111833882B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of voiceprint recognition, and in particular to a voiceprint information management method, device, system, computing device, and storage medium. Background Art
[0002] Voiceprint is a sound wave spectrum that carries speech information and is displayed by an electroacoustic instrument. Different people produce different sound waves when they say the same words, and the corresponding sound wave spectrum, that is, the voiceprint information, is also different. Therefore, by comparing the voiceprint information, it is possible to determine whether the corresponding speakers are the same, that is, to achieve identity authentication based on voiceprint recognition. At present, identity authentication based on voiceprint recognition has been widely used in various scenarios to provide users with convenient and safe technical support.
[0003] However, some voiceprint recognition-based products currently on the market use the same voiceprint information for recognition, which greatly affects the accuracy of voiceprint recognition based on the voiceprint information.
[0004] Therefore, a voiceprint information management solution is needed to improve the accuracy of voiceprint recognition. Summary of the invention
[0005] The purpose of the present disclosure is to provide a voiceprint information management method, device and system to provide support for improving the accuracy of voiceprint recognition.
[0006] According to one aspect of the present disclosure, a voiceprint information management method is provided, comprising: obtaining at least two segments of voice information of a user, the at least two segments of voice information respectively corresponding to at least two voice recording strategies; based on the at least two segments of voice information, respectively obtaining at least two voiceprint information of the user; and storing the at least two voiceprint information in association with the user.
[0007] Optionally, the at least two voiceprint information are stored in association with the identity information of the user; or the at least two voiceprint information are stored in association with a voiceprint account created for the user.
[0008] Optionally, the at least two voice recording strategies are different in at least one dimension.
[0009] Optionally, the at least one dimension includes at least one of the following: a spatial dimension; a scene dimension; a voice content dimension; and a device type dimension.
[0010] Optionally, the spatial dimension includes: the azimuth of the user relative to the electronic device that collects the voice information; and / or the distance of the user relative to the electronic device that collects the voice information.
[0011] Optionally, the method may further include: storing, in association with the user, a device type of the electronic device that collects the voice information.
[0012] Optionally, the method may further include: sending the at least two voice recording strategies to the client device of the user and / or the electronic device that collects the voice information respectively.
[0013] Optionally, the method may further include: matching the voiceprint information obtained from the voice information with at least two pieces of voiceprint information that have been stored; and in case of a successful match, providing the user with a service corresponding to the voice information.
[0014] Optionally, when the voice information is used for voiceprint service, the voiceprint information obtained from the voice information is matched with at least two pieces of voiceprint information that have been stored.
[0015] According to another aspect of the present disclosure, a voiceprint information management method is also provided, including: obtaining voice information of a user; matching the voiceprint information obtained from the voice information with at least two voiceprint information that have been stored; and if the match is successful, providing the user with a service corresponding to the voice information.
[0016] Optionally, the at least two stored voiceprint information correspond to at least two voice recording strategies respectively, and the at least two voice recording strategies are different in at least one dimension.
[0017] According to another aspect of the present disclosure, a voiceprint information management method is provided, including: sending a voiceprint registration request; outputting a first voice recording strategy; in response to voice information corresponding to the first voice recording strategy being uploaded, outputting a second voice recording strategy to the user.
[0018] Optionally, the first voice recording strategy and the second voice recording strategy are acquired locally or from a server; and / or the first voice recording strategy and the second voice recording strategy are different in at least one dimension.
[0019] Optionally, the at least one dimension includes at least one of the following: a spatial dimension; a scene dimension; a voice content dimension; and a device type dimension.
[0020] Optionally, the spatial dimension includes: the azimuth of the user relative to the electronic device that collects the voice information; and / or the distance of the user relative to the electronic device that collects the voice information.
[0021] Optionally, the method is executed on at least one of the following devices or modules: an electronic device; a client device; and a voice module.
[0022] According to another aspect of the present disclosure, a voiceprint information management system is provided, comprising: an electronic device for collecting at least two segments of voice information of a user, wherein the at least two segments of voice information respectively correspond to at least two voice recording strategies; and a server for acquiring the at least two segments of voice information, and based on the at least two segments of voice information, respectively acquiring at least two voiceprint information of the user, and storing the at least two voiceprint information in association with the user.
[0023] Optionally, the system may further include a client device, which is associated with the electronic device, and the electronic device and / or the client device is used to send a voiceprint registration request to a server and output a voice recording strategy to a user.
[0024] Optionally, the at least two voiceprint information are stored in association with the identity information of the user; or the at least two voiceprint information are stored in association with a voiceprint account created for the user.
[0025] According to a fourth aspect of the present disclosure, a server for executing a voiceprint information management method is provided, comprising: a voice acquisition device for acquiring at least two segments of voice information of a user, wherein the at least two segments of voice information respectively correspond to at least two voice recording strategies; a voiceprint information extraction device for acquiring at least two voiceprint information of the user based on the at least two segments of voice information; and a voiceprint account management device for storing the at least two voiceprint information in association with the user.
[0026] Optionally, the at least two voiceprint information are stored in association with the identity information of the user; or the at least two voiceprint information are stored in association with a voiceprint account created for the user.
