WO2019114507A1 - 语音处理方法、装置、存储介质及终端设备 - Google Patents
语音处理方法、装置、存储介质及终端设备 Download PDFInfo
- Publication number
- WO2019114507A1 WO2019114507A1 PCT/CN2018/116587 CN2018116587W WO2019114507A1 WO 2019114507 A1 WO2019114507 A1 WO 2019114507A1 CN 2018116587 W CN2018116587 W CN 2018116587W WO 2019114507 A1 WO2019114507 A1 WO 2019114507A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- voice information
- verification
- information
- verified
- stored
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
Definitions
- the present disclosure relates to the field of computer speech processing technologies, for example, to a voice processing method, apparatus, storage medium, and terminal device.
- the verification method includes a voice-based identification method for verifying the identity of the user, that is, the terminal device acquires the voice spoken by the user, and determines whether the voice is the voice of the user to verify the identity of the user.
- the user's voice may be affected by self-factors or external factors. For example, if the user catches a cold, the user's cold sound will be heavier than the usual sound, and the terminal device may not recognize the user's cold. The accuracy of speech recognition is low.
- the present disclosure provides a voice processing method, apparatus, storage medium, and terminal device, which can improve the accuracy of voice verification.
- the present disclosure provides a voice processing method, including:
- the first verification fails, starting the second verification; wherein the second verification includes a verification mode other than voice verification;
- the first to-be-verified voice information passes the second verification, determining the first to-be-verified voice information as pre-stored voice information, and storing the pre-stored voice information to the identification library.
- the present disclosure also provides a voice processing device, including:
- a voice verification module configured to obtain the first to-be-verified voice information, and perform first verification on the first to-be-verified voice information
- Verifying the startup module configured to initiate a second verification when the first verification fails; wherein the second verification includes a verification mode other than voice verification;
- the verification module is configured to receive the verification information of the second verification, and determine, according to the verification information, whether the first to-be-verified voice information passes the second verification;
- the voice storage module is configured to determine the first to-be-verified voice information as pre-stored voice information when the first to-be-verified voice information passes the second verification, and store the pre-stored voice information to the identification library.
- the present disclosure also provides a computer readable storage medium having stored thereon a computer program that, when executed by a processor, implements a speech processing method as described in embodiments of the present disclosure.
- the present disclosure also provides a terminal device including a memory, a processor, and a computer program stored on the memory and operable by the processor, the processor executing the computer program to implement a voice as described in an embodiment of the present disclosure Approach.
- the voice processing solution provided by the present disclosure can determine whether the voice information failed to be verified is stored in the identification library by determining the user identity when the user's voice information verification fails, and the voice verification can be improved when the subsequent user uses the voice verification again. accuracy.
- FIG. 1 is a schematic flowchart of a voice processing method according to an embodiment
- FIG. 2 is a schematic flowchart of a voice processing method according to another embodiment
- FIG. 3 is a schematic flowchart of a voice processing method according to another embodiment
- FIG. 4 is a schematic flowchart of a voice processing method according to another embodiment
- FIG. 5 is a schematic flowchart of a voice processing method according to another embodiment
- FIG. 6 is a schematic flowchart diagram of a voice processing method according to another embodiment
- FIG. 7 is a structural block diagram of a voice processing apparatus according to an embodiment
- FIG. 8 is a schematic structural diagram of a terminal device according to an embodiment
- FIG. 9 is a schematic structural diagram of a terminal device according to another embodiment.
- Terminal devices can include smartphones, tablets, and other devices with operating systems.
- the unlocking mode of the terminal device may include a voice verification mode, and the opening verification of the application (Application, APP) on the terminal device may also include a voice verification mode, but if the user's voice changes due to a cold, the voice verification may not be performed. At this time, by verifying the identity of the user by other means, the voice of the cold can be used as the verification reference voice, which can improve the accuracy of the subsequent voice verification.
- the unlocking mode of the terminal device may include a voice verification mode
- the opening verification of the application (Application, APP) on the terminal device may also include a voice verification mode, but if the user's voice changes due to a cold, the voice verification may not be performed.
- the voice of the cold can be used as the verification reference voice, which can improve the accuracy of the subsequent voice verification.
- FIG. 1 is a schematic flowchart of a voice processing method according to an embodiment.
- the method may be implemented by a voice processing device.
- the device may be implemented by software and/or hardware, and may be integrated in a terminal device or integrated in other devices.
- the method includes the following steps.
- S1100 Acquire voice information to be verified, and perform first verification on the voice information to be verified.
- the voice information to be verified may be the voice information to be verified that is output by the user of the terminal device during voice verification, and the voice information to be verified is obtained by the terminal device.
- the voice verification may be a verification mode of the terminal device that is verified by the user to be verified by the voice information to be verified by the user.
- the voice information to be verified may be the voice information output by the user for unlocking the terminal device, or may be the voice information output by the user for unlocking the application on the terminal device.
- the application may be WeChat, and the user may judge whether the user passes the first verification according to the voice spoken by the user by speaking a preset text or number.
- the user can analyze and verify the voice information to be verified by outputting the voice information to be verified.
- the verification of the voice information may be performed by comparing the voice information to be verified with the preset voice information of the identification library, where the preset voice information may be the earliest use of the corresponding voice verification by the administrator of the terminal device.
- the voice information to be verified is compared with the standard voice information, and it can be determined whether the voice information to be verified matches the standard voice information, and then it can be determined whether the user is the administrator of the terminal device. .
- the second verification may include at least one of fingerprint verification, password verification, graphics verification, face recognition verification, and iris recognition verification. If the verification of the to-be-verified voice information fails, it is possible to verify whether the user is the administrator himself by initiating the second verification.
- the authentication mode on the terminal device includes multiple modes, and the manner of the second verification may be determined according to the user setting or the system preset. This embodiment is not limited herein.
- S1120 Receive verification information of the second verification, and determine, according to the verification information, whether the to-be-verified voice information passes the second verification.
- the verification information of the second verification is verification information input by the user according to the initiated second verification.
- the verification information of the second verification is fingerprint information of the user; if the second verification is password verification, the verification information of the second verification is password information input by the user. And determining, according to the verification information of the second verification, whether the to-be-verified voice information passes the second verification, whether the user is the administrator himself or not.
- S1130 If the to-be-verified voice information passes the second verification, determine the to-be-verified voice information as pre-stored voice information, and store the pre-stored voice information into the identification library.
- the voice information to be verified passes the second verification, it may be determined that the user is the administrator himself, and the corresponding voice information to be verified is also the voice information of the administrator himself, and the user may change the voice due to some self factors or external factors.
- the voice to be verified may be determined as pre-stored voice information, and the pre-stored voice information is stored in the identification library. If the next time the user changes the sound due to the same self factor or external factors, the output voice information to be verified can be verified.
- the voice information to be verified of the cold may be determined as the pre-stored voice information, and the pre-stored voice information may be stored in the identification library. If the user has a cold again, the identification library is stored. When there is pre-stored voice information when the user catches a cold, the voice information to be verified that the user has a cold can also pass the verification. Therefore, the accuracy of the subsequent voice verification can be improved by determining the voice information to be verified as pre-stored voice information and storing it in the identification library.
- the voice information of the administrator's own friends and relatives may be verified by the technical solution of the embodiment. It is determined to pre-store voice information and store the pre-stored voice information to the identification library.
- the voice information to be verified of another person for example, a relative friend
- the voice information to be verified of another person may be obtained, and because the voice information to be verified of another person is different from the standard voice information of the user, the first verification will definitely fail the verification, and the second verification is initiated accordingly.
- the user (administrator himself/herself) can input the verification information of the second verification.
- the other person's to-be-verified voice information can be stored as a pre-stored voice information in the identification library, and subsequent users can pass their own
- the sound is verified by the voice of the application on the terminal device or the terminal device.
- the use efficiency of the voice recognition can be improved, and the function of the voice verification can also be expanded.
- FIG. 2 is a schematic flowchart diagram of a voice processing method according to another embodiment. As shown in FIG. 2, the method includes the following steps.
- S1200 Acquire a first to-be-verified voice information, and perform first verification on the first to-be-verified voice information.
- S1210 If the first verification fails, start the second verification; wherein the second verification includes a verification mode other than voice verification.
- S1220 Receive verification information of the second verification, and determine, according to the verification information, whether the first to-be-verified voice information passes the second verification.
- the first to-be-verified voice information passes the second verification, determine the first to-be-verified voice information as pre-stored voice information, and store the pre-stored voice information to the identification library.
- S1240 Acquire a second to-be-verified voice information, and verify the second to-be-verified voice information according to the pre-stored voice information in the identification library, where the pre-stored voice information is at least one.
- the second to-be-verified voice information may be voice information to be verified that is output by the user when the voice verification is used later.
- the second to-be-verified voice information is verified according to the pre-stored voice information in the identification library, and the pre-stored voice information may be the first to-be-verified voice information determined by the foregoing embodiment as the pre-stored voice information, or may be identification.
- Other pre-stored voice information stored in the library may be standard voice information recorded by the administrator of the terminal device when the voice verification is first used or initiated. Therefore, the pre-stored voice information in the identification library is at least one, that is, at least the standard voice information recorded by the administrator when the voice verification is first used or started. If the pre-stored voice information is more than one, the first to-be-verified voice information determined to be pre-stored voice information described in the foregoing embodiment may be included.
- the pre-stored voice information in the identification library may also include only the first to-be-verified voice information determined as the pre-stored voice information in the foregoing embodiment; and when the administrator himself uses or initiates voice verification at the earliest,
- the recorded standard voice information can be stored in other storage modules.
- the identification library as long as the first to-be-verified voice information determined as the pre-stored voice information described in the foregoing embodiment is included, the comparison material for voice verification can be added, and the accuracy of the voice verification is improved.
- the verifying the second to-be-verified voice information according to the pre-stored voice information in the identification library may be implemented by:
- S12400 Align the second to-be-verified voice information with the pre-stored voice information.
- the comparing the second to-be-verified voice information with the pre-stored voice information may compare the feature information of the second to-be-verified voice information with the feature information of the pre-stored voice information, where the preset condition may be The comparison error value is less than the set error value. If the error value between the feature information of the second to-be-verified voice information and the feature information of the pre-stored voice information is less than the set error value, it may be determined that the two voice information of the second to-be-verified voice information and the pre-stored voice information are successfully matched, Determining that the second to-be-verified voice information passes verification.
- the comparison may be determined as long as the comparison result of any one of the pre-stored voice information and the second to-be-verified voice information in the library meets a preset condition.
- the second to be verified voice information is verified.
- the identification library must include the first voice information to be verified that is determined to be pre-stored voice information, and may also include the administrator himself when using or starting voice verification at the earliest. The standard voice information recorded.
- the user who outputs the second voice information to be verified may be the administrator of the terminal device, and the user may perform voice verification.
- FIG. 3 is a schematic flowchart of a voice processing method according to another embodiment. As shown in FIG. 3, the method includes the following steps.
- S1300 Acquire voice information to be verified, and perform first verification on the voice information to be verified.
- S1320 Receive verification information of the second verification, and determine, according to the verification information, whether the to-be-verified voice information passes the second verification.
- S1340 Perform pre-processing on the pre-stored voice information to obtain a feature parameter of the pre-stored voice information, where the feature parameter includes a feature parameter that reflects a voice trait.
- the pre-processing process may include performing framing processing on the pre-stored voice information to obtain at least one voice frame.
- the frame length of each speech frame may be any value from 20 milliseconds (ms) to 50 ms. Since the voice information is an unstable signal, the signal changes in the voice information are generally very fast, so the voice information is subjected to frame processing, and the obtained voice frame has a sufficiently long period, and the signal in the voice frame does not change drastically. Can improve the efficiency of subsequent processing.
- the characteristic parameters may include parameters such as a centroid, a root mean square, and a Mel Frequency Cepstrum Coefficient.
- the centroid can reflect the basic frequency band of the signal in the speech frame.
- the root mean square can reflect the signal strength of the speech frame.
- the Mel cepstrum coefficient can simulate the human ear's perception of different frequency speech, which can effectively represent the voice characteristics of human speech.
- the characteristic parameters may also include characteristic parameters such as acoustic energy values, fundamental frequency and resonance peaks that may reflect sound characteristics.
- S1350 Store the feature parameters of the pre-stored voice information to the identification library.
- the size of voice information is generally large, the size of voice information for a period of 10 seconds (s) is generally more than ten kilobytes (Kilobyte, KB). If the stored voice information is directly stored in the identification library, it may lead to occupation. Multiple storage space.
- the pre-stored voice information is generally calculated in the background, and the user does not need to hear the pre-stored voice information again, so the feature parameters of the pre-stored voice information can be stored in the identification library, so that the voice information can be preserved in the voice information.
- the feature parameters enable subsequent calculation processing and occupy a small storage space.
- FIG. 4 is a schematic flowchart diagram of a voice processing method according to another embodiment. As shown in FIG. 4, the method includes the following steps.
- S1400 Acquire a first to-be-verified voice information, and perform first verification on the first to-be-verified voice information.
- S1420 Receive verification information of the second verification, and determine, according to the verification information, whether the first to-be-verified voice information passes the second verification.
- S1440 Perform pre-processing on the pre-stored voice information to obtain a feature parameter of the pre-stored voice information, where the feature parameter includes a feature parameter that reflects a voice trait.
- S1450 Store the feature parameters of the pre-stored voice information to the identification library.
- S1460 Acquire a second to-be-verified voice information, and perform pre-processing on the second to-be-verified voice information to obtain a feature parameter of the second to-be-verified voice information.
- the feature parameter may include a feature parameter that reflects a sound trait. Parameters such as centroid, root mean square, and Mel Frequency Cepstrum Coefficient may be included. It may also include characteristic parameters such as acoustic energy values, pitch frequency and resonance peaks that can reflect sound characteristics. For details, refer to the related description above, and details are not described herein again.
- the Euclidean Distance refers to the true distance between two points in an N-dimensional space or the natural length of a vector formed by two points, and N may be any natural number greater than 0.
- the feature parameters include a centroid, a root mean square, and a Mel frequency cepstral coefficient
- the second to-be-verified speech information is calculated in a three-dimensional space formed by the centroid, the root mean square, and the Mel frequency cepstral coefficient.
- the natural length of the vector formed by (a1, b1, c1) and the pre-stored speech information (a2, b2, c2) is the Euclidean distance; where a1 and a2 are centroids, b1 and b2 are root mean squares, c1 And c2 is the Mel frequency cepstral coefficient. If the Euclidean distance of the feature parameter of the pre-stored voice information and the feature parameter of the second to-be-verified voice information is less than a preset threshold, it may be determined that the pre-stored voice information and the second to-be-verified voice information are relatively close and match each other, and the The second to-be-verified voice information is verified.