[0027] Optionally, the at least two voice recording strategies are different in at least one dimension.
[0028] Optionally, the at least one dimension includes at least one of the following: a spatial dimension; a scene dimension; a voice content dimension; and a device type dimension.
[0029] Optionally, the spatial dimension includes: the azimuth of the user relative to the electronic device that collects the voice information; and / or the distance of the user relative to the electronic device that collects the voice information.
[0030] According to another aspect of the present disclosure, a voiceprint information management device is provided, comprising: a voice acquisition device for acquiring voice information of a user; a voiceprint matching device for matching the voiceprint information acquired from the voice information with at least two voiceprint information that have been stored; and a service device for providing the user with a service corresponding to the voice information when the match is successful.
[0031] According to another aspect of the present disclosure, there is provided a device for executing a voiceprint information management method, comprising: a voiceprint registration device for sending a voiceprint registration request; a first recording strategy output device for outputting a first voice recording strategy; and a second recording strategy output device for outputting a second voice recording strategy to the user in response to voice information corresponding to the first voice recording strategy being uploaded.
[0032] Optionally, the device is at least one of: an electronic device; a client device; and a voice module.
[0033] According to another aspect of the present disclosure, a computing device is provided, including: a processor; and a memory on which executable codes are stored, and when the executable codes are executed by the processor, the processor is caused to execute the method described above.
[0034] According to another aspect of the present disclosure, a non-transitory machine-readable storage medium is provided, on which executable codes are stored. When the executable codes are executed by a processor of an electronic device, the processor is caused to execute the method as described above.
[0035] The technical solution disclosed in the present invention enriches the user's voiceprint information library by performing voice recording and voiceprint information collection from different dimensions, thereby providing support for improving the stability and accuracy of voiceprint services based on voiceprints. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] The above and other objects, features and advantages of the present disclosure will become more apparent through a more detailed description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components in the exemplary embodiments of the present disclosure.
[0037] Figure 1 A schematic diagram of a voiceprint information management system according to an embodiment of the present disclosure is shown.
[0038] Figure 2 A schematic diagram of the process of a voiceprint information management method according to an embodiment of the present disclosure is shown.
[0039] Figure 3 A schematic diagram of the process of a voiceprint information management method according to an embodiment of the present disclosure is shown.
[0040] Figures 4A-4C A schematic diagram of a voiceprint registration process according to an embodiment of the present disclosure is shown.
[0041] Figure 5 A schematic block diagram of a server according to an embodiment of the present disclosure is shown.
[0042] Figure 6 A schematic block diagram of a client device according to an embodiment of the present disclosure is shown.
[0043] Figure 7 A schematic diagram of the structure of a computing device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0044] The preferred embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.
[0045] Figure 1 A schematic diagram of a voiceprint information management system according to an embodiment of the present disclosure is shown.
[0046] In one embodiment, the voiceprint information management system 100 of the present disclosure may include at least one server 130 , a plurality of client devices 110 , and a plurality of electronic devices 120 .
[0047] like Figure 1 As shown, the client device 110 and / or the electronic device 120 can transmit and receive information with the server 130 via the network. The server 130 can obtain the content required by the client device and / or the electronic device by accessing a database, for example. The connection between the client device 110 and the electronic device 120 indicates that there can be information exchange between the two, and the connection can be a wired connection, a wireless connection, or any form of connection that can transmit information.
[0048] Therein, the client device 110 is any suitable portable electronic device that can be used for network access, including but not limited to smart phones, tablet computers or other portable clients. Electronic device 120 is an electronic device that can be used for network access and has voice interaction function, including but not limited to smart speakers, smart home appliances, etc. Server 130 is any server that can provide information required for interactive services through network access. The network can be a network for information transmission in a broad sense, and can include one or more communication networks, such as wireless communication networks, the Internet, private domain networks, local area networks, metropolitan area networks, wide area networks or cellular data networks. It should be noted that if additional modules are added to the illustrated environment or individual modules are removed therefrom, the underlying concepts of the exemplary embodiments of the present invention will not be changed.
[0049] In the following description, one or some of the client devices or electronic devices will be selected for description, but those skilled in the art should understand that the above-mentioned multiple client devices or electronic devices are intended to represent a large number of terminals existing in a real network, and the single server shown is intended to represent the operation of the server involved in the technical solution of the present invention. The detailed description of a single client device or a single electronic device and a single server is at least for the convenience of explanation, and does not imply any limitation on the type or location of the mobile terminal and the server.
[0050] In one embodiment, the user can send a voiceprint registration request to the server via an electronic device 120 (e.g., a smart speaker), and / or output a voice recording strategy from the server to the user, and then collect at least two segments of voice information of the user in response to the voice recording strategy via the electronic device, and upload the collected at least two ends of voice information to the server, wherein the at least two segments of voice information correspond to at least two voice recording strategies respectively. The server 130 can obtain at least two segments of voice information of the user, and based on the at least two segments of voice information, respectively obtain at least two voiceprint information of the user, and store the at least two voiceprint information in association with the user.
[0051] In another embodiment, the electronic device 120 can cooperate with the above-mentioned client device 110 to implement the above-mentioned processing. For example, a user can issue a voiceprint registration request via the client device 110 (such as a smart phone), and output a voice recording strategy to the user. The voice recording strategy can be stored locally on the client device or obtained from a server. The user can record voice information based on the voice recording strategy output by the client device 110. The electronic device 120 can collect at least two segments of voice information of the user in response to the voice recording strategy, and the at least two segments of voice information correspond to at least two voice recording strategies respectively. The server 130 can obtain at least two segments of voice information of the user, and based on the at least two segments of voice information, respectively obtain at least two voiceprint information of the user, and store the at least two voiceprint information in association with the user.
[0052] The at least two voice recording strategies mentioned above may be different in at least one dimension. The at least one dimension may include, but is not limited to, the spatial dimension, scene dimension, voice content dimension, device type dimension, etc. shown in Table 1 below. Different dimensions may also include sub-dimensions or dimensional factors accordingly. For example, the spatial dimension may include the azimuth, distance, etc. of the user relative to the electronic device that collects voice information; for example, the scene dimension may include a quiet environment scene, an environment scene with background noise, etc.; for example, the voice content dimension may include text, numbers, tones, or any combination thereof; for example, the device type dimension may include the device type of the device that collects voice information and / or the module type of the voice module configured by the device, etc. It should be understood that the above dimensions and / or dimensional factors are only illustrative examples of different dimensions of the present disclosure, and are not any limitations on the dimensions and / or dimensional factors of the present disclosure.
[0053] Table 1
[0054]
[0055]
[0056] The voiceprint information management solution disclosed in the present invention will be described in detail below with reference to the accompanying drawings and embodiments.
[0057] Figure 2 A schematic diagram of the process of a voiceprint information management method according to an embodiment of the present disclosure is shown. Figure 5 A schematic block diagram of a server according to an embodiment of the present disclosure is shown. Figure 5 The server 500 shown can be used to implement Figure 2 The voiceprint information management method shown in the figure can be used in the stage of user voiceprint registration (i.e. extracting the user's voiceprint information and associating it with the user). It should be understood that Figure 2 The voiceprint information management method may also be executed locally on the device side, and the present disclosure does not impose any limitation on this.
[0058] In the case where a user needs to register a voiceprint, the user can request voiceprint registration through interaction with the electronic device (e.g., voice interaction), and the electronic device can send a voiceprint registration request to the server in response to the interaction. In response to the voiceprint registration request, the server can provide a voice recording strategy to the user's electronic device to provide support and / or guidance for the user to record voice information.
[0059] In one embodiment, the electronic device may output (for example, in the form of voice output) a voice recording strategy to the user to guide the user to record the voice information required for voiceprint registration according to the voice recording strategy, and upload the recorded voice information to the server.
[0060] As mentioned above, due to the different device types of electronic devices and / or the environments and / or scenarios in which voice recording is performed, differences may occur in the voiceprint information extracted from the collected voice information, thereby affecting the accuracy of voiceprint services such as recognition and authentication based on the voiceprint information.
[0061] Therefore, in the embodiment of the present disclosure, in the voiceprint registration stage, for example, the entire voiceprint registration process can be completed by recording voice in multiple rounds in a segmented manner. Among them, at least two segments of voice information of the user can be collected in segments, each segment of voice information can be recorded based on a voice recording strategy, and the at least two segments of voice information can correspond to at least two voice recording strategies respectively.
[0062] See also Figure 2 In step S210, for example, Figure 5 The voice acquisition device 510 shown acquires at least two segments of voice information of the user.
[0063] The voice message and the first / second / third / fourth voice message described below may be composed of Figure 1 The electronic device shown may have an audio collection / recording function and may collect / record the user's voice information. In addition, the electronic device may upload the collected user's voice information to a server so that the server can obtain the user's voiceprint information from the received voice information. It should be understood that the descriptions of "first", "second", "third" and "fourth" mentioned in this article are intended to distinguish the described objects, rather than to have any explicit or implicit provisions on their order and size.
[0064] Then, in step S220, for example, Figure 5 The voiceprint information extraction device 520 shown obtains at least two voiceprint information of the user based on the at least two pieces of voice information.
[0065] In step S230, for example, Figure 5 The voiceprint account management device 530 shown stores the at least two voiceprint information in association with the user.
[0066] The storage of user voiceprint information can be implemented in a variety of ways. For example, the at least two voiceprint information can be stored in association with the user's identity information, or a voiceprint account can be created for the user and the at least two voiceprint information of the user can be stored in association with the voiceprint account created for the user. The specific way of storing voiceprint information can be determined according to the specific application scenario, and the present disclosure does not limit this.
[0067] Therefore, by collecting multiple voice information based on different voice recording strategies to obtain multiple voiceprint information of the user, the user's voiceprint information database is further enriched to provide support for improving the accuracy and security of voiceprint services such as voiceprint-based recognition, authentication, and payment.
[0068] In one embodiment, the server may send the at least two voice recording strategies to the electronic device of the user respectively, so as to collect at least two segments of voice information based on segmented recording.