- FIG. 5 is a schematic flowchart of a voice processing method according to another embodiment. As shown in FIG. 5, the method includes the following steps.
- S1500 Acquire a first to-be-verified voice information, and perform first verification on the first to-be-verified voice information.
- S1520 Receive verification information of the second verification, and determine, according to the verification information, whether the first to-be-verified voice information passes the second verification.
- the failure category information may be a plurality of category information pre-stored by the system, and the failure category information corresponding to the first to-be-verified voice information may be sent to the user by using a plurality of category information pre-stored by the system, and according to The user's selection determines that at least one of the plurality of category information is the failure category information.
- the failure category information can also be a custom content entered by the user.
- the plurality of category information pre-stored by the system may include a cold, a hoarse voice, and a throat irritation, etc., and the user may also input the customized content “cold”, “sound dumb”, and “friends and relatives” as the failure category. information.
- S1540 Determine the first to-be-verified voice information as pre-stored voice information, and store the pre-stored voice information and the failure category information to the identification library.
- the verification information of the second verification input by the user passes the second verification, it may be determined that the first to-be-verified voice information that is failed to be verified is managed.
- the administrator himself or the administrator can obtain the reason for the failure of the first to-be-verified voice information verification by obtaining the failure category information corresponding to the first to-be-verified voice information. And storing the failure category information and the first to-be-verified voice information determined to be pre-stored voice information to the identification library, and providing the user with corresponding service information when the subsequent user performs voice verification.
- S1550 Align the second to-be-verified voice information with the pre-stored voice information.
- the state of the user when outputting the second to-be-verified voice information may be determined. .
- the failure category information corresponding to the target voice information is a cold, it may be determined that the user is likely to have a cold at this time. The user may be provided with corresponding service information according to the obtained failure category information.
- the failure category information when the failure category information is obtained as a cold, the warm reminder information “drink hot water” and the like may be pushed to the user.
- the failure category information may also be transmitted to the health application on the terminal device, and the health application may record the health status of the user according to the failure category information, and provide the corresponding service information for the user, which may make the terminal device more intelligent.
- the operation after the failure category information is obtained may be set according to the requirements of the actual application, and the embodiment is not limited herein.
- FIG. 6 is a schematic flowchart diagram of a voice processing method according to another embodiment. As shown in FIG. 6, the method includes the following steps.
- S1600 Acquire a first to-be-verified voice information, and perform first verification on the first to-be-verified voice information.
- S1620 Receive verification information of the second verification, and determine, according to the verification information, whether the first to-be-verified voice information passes the second verification.
- S1640 Send the first to-be-verified voice information and the failure category information to a backend server.
- the first to-be-verified voice information and the failure category information may also be uploaded to the background server, and the background R&D staff may analyze and research the first to-be-verified voice information according to the failure category information, which may assist in research and development. Voice development and management corresponding to the failure category information.
- FIG. 7 is a structural block diagram of a voice processing apparatus according to an embodiment.
- the apparatus may perform a voice processing method.
- the apparatus includes: a first voice verification module 210, configured to acquire a first voice to be verified. And performing a first verification on the first to-be-verified voice information; the verification startup module 211 is configured to start the second verification when the first verification fails; wherein the second verification includes verification other than voice verification
- the verification module 212 is configured to receive the verification information of the second verification, and determine, according to the verification information, whether the first to-be-verified voice information passes the second verification, and the voice storage module 213 is configured to be in the first
- the voice information to be verified passes the second verification
- the first to-be-verified voice information is determined as pre-stored voice information, and the pre-stored voice information is stored in the identification library.
- the technical solution provided by the embodiment can be used to determine whether the voice information of the failed verification is stored in the identification library when the user's voice information fails to be verified, and the voice verification can be improved when the subsequent user uses the voice verification again. accuracy.
- the device further includes: a second voice verification module, configured to: after determining the first to-be-verified voice information as pre-stored voice information, and storing the pre-stored voice information to the identification library, acquiring The second to-be-verified voice information is verified according to the pre-stored voice information in the identification library, wherein the pre-stored voice information is at least one.
- a second voice verification module configured to: after determining the first to-be-verified voice information as pre-stored voice information, and storing the pre-stored voice information to the identification library, acquiring The second to-be-verified voice information is verified according to the pre-stored voice information in the identification library, wherein the pre-stored voice information is at least one.
- the verification module includes: a comparison unit configured to compare the second to-be-verified voice information with the pre-stored voice information; and a verification unit configured to be in any of the pre-stored voice information and the When the comparison result of the second to-be-verified voice information meets the preset condition, it is determined that the second to-be-verified voice information passes the verification.
- the voice storage module includes: a determining unit, configured to determine the first to-be-verified voice information as pre-stored voice information; and a pre-processing unit configured to pre-process the pre-stored voice information to obtain the The feature parameter of the pre-stored voice information; wherein the feature parameter includes a feature parameter embodying a voice trait; and the storage unit is configured to store the feature parameter of the pre-stored voice information to the recognition library.
- the comparison unit includes: a preprocessing subunit, configured to preprocess the second to-be-verified speech information, to obtain a feature parameter of the second to-be-verified speech information; and to calculate a sub-unit, set to Calculating an Euclidean distance between the feature parameter of the second to-be-verified speech information and the feature parameter of the pre-stored speech information; the verification unit is configured to: the feature parameter of any of the pre-stored voice information and the second to-be-verified speech When the Euclidean distance of the characteristic parameter of the information is less than the preset threshold, determining that the second to-be-verified voice information passes the verification.
- the device further includes: a category information acquiring module, configured to acquire, after the second verification verification information passes the second verification, failure category information corresponding to the first to-be-verified voice information;
- the voice storage module Is configured to: determine the first to-be-verified voice information as pre-stored voice information, and store the pre-stored voice information and the failure category information to the identification library;
- the device further includes: a category information determining module, configured to determine After the second to-be-verified voice information is verified, the failure category information corresponding to the target voice information is obtained from the identification library, and the target voice information is preset voice information corresponding to the comparison result of the preset condition.
- the device further includes: a category information acquiring module, configured to acquire, after the first to-be-verified voice information passes the second verification, failure category information corresponding to the first to-be-verified voice information; And configured to send the first to-be-verified voice information and the failure category information to a background server.
- a category information acquiring module configured to acquire, after the first to-be-verified voice information passes the second verification, failure category information corresponding to the first to-be-verified voice information.
- the second verification comprises at least one of fingerprint verification, password verification, graphics verification, face recognition verification, and iris recognition verification.
- the apparatus further includes: an information pushing module, configured to: after acquiring the failure category information corresponding to the target voice information from the identification library, according to the failure category corresponding to the target voice information The information pushes service information corresponding to the failure category information.
- the present invention provides a storage medium containing computer executable instructions that are not limited to the voice processing operations as described above, and may also perform related operations in the voice processing methods provided by any of the embodiments of the present disclosure.
- the embodiment further provides a storage medium containing computer executable instructions for performing a voice processing method when executed by a computer processor, the method comprising: acquiring voice information to be verified, and The first verification is performed on the voice information to be verified; if the first verification fails, the second verification is initiated; wherein the second verification includes a verification mode other than voice verification; and the verification information of the second verification is received, according to the verification information Determining whether the to-be-verified voice information passes the second verification; if the to-be-verified voice information passes the second verification, determining the to-be-verified voice information as pre-stored voice information, and storing the pre-stored voice information to Identify the library.
- Storage medium any type of storage device or storage device.
- the term "storage medium” is intended to include: a mounting medium such as a Compact Disc Read-Only Memory (CD-ROM), a floppy disk or a tape device; a computer system memory or a random access memory such as a dynamic random Random Access Memory (DRAM), Double Data Rate Random Access Memory (DDR RAM), Static Random Access Memory (SRAM), Extended Data Output Random Extended Data Output Random Access Memory (EDO RAM), Rambus Random Access Memory (RAM), etc.; non-volatile memory such as flash memory, magnetic media (such as hard disk or light) Storage); registers or other similar types of memory elements, etc.
- the storage medium may also include other types of memory or a combination thereof.
- the storage medium may be located in a first computer system in which the program is executed, or may be located in a different second computer system, the second computer system being coupled to the first computer system via a network, such as the Internet.
- the second computer system can provide program instructions to the first computer for execution.
- the term "storage medium" can include two or more storage media that can reside in different locations (eg, in different computer systems connected through a network).
- a storage medium may store program instructions (eg, program instructions implemented as a computer program) executable by one or more processors.
- the embodiment provides a terminal device, and the voice processing device provided by the embodiment is integrated into the terminal device.
- FIG. 8 is a schematic structural diagram of a terminal device according to an embodiment.
- This embodiment provides a terminal device, including a memory 31, a processor 32, and a computer program stored in the memory 31 and operable on the processor.
- the speech processing method described in the above embodiments is implemented when the processor executes the computer program.
- the terminal device provided in this embodiment can determine whether the voice information failed to be verified is stored in the identification library by determining the identity of the user when the voice information of the user fails to be verified, and the voice verification can be improved when the subsequent user uses the voice verification again. Sex.
- FIG. 9 is a schematic structural diagram of a terminal device according to another embodiment.
- the terminal device may include: a casing (not shown in FIG. 9), a touch screen (not shown in FIG. 9), a touch button (not shown in FIG. 9), a memory 301, and a central processing unit.
- CPU Central Processing Unit
- CPU 302 also referred to as a processor, hereinafter referred to as CPU
- circuit board not shown in FIG. 9
- a power supply circuit not shown in FIG. 9
- the circuit board is disposed inside a space enclosed by the casing; the CPU 302 and the memory 301 are disposed on the circuit board; and the power circuit is configured to be at least one circuit or device of the terminal device
- the memory 301 is configured to store executable program code; the CPU 302 runs a computer program corresponding to the executable program code by reading executable program code stored in the memory 301 to implement the following steps Obtaining the voice information to be verified, and performing the first verification on the voice information to be verified; if the first verification fails, starting the second verification; wherein the second verification includes a verification mode other than voice verification; And verifying, by the verification information, whether the voice information to be verified passes the second verification, and if the voice information to be verified passes the second verification, determining the voice information to be verified as The voice information is pre-stored and the pre-stored voice information is stored in the identification library.
- the terminal device further includes: a peripheral interface 303, a radio frequency (RF) circuit 305, an audio circuit 306, a speaker 311, a power management chip 308, an input/output (I/O) subsystem 309, a touch screen 312, and others.
- Input/control device 310 and external port 304 are communicated via one or more communication buses or signal lines 307.
- the terminal device 300 illustrated in FIG. 9 is merely an example of a terminal device, and the terminal device 300 may have more or fewer components than those illustrated in FIG. 9, and two or more components may be combined, or Can have different component configurations.
- the various components shown in Figure 9 can be implemented in hardware, software, or a combination of hardware and software, including one or more signal processing and/or application specific integrated circuits.
- the following describes the terminal device for implementing voice processing provided by this embodiment, where the terminal device takes a mobile phone as an example.
- the memory 301 can be accessed by the CPU 302 and the peripheral interface 303, etc., and the memory 301 can include a high speed random access memory, and can also include a nonvolatile memory such as one or more magnetic disk storage devices, flash memory devices. Or other volatile solid-state storage devices.
- a nonvolatile memory such as one or more magnetic disk storage devices, flash memory devices. Or other volatile solid-state storage devices.
- Peripheral interface 303 which can connect the input and output peripherals of the device to CPU 302 and memory 301.
- I/O subsystem 309 which can connect input peripherals and output peripherals on the device, such as touch screen 312 and other input/control devices 310, to peripheral interface 303.
- I/O subsystem 309 can include display controller 3091 and one or more input controllers 3092 that are configured to control other input/control devices 310.
- one or more input controllers 3092 receive electrical signals from other input/control devices 310 or transmit electrical signals to other input/control devices 310, and other input/control devices 310 may include physical buttons (pressing buttons and rocker buttons, etc.) ), dial, slide switch, joystick, and click wheel.
- the input controller 3092 can be connected to any of the following: a keyboard, an infrared port, a USB interface, and a pointing device such as a mouse.
- the touch screen 312 is an input interface and an output interface between the user terminal device and the user, and displays the visual output to the user.
- the visual output may include graphics, text, icons, videos, and the like.
- Display controller 3091 in I/O subsystem 309 receives an electrical signal from touch screen 312 or an electrical signal to touch screen 312.
- the touch screen 312 detects the contact on the touch screen 312, and the display controller 3091 converts the detected contact into an interaction with the user interface object displayed on the touch screen 312, that is, realizes human-computer interaction, and the user interface object displayed on the touch screen 312 may be Run the game's icons, connect to the corresponding network icons, and more.
- the device may further include a light mouse, which is a touch sensitive surface that does not display a visual output, or an extension of a touch sensitive surface formed by the touch screen.
- the RF circuit 305 is mainly configured to establish communication between the mobile phone and the wireless network (ie, the network side), and implement data reception and transmission between the mobile phone and the wireless network. For example, sending and receiving short messages and emails.
- the RF circuit 305 receives and transmits an RF signal, which is also referred to as an electromagnetic signal, and the RF circuit 305 converts the electrical signal into an electromagnetic signal or converts the electromagnetic signal into an electrical signal, and through the electromagnetic signal and the communication network and Other devices communicate.
- RF circuitry 305 may include known circuitry for performing these functions, including but not limited to an antenna system, an RF transceiver, one or more amplifiers, a tuner, one or more oscillators, a digital signal processor, CODER-DECoder (CODEC) chipset, Subscriber Identity Module (SIM), etc.
- an antenna system an RF transceiver, one or more amplifiers, a tuner, one or more oscillators, a digital signal processor, CODER-DECoder (CODEC) chipset, Subscriber Identity Module (SIM), etc.
- CODER-DECoder CODER-DECoder
- SIM Subscriber Identity Module
- the audio circuit 306 is primarily configured to receive audio data from the peripheral interface 303, convert the audio data into an electrical signal, and transmit the electrical signal to the speaker 311.
- the speaker 311 is arranged to restore the voice signal received by the handset from the wireless network via the RF circuit 305 to sound and play the sound to the user.
- the power management chip 308 is configured to provide power and power management for the hardware connected to the CPU 302, the I/O subsystem, and the peripheral interface.
- the terminal device provided in the embodiment of the present application can improve the accuracy of voice verification.
- the voice processing device, the storage medium, and the terminal device provided in the foregoing embodiments may perform the voice processing method provided by any embodiment of the present disclosure, and have corresponding functional modules and beneficial effects for performing the method.