[0069] Specifically, for example, a first voice recording strategy may be output to the user first, and in response to the voice information recorded by the electronic device being uploaded, a second voice recording strategy may be output to the user.
[0070] Thus, at least two segments of voice information collected in segments can be uploaded to the server, and at least two voiceprint information obtained from the at least two segments of voice information can also be stored in association with the user to enrich the user's voiceprint information library.
[0071] In order to fully consider various application scenarios, multiple dimensions can be preset to facilitate the formulation of the above-mentioned voice recording strategy, and the at least two voice recording strategies are different in at least one dimension. Among them, the at least one dimension can include, for example, a spatial dimension, a scene dimension, a voice content dimension, a device type dimension, etc.
[0072] As mentioned above, different dimensions can correspond to their respective sub-dimensions or dimensional factors. For example, the spatial dimension can correspond to spatial factors such as the distance and azimuth angle between the user and the electronic device that collects voice information, the scene dimension can correspond to scene factors such as quiet scenes and scenes with background noise (such as watching TV, running on a treadmill, cooking in the kitchen, playing music, etc.), the voice content dimension can correspond to voice content factors of different voice contents (including language, text, numbers, tones or any combination thereof, etc.), and the device type dimension can correspond to the device type factor of the device that collects voice information and / or the module type of the voice module configured by the device, etc.
[0073] It should be understood that the above dimensions and / or dimensional factors are only illustrative examples of different dimensions of the present disclosure, and are not any limitations on the dimensions and / or dimensional factors of the present disclosure. The dimensions and / or dimensional factors of the present disclosure may also be other dimensions and their corresponding dimensional factors. In addition, the voice recording strategy may correspond to any single dimension of the preset multiple dimensions, or may correspond to at least two or more of the preset multiple dimensions, and different voice recording strategies may correspond to random combinations of at least two or more dimensions. The present disclosure does not impose any restrictions on this.
[0074] In the embodiment of the present disclosure, the at least two voice recording strategies mentioned above may be pre-formulated and stored, or may be generated in real time in response to a user's voiceprint registration request, and the at least two voice recording strategies may be obtained locally or from a server.
[0075] In one embodiment, at least two voice recording strategies can be pre-formulated and stored locally in the electronic device or in a server or other device accessible to the server. When a user initiates a voiceprint registration request via the electronic device, the electronic device can obtain at least two voice recording strategies from the local device or the server and output them to the user to guide the user to perform voice recording respectively.
[0076] In one embodiment, at least two voice recording strategies can also be generated in real time by the server in response to the voiceprint registration request of the electronic device, and sent to the electronic device for output to the user, thereby guiding the user to perform voice recording respectively. When the user initiates a voiceprint registration request via the electronic device, the electronic device can collect information about the current scene (such as background noise information, etc.) in real time and report it to the server, so that the server can formulate a voice recording strategy in real time according to the current scene of the electronic device, and send it to the electronic device for real-time output to the user.
[0077] Therefore, in the embodiments of the present disclosure, voice recording can be performed based on at least two pre-established and stored voice recording strategies, or a personalized voice recording strategy can be generated in real time according to the scenario or environment in which the user and his or her electronic device are located to perform voice recording, thereby providing multiple possibilities for obtaining voiceprint information, thereby providing support and guarantee for the accuracy, security, etc. of voiceprint services based on voiceprints.
[0078] In some application scenarios (such as smart home), multiple voiceprint information can be stored in association with the user (such as user identity information), and different voiceprint information may be collected via different electronic devices. To facilitate the management of the user's voiceprint information, these multiple voiceprint information can be associated and collected and stored in categories based on a certain dimension (such as the device type dimension of the electronic device that performs voice recording). In one embodiment, a voiceprint account can also be created for the user, and the voiceprint account can correspond to the voiceprint information obtained based on multiple different types of electronic devices.
[0079] Specifically, if the user's voiceprint account has not been registered, in response to the voiceprint registration request, the server can create a voiceprint account for the user, and can store the device type (such as device identification) of the electronic device and the corresponding collected voiceprint information in association with the voiceprint account created for the user. If the user's voiceprint account has been created, in response to the voiceprint registration request, the server can create a sub-voiceprint account corresponding to the electronic device that initiated the voiceprint registration request in association with the voiceprint account, and store the voiceprint information obtained based on the electronic device in association with the sub-voiceprint account.
[0080] Therefore, by providing users with a voiceprint account, it is possible to manage multiple voiceprint information collected through different electronic devices, thereby avoiding to a certain extent the differences in user voiceprint information caused by different structural types of voice devices having different appearances, microphone arrays, and voice processing capabilities under a single voiceprint information collection mechanism, thereby providing support for improving the accuracy and security of voiceprint-based services.
[0081] After the user completes voiceprint registration, the voiceprint information stored in association with the user can be used for some specific voiceprint services, such as voiceprint lock, identity authentication, voiceprint payment and other services.