- the voice processing method provided by any embodiment of the present disclosure.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Telephone Function (AREA)
- Telephonic Communication Services (AREA)
Abstract
一种语音处理方法，该语音处理方法通过获取待验证语音信息，并对待验证语音信息进行第一验证；如果第一验证失败，则启动第二验证；其中，第二验证包括除语音验证以外的验证方式；接收第二验证的验证信息，根据验证信息判断待验证语音信息是否通过第二验证；如果待验证语音信息通过第二验证，则将待验证语音信息确定为预存语音信息，并将预存语音信息存储至识别库。
Description
本公开要求在2017年12月14日提交中国专利局、申请号为201711339174.0的中国专利申请的优先权，该申请的全部内容通过引用结合在本公开中。
本公开涉及计算机语音处理技术领域，例如涉及一种语音处理方法、装置、存储介质及终端设备。
随着终端设备的发展，终端设备逐渐成为人们日常生活必不可少且随身携带的工具，终端设备上也有诸多关于用户的私密或重要的信息，用户一般会在终端设备上设置身份验证，以防被不怀好意的人通过终端设备窃取用户的信息。验证方式包括基于语音的识别来进行验证用户身份的方法，即终端设备获取用户说的语音，并判断该语音是不是用户本人的声音来验证用户的身份。而用户的声音会受到自身因素或外部因素的干扰而发生变化，例如用户如果感冒了，用户感冒的声音会比平时的声音显得鼻音较重，而终端设备可能就无法识别出用户感冒的声音，导致语音识别的准确率低下。
发明内容
本公开提供一种语音处理方法、装置、存储介质及终端设备，可以提高语音验证的准确性。
本公开提供了一种语音处理方法，包括：
获取第一待验证语音信息，并对所述第一待验证语音信息进行第一验证；
如果所述第一验证失败，则启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式；
接收所述第二验证的验证信息，根据所述验证信息判断所述第一待验证语音信息是否通过所述第二验证；
如果所述第一待验证语音信息通过所述第二验证，则将所述第一待验证语音信息确定为预存语音信息，并将预存语音信息存储至识别库。
本公开还提供了一种语音处理装置，包括：
语音验证模块，设置为获取第一待验证语音信息，并对所述第一待验证语音信息进行第一验证；
验证启动模块，设置为在第一验证失败时，启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式；
验证模块，设置为接收第二验证的验证信息，根据所述验证信息判断所述第一待验证语音信息是否通过所述第二验证；
语音存储模块，设置为在所述第一待验证语音信息通过所述第二验证时，将所述第一待验证语音信息确定为预存语音信息，并将预存语音信息存储至识别库。
本公开还提供了一种计算机可读存储介质，计算机可读存储介质上存储有计算机程序，该计算机程序被处理器执行时实现如本公开实施例所述的语音处理方法。
本公开还提供了一种终端设备，包括存储器，处理器及存储在存储器上并可在处理器运行的计算机程序，所述处理器执行所述计算机程序时实现如本公开实施例所述的语音处理方法。
本公开提供的一种语音处理方案可以在用户的语音信息验证失败时，通过判断用户身份决定是否将验证失败的语音信息存入识别库，可以在后续用户再次使用语音验证时候，提高语音验证的准确性。
图1为一实施例提供的一种语音处理方法的流程示意图；
图2为另一实施例提供的一种语音处理方法的流程示意图；
图3为另一实施例提供的一种语音处理方法的流程示意图；
图4为另一实施例提供的一种语音处理方法的流程示意图；
图5为另一实施例提供的一种语音处理方法的流程示意图；
图6为另一实施例提供的一种语音处理方法的流程示意图；
图7为一实施例提供的一种语音处理装置的结构框图；
图8为一实施例提供的一种终端设备的结构示意图；
图9为另一实施例提供的一种终端设备的结构示意图。
下面结合附图并通过具体实施方式来说明本公开的技术方案。此处所描述的具体实施例仅仅用于解释本公开，而非对本公开的限定。另外，为了便于描述，附图中仅示出了与本公开相关的部分而非全部结构。
终端设备可包括智能手机、平板电脑和其他具有操作系统的设备。终端设备的解锁方式可包括语音验证方式，终端设备上的应用程序(Application，APP)的开启验证也可包括语音验证方式，但是如果用户的声音因为感冒发生一些变化时，可能无法通过语音验证。而此时通过其他方式验证用户的身份，可以将感冒的语音作为验证参考语音，可以提高后续语音验证的准确性。
图1为一实施例提供的一种语音处理方法的流程示意图，该方法可以由语音处理装置执行，该装置可以由软件和/或硬件实现，一般可以集成在终端设备中，也可以集成在其他安装有操作系统的设备中。如图1所示，该方法包括如下步骤。
S1100、获取待验证语音信息，并对所述待验证语音信息进行第一验证。
所述待验证语音信息可为终端设备的用户在语音验证时输出的待验证语音信息，通过终端设备获取该待验证语音信息。语音验证可以是终端设备通过用户输出的待验证语音信息来验证是否为管理员本人的验证方式。待验证语音信息可以是所述用户用于解锁终端设备所输出的语音信息，还可以是所述用户用于解锁终端设备上的应用程序所输出的语音信息。在一 实施例中，所述应用程序可以是微信，用户可以通过说出预先设置的文字或数字，根据用户说的语音来判断用户是否通过第一验证。
用户通过输出待验证语音信息，终端设备可以对该待验证语音信息进行分析和验证。在一实施例中，可以通过将待验证语音信息与识别库的预设语音信息进行比对来对待验证语音信息进行验证，预设语音信息可以是终端设备的管理员本人最早使用对应的语音验证时所录制的标准语音信息，通过将待验证语音信息与标准语音信息来进行比对，可以判断该待验证语音信息是否与标准语音信息是否匹配，进而可以判断用户是否是终端设备的管理员本人。
S1110、如果第一验证失败，则启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式。
在一实施例中，所述第二验证可包括指纹验证、密码验证、图形验证、人脸识别验证和虹膜识别验证中的至少一种。如果所述待验证语音信息验证失败了，则可以通过启动第二次验证来验证用户是否为管理员本人。终端设备上的验证方式包括多种方式，可以根据用户设置或系统预设来确定第二次验证的方式，本实施例在此不作限定。
S1120、接收第二验证的验证信息，根据所述验证信息判断所述待验证语音信息是否通过所述第二验证。
所述第二验证的验证信息为用户根据启动的第二验证所输入的验证信息。在一实施例中，如果第二验证为指纹验证，则第二验证的验证信息为用户的指纹信息；如果第二验证为密码验证，则第二验证的验证信息为用户输入的密码信息。根据所述第二验证的验证信息判断所述待验证语音信息是否通过所述第二验证，可以确定用户是否为管理员本人。
S1130、如果所述待验证语音信息通过所述第二验证，则将所述待验证语音信息确定为预存语音信息，并将预存语音信息存储至识别库。
如果待验证语音信息通过所述第二验证，则可以确定用户为管理员本人，相应的所述待验证语音信息也是管理员本人的语音信息，用户可能因为一些自身因素或外部因素导致声音发生变化，可以将所述待验证语音确定为预存语音信息，并将预存语音信息存储至识别库。如果下次用户还是因为同样的自身因素或外部因素导致声音发生变化时，输出的待验证语音信息可以通过验证。