[0082] Specifically, after obtaining the user's voice information, the voice information can be first analyzed and identified. When it is determined that the voice information is used for voiceprint registration, the user's voiceprint information can be extracted and stored based on the above-mentioned solution. When it is determined that the voice information is used for voiceprint service, the voiceprint information obtained from the voice information can be matched one by one with at least two stored voiceprint information. The voiceprint information extracted from the voice information can be considered to be a successful match if it matches any of the stored voiceprint information. When the match is successful, the service corresponding to the voice information can be provided to the user. In the case of a failed match, the user can be guided to register for voiceprint by providing at least two voice recording strategies to the user as described above, which will not be repeated here.
[0083] Therefore, the present invention enriches the user's voiceprint information library by collecting at least two voice information segments during the voiceprint registration stage, thereby providing support and guarantee for the accuracy and security of the subsequent voiceprint service based on the voiceprint information, thereby providing users with a better voiceprint experience.
[0084] In another embodiment, the voiceprint information management solution disclosed above can also be implemented by Figure 1 The electronic device, client device, and server shown are implemented in collaboration.
[0085] The electronic device may be associated with a client device. A related application client may be installed or run on the client device, and the user may, for example, log in to the application on the client device side and manage his voiceprint. The electronic device may have an audio recording function and a network communication function. In one embodiment, the user may initiate a voiceprint registration request to the server through interaction with the client device, and based on the voice recording strategy output by the client device side, realize the recording of voice information through voice interaction with the electronic device.
[0086] Figure 3 A schematic diagram of the process of a voiceprint information management method according to an embodiment of the present disclosure is shown. Figure 6 A schematic block diagram of a voiceprint information management device according to an embodiment of the present disclosure is shown. Figure 3 The method shown can be found in Figure 6 In one embodiment, the voiceprint information management device may be Figure 1 The electronic device and / or client device may also be a predetermined voice module, and the voice module may be set on any terminal device that can access the network, and the present disclosure does not impose any restrictions on this.
[0087] See also Figure 3 In step S310, for example, Figure 6 The voiceprint registration device 610 shown sends a voiceprint registration request to the server.
[0088] In step S320, for example, Figure 6 The first recording strategy output device 620 shown outputs the first voice recording strategy to the user.
[0089] The first voice recording strategy can be stored locally in the voiceprint information management device or received from a server. The voiceprint information management device can present the first voice recording strategy to the user based on a display screen, or can play the first voice recording strategy to the user in the form of voice broadcast. The specific output method can be determined based on the implementation form of the voiceprint information management device, and the present disclosure does not limit this.
[0090] According to the output first voice recording strategy, the user can perform voice interaction with the electronic device to achieve the recording of voice information. The electronic device can upload the recorded voice information to the server. In one embodiment, when the voiceprint information management device is Figure 1 When the client device is shown, the electronic device may be an electronic device associated with the client device, and the client device and the electronic device based on the association may collaboratively implement the voiceprint information management solution of the present disclosure.
[0091] Then, in step S330, for example, Figure 6 The second recording strategy output device 630 shown outputs the second voice recording strategy to the user in response to the voice information corresponding to the first voice recording strategy being uploaded.
[0092] The second voice recording strategy is the same as or similar to the first voice recording strategy. The second voice recording strategy can also be stored locally in the voiceprint information management device or obtained from a server. The voiceprint information management device can present the second voice recording strategy to the user based on a display screen, or can play the second voice recording strategy to the user in the form of voice announcement.
[0093] In one embodiment, the first voice recording strategy and the second voice recording strategy may be different in at least one dimension. The at least one dimension includes at least one of the following: space dimension; scene dimension; voice content dimension; and device type dimension. The dimension or its related content can refer to the relevant description above and will not be repeated here.
[0094] In order to more clearly understand the implementation process of the voiceprint information management solution disclosed in this disclosure, Figures 4A-4C A schematic diagram of the voiceprint registration process according to an embodiment of the present disclosure is shown for schematic illustration. The voiceprint registration process is implemented by the user, the client device, the electronic device and the server in collaboration, and the entire voiceprint registration process can be completed in stages. The voiceprint information collected in the stages after the first stage of voiceprint registration can be used as a supplement to the voiceprint information in the first stage of voiceprint registration to enhance the user's voiceprint information. It should be understood that Figures 4A-4C The flowchart shown is merely an exemplary description of the voiceprint information management solution of the present disclosure and is not intended to be limiting.
[0095] See also Figure 4A In step S401, the user may enter his / her identity information, the voiceprint name of the voiceprint account to be created, etc. on the voiceprint registration interface presented on the client device.
[0096] Afterwards, in step S402, the client device may initiate a voiceprint registration request in response to the user's voiceprint registration operation, so as to request voiceprint registration from the server.
[0097] In step S403, the client device receives a first voice recording strategy sent by the server, which may include, for example, "keep the environment quiet", "distance 1 from the electronic device", "azimuth 1 relative to the electronic device", "voice content 1", etc.
[0098] In step S404, the user adjusts the voice collection environment to a quiet scene according to the first voice recording strategy, adjusts the relative position of the user and the electronic device that collects voice information (for example, the azimuth 1 and distance 1 of the user relative to the electronic device), and sends a first voice to the electronic device to record voice content 1. After that, the electronic device sends the recorded first voice information to the server.
[0099] In step S405, the server parses the received first voice information, and extracts voiceprint information based on the received first voice information when the content of the first voice information matches the voice content 1 sent to the client device. If the content of the first voice information does not match the voice content 1 sent to the client device, the server returns to step S403 and reminds the user to record the voice content 1 again according to the first voice recording strategy until the first voice information matching the voice content 1 can be collected.
[0100] In step S406, it is determined whether the voiceprint information is extracted successfully. If the voiceprint information is extracted successfully, the process proceeds to step S407, where the extracted voiceprint information is associated with the user (eg, with the user identity information or with the voiceprint account created for the user, etc., the same below) and stored, completing the first stage of voiceprint registration.
[0101] In the case where the voiceprint information extraction fails, the process returns to step S403 and reminds the user to record the voice content 1 again according to the first voice recording strategy until the voiceprint information extraction succeeds.
[0102] After completing the first stage of voiceprint registration, the client device can also provide users with "voiceprint enhancement" services. For example, it can provide users with "voiceprint enhancement from the spatial dimension", "voiceprint enhancement from the scene dimension", "voiceprint enhancement from the device type dimension", etc., so that by recording multiple first voices of the same user other than the first stage of voiceprint registration, more voiceprint information can be extracted as a supplement, thereby enhancing the user's voiceprint information.
[0103] In one embodiment, see Figure 4B In step S408, the user may, for example, select "Enhance voiceprint from spatial dimension" on the voiceprint registration interface provided to the user on the client device. In step S409, the client device may initiate a voiceprint enhancement request to the server to request enhancement of the user's voiceprint information. In other embodiments, after the user completes the first stage of voiceprint registration, the server may also directly send the second voice recording strategy to the user's client device and guide the user to complete the enhancement of his voiceprint information. The present disclosure does not limit the specific implementation method thereof.
[0104] In step S410, the server sends a second voice recording strategy to the client device, which may include, for example, "keep the environment quiet", "distance 2 from the electronic device", "azimuth 2 relative to the electronic device", "voice content 2", etc.
[0105] In step S411, the user adjusts the voice collection environment to a quiet scene according to the second voice recording strategy, and adjusts the relative position of the user and the electronic device (for example, the user's azimuth angle 2 and distance 2 relative to the electronic device), and records the voice content 2 by sending the first voice to the electronic device. After that, the electronic device sends the recorded second voice information to the server. In the case that the content of the second voice information does not match the voice content 2 sent to the client device, it returns to step S410 and reminds the user to record the voice content 2 again according to the second voice recording strategy until the second voice information matching the voice content 2 can be collected.
[0106] In step S412, the server parses the received second voice information, and extracts voiceprint information based on the received second voice information when the content of the second voice information matches the voice content 2 sent to the client device.
[0107] In step S413, it is determined whether the voiceprint information is extracted successfully. If the voiceprint information is extracted successfully, the process proceeds to step S414, where the extracted voiceprint information is associated with the user and stored, completing the second stage of voiceprint registration, namely, the voiceprint enhancement stage.
[0108] In the case where the voiceprint information extraction fails, the process returns to step S410 and reminds the user to record the voice content 2 again according to the second voice recording strategy until the voiceprint information extraction succeeds.
[0109] In another embodiment, see Figure 4C In step S415, the user may select "enhance voiceprint from scene dimension" on the voiceprint registration interface provided to the user on the client device.
[0110] In step S416, the client device initiates a voiceprint enhancement request to the server to request enhancement of the user's voiceprint information. In other embodiments, after the user completes the first stage of voiceprint registration and / or the second stage of voiceprint registration, the server may also directly send the third voice recording strategy to the user's client device and guide the user to complete the enhancement of his voiceprint information. The present disclosure does not limit the specific implementation method thereof.
[0111] In step S417, the server sends a third voice recording strategy to the client device, which voice recording may include, for example, "adding scene interference (for example, turning on the TV and adjusting the TV volume to the volume for daily use)", "distance 1 from the electronic device", "azimuth 1 relative to the electronic device", "voice content 1", etc.
[0112] In step S418, according to the third voice recording strategy, the user places the user and the electronic device that collects voice information in the scene indicated by the third voice recording strategy, such as adjusting the volume of the TV to the volume used in daily life, and adjusting the relative position of the user and the electronic device, such as the distance 1 between the user and the electronic device and the azimuth 1 relative to the electronic device, and sends the third voice to the electronic device to record the voice content 1. Afterwards, the electronic device sends the recorded third voice information to the server.
[0113] In step S419, the server parses the received third voice information, and extracts voiceprint information based on the received third voice information when the scene information (e.g., background noise) in the third voice information matches the scene interference in the third voice recording strategy, and the content in the third voice information matches the voice content 1 sent to the client device. In the case that the scene information (e.g., background noise) in the third voice information matches the scene interference in the third voice recording strategy, and / or the content in the third voice information does not match the voice content 1 sent to the client device, the server returns to step S417 and reminds the user to record the voice content 1 again according to the third voice recording strategy until the third voice information collected based on the scene interference of the third voice recording strategy and matching the voice content 1 can be collected.