例如，用户因为感冒，声音发生了变化，则可以将感冒的待验证语音信息确定为预存语音信息，并将预存语音信息存储至识别库，如果下次用户再发生感冒时，识别库中已存储有用户在感冒时的预存语音信息，则用户感冒的待验证语音信息也能通过验证。所以通过将所述待验证语音信息确定为预存语音信息并存储至识别库可以提高后续的语音验证的准确性。
在一实施例中，如果管理员本人的亲友需要与管理员本人共同使用终端设备或终端设备上的应用程序，也可以通过本实施例的技术方案，将管理员本人的亲友的待验证语音信息确定为预存语音信息，并将预存语音信息存储至识别库。在一实施例中，可以通过获取他人(例如亲友)的待验证语音信息，而因为他人的待验证语音信息与用户的标准语音信息不同，第一验证肯定会验证失败，相应的启动第二验证。而此时，用户(管理员本人)可以输入第二 验证的验证信息，通过第二验证后，则可以将他人的待验证语音信息作为预存语音信息存储至识别库，后续他人便可以通过自己的声音通过终端设备或终端设备上的应用程序的语音验证。通过本实施例可以提高语音识别的使用效率，还可以扩充语音验证的功能。
图2为另一实施例提供的一种语音处理方法的流程示意图。如图2所示，该方法包括如下步骤。
S1200、获取第一待验证语音信息，并对所述第一待验证语音信息进行第一验证。
S1210、如果第一验证失败，则启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式。
S1220、接收第二验证的验证信息，根据所述验证信息判断所述第一待验证语音信息是否通过所述第二验证。
S1230、如果所述第一待验证语音信息通过所述第二验证，则将所述第一待验证语音信息确定为预存语音信息，并将预存语音信息存储至识别库。
上述操作的具体实施方式可以参考上文的相关描述，在此不再赘述。
S1240、获取第二待验证语音信息，根据识别库中的预存语音信息对第二待验证语音信息进行验证；其中，所述预存语音信息为至少一个。
在一实施例中，所述第二待验证语音信息可以是用户在后续使用语音验证时所输出的待验证语音信息。根据识别库中的预存语音信息对第二待验证语音信息进行验证，所述预存语音信息可以是通过上述实施例中所述的确定为预存语音信息的第一待验证语音信息，也可以是识别库中存储的其他预存语音信息。在一实施例，其他预存语音信息可以是终端设备的管理员本人在最早使用或启动语音验证时，所录制的标准语音信息。所以识别库中的预存语音信息为至少一个，即至少包括管理员本人在最早使用或启动语音验证时，所录制的标准语音信息。如果预存语音信息为一个以上，则可包括通过上述实施例中所述的确定为预存语音信息的第一待验证语音信息。
在一实施例中，识别库中的预存语音信息也可以只包括上述实施例中所述的确定为预存语音信息的第一待验证语音信息；而管理员本人在最早使用或启动语音验证时，所录制的标准语音信息可以存储在其他存储模块中。识别库中只要包括上述实施例中所述的确定为预存语音信息的第一待验证语音信息，便可以增加用于语音验证的比对素材，提高语音验证的准确性。
在一实施例中，所述根据识别库中的预存语音信息对第二待验证语音信息进行验证可以通过下述方式进行实施：
S12400、将所述第二待验证语音信息和所述预存语音信息进行比对。
S12410、如果任一所述预存语音信息和所述第二待验证语音信息的比对结果符合预设条件，则确定所述第二待验证语音信息通过验证。
所述将所述第二待验证语音信息和所述预存语音信息进行比对，可以是将第二待验证语音信息的特征信息和预存语音信息的特征信息进行比对，所述预设条件可以是比对误差值小 于设定误差值。如果第二待验证语音信息的特征信息和预存语音信息的特征信息之间的误差值小于设定误差值，则可以确定第二待验证语音信息和预存语音信息两个语音信息匹配成功，则可以确定所述第二待验证语音信息通过验证。
在一实施例中，识别库中存储的预存语音信息为一个以上，则只要识别库中任意一个预存语音信息和第二待验证语音信息的比对结果符合预设条件，则可以确定所述第二待验证语音信息通过验证。如果识别库中存储的预存语音信息为一个以上，则识别库中一定包括所述的确定为预存语音信息的第一待验证语音信息，还可包括管理员本人在最早使用或启动语音验证时，所录制的标准语音信息。所以只要任意一个预存语音信息和第二待验证语音信息的比对结果符合预设条件，则可以确定输出第二待验证语音信息的用户为终端设备的管理员本人，用户可以通过语音验证。
通过将所述第二待验证语音信息和所述预存语音信息进行比对，如果任一所述预存语音信息和所述第二待验证语音信息的比对结果符合预设条件，则确定所述第二待验证语音信息通过验证，可以使用户进行语音验证时，将用户输出的第二待验证语音信息与不同的比对素材进行比对，可以提高语音验证的准确性。
图3为另一实施例提供的一种语音处理方法的流程示意图。如图3所示，该方法包括如下步骤。
S1300、获取待验证语音信息，并对所述待验证语音信息进行第一验证。
S1310、如果第一验证失败，则启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式。
S1320、接收第二验证的验证信息，根据所述验证信息判断所述待验证语音信息是否通过所述第二验证。
上述操作的具体实施方式可以参考上文的相关描述，在此不再赘述。
S1330、将所述待验证语音信息确定为预存语音信息。
S1340、对所述预存语音信息进行预处理，得到所述预存语音信息的特征参数；其中，所述特征参数包括体现声音特质的特征参数。
在一实施例中，预处理过程可包括：对预存语音信息进行分帧处理，以得到至少一个的语音帧。在一实施例中，每个语音帧的帧长可以是20毫秒(ms)至50ms中的任意数值。因为语音信息是不稳定的信号，语音信息中的信号变化一般都非常快速，所以将语音信息进行分帧处理，得到的语音帧具有足够长的周期，语音帧内的信号也不会发生剧烈变化，可以提高后续处理的效率。
分别对每一个语音帧进行提取特征参数的处理。在一实施例中，特征参数可包括质心、均方根和梅尔频率倒谱系数(Mel Frequency Cepstrum Coefficient)等参数。质心可以体现语音帧内信号的基本频率带，均方根可以体现语音帧的信号强弱，梅尔倒谱系数可以模拟体现人耳对不同频率语音的感知，可以有效表示人的语音的声音特质。特征参数还可以包括声能值、基音频率和共振峰值等可以体现声音特质的特征参数。
S1350、将所述预存语音信息的特征参数存储至识别库。
因为语音信息的大小一般较大，一段10秒(s)左右的语音信息的大小一般为十几千字节(Kilobyte，KB)，如果直接将预存语音信息存储至识别库，可能会导致占用过多存储空间。而使用预存语音信息一般是在后台进行计算处理，用户无需再次听到预存语音信息，所以可以将所述预存语音信息的特征参数存储至识别库，如此可以保存语音信息中的可以体现声音特质的特征参数，能够实现后续的计算处理，且占用很小的存储空间。
图4为另一实施例提供的一种语音处理方法的流程示意图。如图4所示，该方法包括如下步骤。
S1400、获取第一待验证语音信息，并对所述第一待验证语音信息进行第一验证。
S1410、如果第一验证失败，则启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式。
S1420、接收第二验证的验证信息，根据所述验证信息判断所述第一待验证语音信息是否通过所述第二验证。