[0114] In step S420, it is determined whether the voiceprint information is extracted successfully. If the voiceprint information is extracted successfully, the process proceeds to step S421, where the extracted voiceprint information is associated with the user and stored, completing the third stage of voiceprint registration, namely, the voiceprint enhancement stage.
[0115] In the case where the voiceprint information extraction fails, the process returns to step S417 and reminds the user to record the voice content 1 again according to the third voice recording strategy until the voiceprint information extraction succeeds.
[0116] Therefore, in the technical solution of the present invention, segmented voice recording can be performed from different dimensions to realize the user's voiceprint registration and enrich the user's voiceprint information database, thereby providing support and guarantee for the accuracy and stability of services based on user voiceprints.
[0117] It should be understood that the above is only an example of the present disclosure and not any limitation of the present disclosure. The above-mentioned spatial dimensions are not limited to fixed distances, azimuth angles and voice content, but can also be a random combination of these factors. The above-mentioned scene dimensions are not limited to the exemplified scenes of watching TV, but can also include scenes such as running and cooking, and can be a random combination of these multiple scenes. The above-mentioned voiceprint enhancement from the spatial dimension or scene dimension is not limited to voice recording in a single dimension, but can also be other dimensions, or can also be a random combination of spatial dimensions, scene dimensions or other dimensions. The present disclosure does not impose any restrictions on this.
[0118] Therefore, through the voiceprint information management solution disclosed above, in the voiceprint registration stage, voice collection is carried out by adopting a segmented multi-round recording method to store a rich voiceprint information library corresponding to different dimensions for users, thereby providing support and guarantee for the accuracy and security of the user's subsequent voiceprint service based on voiceprint information.
[0119] Figure 7 A schematic diagram of the structure of a computing device according to an embodiment of the present invention is shown.
[0120] See also Figure 7 , the computing device 700 includes a memory 710 and a processor 720 .
[0121] The processor 720 may be a multi-core processor or may include multiple processors. In some embodiments, the processor 720 may include a general-purpose main processor and one or more special coprocessors, such as a graphics processing unit (GPU), a digital signal processor (DSP), etc. In some embodiments, the processor 720 may be implemented using a customized circuit, such as an application-specific integrated circuit (ASIC) or a field programmable gate array (FPGA).
[0122] The memory 710 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. Among them, ROM can store static data or instructions required by the processor 720 or other modules of the computer. The permanent storage device may be a readable and writable storage device. The permanent storage device may be a non-volatile storage device that does not lose the stored instructions and data even after the computer is powered off. In some embodiments, the permanent storage device uses a large-capacity storage device (such as a magnetic or optical disk, flash memory) as a permanent storage device. In some other embodiments, the permanent storage device may be a removable storage device (such as a floppy disk, optical drive). The system memory may be a readable and writable storage device or a volatile readable and writable storage device, such as a dynamic random access memory. The system memory may store some or all instructions and data required by the processor at run time. In addition, the memory 710 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory 710 may include a readable and / or writable removable storage device, such as a laser disc (CD), a read-only digital versatile disc (e.g., DVD-ROM, double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (e.g., SD card, mini SD card, Micro-SD card, etc.), a magnetic floppy disk, etc. The computer-readable storage medium does not include carrier waves and transient electronic signals transmitted wirelessly or wired.
[0123] The memory 710 stores processable codes, and when the processable codes are processed by the processor 720, the processor 720 can execute the voiceprint information management method mentioned above.
[0124] The voiceprint information management method, device and system according to the present invention have been described above in detail with reference to the accompanying drawings.
[0125] In addition, the method according to the present invention may also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing the above steps defined in the above method of the present invention.
[0126] Alternatively, the present invention may also be implemented as a non-temporary machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) on which executable code (or computer program, or computer instruction code) is stored. When the executable code (or computer program, or computer instruction code) is executed by a processor of an electronic device (or computing device, server, etc.), the processor executes the various steps of the above-mentioned method according to the present invention.
[0127] Those skilled in the art will further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the disclosure herein may be implemented as electronic hardware, computer software, or a combination of both.
[0128] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system and method according to multiple embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or code, and the part of the module, program segment or code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0129] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.
Claims
1. A voiceprint information management method, characterized in that: include: Outputting at least two voice recording strategies to a user, wherein the at least two voice recording strategies are different in at least two dimensions; Acquire at least two voice messages recorded by the user according to the at least two voice recording strategies respectively; Based on the at least two pieces of voice information, respectively obtain at least two voiceprint information of the user; The at least two voiceprint information are stored in association with the user.
2. The method according to claim 1, characterized in that storing the at least two voiceprint information in association with the user's identity information; or The at least two voiceprint information are stored in association with the voiceprint account created for the user.
3. The method according to claim 1, characterized in that The at least two dimensions include at least two of the following: spatial dimension; Scene dimension; Speech content dimension; and Device type dimension.
4. The method according to claim 3, characterized in that: The spatial dimensions include: The azimuth angle of the user relative to the electronic device collecting the voice information; and / or The distance between the user and the electronic device that collects the voice information.
5. The method according to claim 1, characterized in that Also includes: The device type of the electronic device that collects the voice information is stored in association with the user.
6. The method according to claim 1, characterized in that Also includes: The at least two voice recording strategies are sent to the client device of the user and / or the electronic device that collects the voice information respectively.
7. The method according to claim 1, characterized in that Also includes: Matching the voiceprint information obtained from the acquired voice information with at least two stored voiceprint information; as well as In case of successful matching, the service corresponding to the voice information is provided to the user.
8. The method according to claim 7, characterized in that In the case where the voice information is used for voiceprint service, the voiceprint information obtained from the voice information is matched with at least two pieces of voiceprint information that have been stored.
9. A voiceprint information management method, characterized in that: include: Get the user's voice information; matching the voiceprint information obtained from the voice information with at least two pieces of voiceprint information that have been stored, the at least two pieces of voiceprint information being obtained based on at least two pieces of voice information recorded by the user according to at least two voice recording strategies output to the user, respectively, the at least two voice recording strategies being different in at least two dimensions; In case of successful matching, the service corresponding to the voice information is provided to the user.
10. A voiceprint information management method, characterized in that: include: Send voiceprint registration request; Outputting a first voice recording strategy; In response to voice information recorded by a user according to the first voice recording strategy being uploaded, a second voice recording strategy is output to the user, wherein the first voice recording strategy and the second voice recording strategy are different in at least two dimensions.
11. The method according to claim 10, characterized in that The first voice recording strategy and the second voice recording strategy are acquired locally or from a server.
12. The method according to claim 11, characterized in that The at least two dimensions include at least two of the following: spatial dimension; Scene dimension; Speech content dimension; and Device type dimension.
13. The method according to claim 12, characterized in that The spatial dimensions include: The azimuth angle of the user relative to the electronic device collecting the voice information; and / or The distance between the user and the electronic device that collects the voice information.
14. The method according to claim 12, characterized in that The method is performed on at least one of the following devices or modules: Electronic equipment; Client devices; and Voice module.
15. A voiceprint information management system, characterized in that: include: An electronic device, configured to output at least two voice recording strategies to a user, collect at least two segments of voice information recorded by the user according to the at least two voice recording strategies, and send the at least two segments of voice information to a server, wherein the at least two voice recording strategies are different in at least two dimensions; as well as The server obtains the at least two pieces of voice information, and based on the at least two pieces of voice information, respectively obtains at least two pieces of voiceprint information of the user, and stores the at least two pieces of voiceprint information in association with the user.
16. The system according to claim 15, characterized in that Also includes: a client device, associated with the electronic device, The electronic device and / or the client device is used to send a voiceprint registration request to a server and output a voice recording strategy to a user.
17. A server for executing a voiceprint information management method, characterized in that: include: A voice acquisition device, used to output at least two voice recording strategies to a user, and acquire at least two pieces of voice information recorded by the user according to the at least two voice recording strategies, respectively, wherein the at least two voice recording strategies are different in at least two dimensions; A voiceprint information extraction device, used to obtain at least two voiceprint information of the user based on the at least two pieces of voice information; as well as The voiceprint account management device is used to store the at least two voiceprint information in association with the user.
18. The server according to claim 17, characterized in that: storing the at least two voiceprint information in association with the user's identity information; or The at least two voiceprint information are stored in association with the voiceprint account created for the user.
19. The server according to claim 18, characterized in that The at least two dimensions include at least two of the following: spatial dimension; Scene dimension; Speech content dimension; and Device type dimension.
20. The server according to claim 19, characterized in that The spatial dimensions include: The azimuth angle of the user relative to the electronic device collecting the voice information; and / or The distance between the user and the electronic device that collects the voice information.
21. A voiceprint information management device, characterized in that: include: A voice acquisition device, used to acquire the user's voice information; A voiceprint matching device, configured to match the voiceprint information obtained from the voice information with at least two pieces of voiceprint information that have been stored, wherein the at least two pieces of voiceprint information are obtained based on at least two pieces of voice information recorded by the user according to at least two voice recording strategies output to the user, respectively, and the at least two voice recording strategies are different in at least two dimensions; The service device is used to provide the user with the service corresponding to the voice information when the matching is successful.
22. A voiceprint information management device, characterized in that: include: A voiceprint registration device, used to send a voiceprint registration request; A first recording strategy output device, used to output a first voice recording strategy; as well as The second recording strategy output device is used to output a second voice recording strategy to the user in response to the voice information recorded by the user according to the first voice recording strategy being uploaded, and the first voice recording strategy and the second voice recording strategy are different in at least two dimensions.
23. The device according to claim 22, characterized in that The device is at least one of the following: Electronic equipment; Client devices; and Voice module.
24. A computing device comprising: processor; as well as A memory having executable codes stored thereon, which, when executed by the processor, causes the processor to execute the method according to any one of claims 1 to 14.
25. A non-transitory machine-readable storage medium having executable codes stored thereon, which, when executed by a processor of an electronic device, causes the processor to execute the method according to any one of claims 1 to 14.