S1430、将所述第一待验证语音信息确定为预存语音信息。
S1440、对所述预存语音信息进行预处理，得到所述预存语音信息的特征参数；其中，所述特征参数包括体现声音特质的特征参数。
S1450、将所述预存语音信息的特征参数存储至识别库。
上述操作的具体实施方式可以参考上文的相关描述，在此不再赘述。
S1460、获取第二待验证语音信息，对所述第二待验证语音信息进行预处理，得到所述第二待验证语音信息的特征参数。
本实施例中，所述特征参数可包括体现声音特质的特征参数。可包括质心、均方根和梅尔频率倒谱系数(Mel Frequency Cepstrum Coefficient)等参数。还可以包括声能值、基音频率和共振峰值等可以体现声音特质的特征参数。具体实施方式可以参考上文的相关描述，在此不再赘述。
S1470、计算所述第二待验证语音信息的特征参数与所述预存语音信息的特征参数的欧氏距离。
S1480、如果任一所述预存语音信息的特征参数和所述第二待验证语音信息的特征参数的欧氏距离小于预设阈值，则确定所述第二待验证语音信息通过验证。
本实施例中，所述欧氏距离(Euclidean Distance)指的是在一个N维空间中的两点之间的真实距离或两点形成的向量的自然长度，N可以是大于0的任意自然数。在一实施例中，如果特征参数包括质心、均方根和梅尔频率倒谱系数，则在质心、均方根和梅尔频率倒谱系数形成的三维空间中，计算第二待验证语音信息的(a1，b1，c1)和预存语音信息的(a2，b2，c2)两点形成的向量的自然长度即欧氏距离；其中，a1和a2为质心，b1和b2为均方根，c1和c2为梅尔频率倒谱系数。如果预存语音信息的特征参数和第二待验证语音信息的特征参数的欧氏距离小于预设阈值，则可以确定该预存语音信息和该第二待验证语音信息比较接近，相互匹配， 可以确定所述第二待验证语音信息通过验证。
通过计算所述第二待验证语音信息的特征参数与所述预存语音信息的特征参数的欧氏距离，如果任一所述预存语音信息的特征参数和所述第二待验证语音信息的特征参数的欧氏距离小于预设阈值，则确定所述第二待验证语音信息通过验证，可以比较两个语音信息中的体现声音特质的特征参数，能够有效提高比对的效率。
图5为另一实施例提供的一种语音处理方法的流程示意图。如图5所示，该方法包括如下步骤。
S1500、获取第一待验证语音信息，并对所述第一待验证语音信息进行第一验证。
S1510、如果第一验证失败，则启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式。
S1520、接收第二验证的验证信息，根据所述验证信息判断所述第一待验证语音信息是否通过所述第二验证。
上述操作的具体实施方式可以参考上文的相关描述，在此不再赘述。
S1530、如果所述第一待验证语音信息通过所述第二验证，获取与第一待验证语音信息对应的失败类别信息。
在一实施例中，所述失败类别信息可以是系统预存的多个类别信息，获取与第一待验证语音信息对应的失败类别信息可以是将系统预存的多个类别信息发送给用户，并根据用户的选择确定多个类别信息中至少一个类别信息为失败类别信息。失败类别信息还可以是用户输入的自定义内容。在一实施例中，系统预存的多个类别信息可包括感冒、声音沙哑和喉咙发炎等，用户也可以通过输入自定义内容“感冒了”、“声音哑了”和“亲友”来作为失败类别信息。
S1540、将所述第一待验证语音信息确定为预存语音信息，并将所述预存语音信息和所述失败类别信息存储至识别库。
如果用户输入的第一待验证语音信息未通过操作S1500中的第一验证，但是用户输入的第二验证的验证信息通过了第二验证，则可以确定验证失败的第一待验证语音信息为管理员本人或获得管理员授权，可以通过获取第一待验证语音信息对应的失败类别信息，来确定第一待验证语音信息验证失败的原因。并将失败类别信息以及确定为预存语音信息的第一待验证语音信息存储至识别库，可以给后续用户进行语音验证时，为用户提供相应的服务信息。
S1550、将所述第二待验证语音信息和所述预存语音信息进行比对。
S1560、如果任一所述预存语音信息和所述第二待验证语音信息的比对结果符合预设条件，则确定所述第二待验证语音信息通过验证。
上述操作的具体实施方式可以参考上文的相关描述，在此不再赘述。
S1570、从所述识别库中获取与目标语音信息对应的失败类别信息，所述目标语音信息为符合预设条件的比对结果对应的预设语音信息。
如果第二待验证语音信息验证通过，则确定与第二待验证语音信息比对结果符合预设条 件的目标语音信息对应的失败类别信息，可以确定用户在输出第二待验证语音信息时的状态。在一实施例中，如果用户输入的第二待验证语音信息和目标语音信息比对结果符合预设条件，且该目标语音信息对应的失败类别信息为感冒，则可以确定用户此时很可能感冒了；可以根据获取到的失败类别信息为用户提供相应的服务信息，例如可以在获取到失败类别信息为感冒时，推送温馨提醒信息“多喝热水”等信息给用户。还可以将失败类别信息传输至终端设备上的健康类应用程序，健康类应用程序可以根据失败类别信息记录用户的健康状态，并为用户提供相应的服务信息，可以使终端设备更智能化。
获取到失败类别信息后的操作可以根据实际应用的需求而设定，本实施例在此不作限定。
图6为另一实施例提供的一种语音处理方法的流程示意图。如图6所示，该方法包括如下步骤。
S1600、获取第一待验证语音信息，并对所述第一待验证语音信息进行第一验证。
S1610、如果第一验证失败，则启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式。
S1620、接收第二验证的验证信息，根据所述验证信息判断所述第一待验证语音信息是否通过所述第二验证。
S1630、在所述第一待验证语音信息通过所述第二验证之后，获取与第一待验证语音信息对应的失败类别信息。
上述操作的具体实施方式可以参考上文的相关描述，在此不再赘述。
S1640、将所述第一待验证语音信息和所述失败类别信息发送至后台服务器。
在一实施例中，还可以将第一待验证语音信息和所述失败类别信息上传至后台服务器，后台研发工作人员可以根据失败类别信息对第一待验证语音信息进行分析和研发，可以辅助研发人员关于失败类别信息对应的语音研发和管理。
图7为一实施例提供的一种语音处理装置的结构框图，该装置可以执行语音处理方法，如图7所示，该装置包括：第一语音验证模块210，设置为获取第一待验证语音信息，并对所述第一待验证语音信息进行第一验证；验证启动模块211，设置为在第一验证失败时，启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式；验证模块212，设置为接收第二验证的验证信息，根据所述验证信息判断所述第一待验证语音信息是否通过所述第二验证；语音存储模块213，设置为在所述第一待验证语音信息通过所述第二验证时，将所述第一待验证语音信息确定为预存语音信息，并将所述预存语音信息存储至识别库。
通过本实施例提供的技术方案，可以在用户的语音信息验证失败时，通过判断用户身份决定是否将验证失败的语音信息存入识别库，可以在后续用户再次使用语音验证时候，提高语音验证的准确性。
在一实施例中，上述装置还包括：第二语音验证模块，设置为在将所述第一待验证语音信息确定为预存语音信息，并将所述预存语音信息存储至识别库之后，获取第二待验证语音信息，根据所述识别库中的预存语音信息对所述第二待验证语音信息进行验证；其中，所述 预存语音信息为至少一个。
在一实施例中，验证模块包括：比对单元，设置为将所述第二待验证语音信息和所述预存语音信息进行比对；验证单元，设置为在任一所述预存语音信息和所述第二待验证语音信息的比对结果符合预设条件时，确定所述第二待验证语音信息通过验证。
在一实施例中，语音存储模块包括：确定单元，设置为将所述第一待验证语音信息确定为预存语音信息；预处理单元，设置为对所述预存语音信息进行预处理，得到所述预存语音信息的特征参数；其中，所述特征参数包括体现声音特质的特征参数；存储单元，设置为将所述预存语音信息的特征参数存储至识别库。
在一实施例中，比对单元包括：预处理子单元，设置为对所述第二待验证语音信息进行预处理，得到所述第二待验证语音信息的特征参数；计算子单元，设置为计算所述第二待验证语音信息的特征参数与所述预存语音信息的特征参数的欧氏距离；验证单元是设置为：在任一所述预存语音信息的特征参数和所述第二待验证语音信息的特征参数的欧氏距离小于预设阈值时，确定所述第二待验证语音信息通过验证。
在一实施例中，上述装置还包括：类别信息获取模块，设置为在第二验证的验证信息通过所述第二验证之后，获取与第一待验证语音信息对应的失败类别信息；语音存储模块是设置为：将所述第一待验证语音信息确定为预存语音信息，并所述预存语音信息和所述失败类别信息存储至识别库；上述装置还包括：类别信息确定模块，设置为在确定第二待验证语音信息通过验证之后，从所述识别库中获取与目标语音信息对应的失败类别信息，所述目标语音信息为符合预设条件的比对结果对应的预设语音信息。
在一实施例中，上述装置还包括：类别信息获取模块，设置为在第一待验证语音信息通过所述第二验证之后，获取与第一待验证语音信息对应的失败类别信息；后台发送模块，设置为将所述第一待验证语音信息和所述失败类别信息发送至后台服务器。
在一实施例中，所述第二验证包括指纹验证、密码验证、图形验证、人脸识别验证和虹膜识别验证中的至少一种。
在一实施例中，所述装置还包括：还包括：信息推送模块，设置为在从所述识别库中获取与目标语音信息对应的失败类别信息之后，根据所述目标语音信息对应的失败类别信息推送与所述失败类别信息对应的服务信息。
本实施例所提供的一种包含计算机可执行指令的存储介质，计算机可执行指令不限于如上所述的语音处理操作，还可以执行本公开任意实施例所提供的语音处理方法中的相关操作。
本实施例还提供一种包含计算机可执行指令的存储介质，所述计算机可执行指令在由计算机处理器执行时用于执行语音处理方法，该方法包括：获取待验证语音信息，并对所述待验证语音信息进行第一验证；如果第一验证失败，则启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式；接收第二验证的验证信息，根据所述验证信息判断所述待验证语音信息是否通过所述第二验证；如果所述待验证语音信息通过所述第二验证，则将所述待验证语音信息确定为预存语音信息，并将预存语音信息存储至识别库。
存储介质——任何的一种类型的存储器设备或存储设备。术语“存储介质”旨在包括：安装介质，例如紧凑型光盘只读储存器(Compact Disc Read-Only Memory，CD-ROM)、软盘或磁带装置；计算机系统存储器或随机存取存储器，诸如动态随机存取存储器(Dynamic Random Access Memory，DRAM)、双倍数据速率随机存取存储器(Double Data Rate Random Access Memory，DDR RAM)、静态随机存取存储器(Static Random Access Memory，SRAM)、扩展数据输出随机存取存储器(Extended Data Output Random Access Memory，EDO RAM)，兰巴斯(Rambus)随机存取存储器(Random Access Memory，RAM)等；非易失性存储器，诸如闪存、磁介质(例如硬盘或光存储)；寄存器或其它相似类型的存储器元件等。存储介质可以还包括其它类型的存储器或其组合。另外，存储介质可以位于程序在其中被执行的第一计算机系统中，或者可以位于不同的第二计算机系统中，第二计算机系统通过网络(诸如因特网)连接到第一计算机系统。第二计算机系统可以提供程序指令给第一计算机用于执行。术语“存储介质”可以包括可以驻留在不同位置中(例如在通过网络连接的不同计算机系统中)的两个或更多存储介质。存储介质可以存储可由一个或多个处理器执行的程序指令(例如程序指令实现为计算机程序)。
本实施例提供了一种终端设备，该终端设备中可集成公开实施例提供的语音处理装置。
图8为一实施例提供的一种终端设备的结构示意图，本实施例提供了一种终端设备，包括存储器31，处理器32及存储在存储器31上并可在处理器运行的计算机程序，所述处理器执行所述计算机程序时实现上述实施例所述的语音处理方法。本实施例提供的终端设备，可以在用户的语音信息验证失败时，通过判断用户身份决定是否将验证失败的语音信息存入识别库，可以在后续用户再次使用语音验证时候，提高语音验证的准确性。
图9为另一实施例提供的一种终端设备的结构示意图。如图9所示，该终端设备可以包括：壳体(图9中未示出)、触摸屏(图9中未示出)、触摸按键(图9中未示出)、存储器301、中央处理器(Central Processing Unit，CPU)302(又称处理器，以下简称CPU)、电路板(图9中未示出)和电源电路(图9中未示出)。所述电路板安置在所述壳体围成的空间内部；所述CPU302和所述存储器301设置在所述电路板上；所述电源电路，设置为为所述终端设备的至少一个电路或器件供电；所述存储器301，设置为存储可执行程序代码；所述CPU302通过读取所述存储器301中存储的可执行程序代码来运行与所述可执行程序代码对应的计算机程序，以实现以下步骤：获取待验证语音信息，并对所述待验证语音信息进行第一验证；如果第一验证失败，则启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式；接收第二验证的验证信息，根据所述验证信息判断所述待验证语音信息是否通过所述第二验证；如果所述待验证语音信息通过所述第二验证，则将所述待验证语音信息确定为预存语音信息，并将预存语音信息存储至识别库。
所述终端设备还包括：外设接口303、射频(Radio Frequency，RF)电路305、音频电路306、扬声器311、电源管理芯片308、输入/输出(I/O)子系统309、触摸屏312、其他输入/控制设备310以及外部端口304，这些部件通过一个或多个通信总线或信号线307来通信。
图9示出的终端设备300仅仅是终端设备的一个范例，并且终端设备300可以具有比图9中所示出的更多的或者更少的部件，可以组合两个或更多的部件，或者可以具有不同的部件配置。图9中所示出的多种部件可以在包括一个或多个信号处理和/或专用集成电路在内的硬件、软件、或硬件和软件的组合中实现。
下面就本实施例提供的用于实现语音处理的终端设备进行描述，该终端设备以手机为例。
存储器301，所述存储器301可以被CPU302以及外设接口303等访问，所述存储器301可以包括高速随机存取存储器，还可以包括非易失性存储器，例如一个或多个磁盘存储器件、闪存器件、或其他易失性固态存储器件。
外设接口303，所述外设接口303可以将设备的输入和输出外设连接到CPU302和存储器301。
I/O子系统309，所述I/O子系统309可以将设备上的输入外设和输出外设，例如触摸屏312和其他输入/控制设备310，连接到外设接口303。I/O子系统309可以包括显示控制器3091和设置为控制其他输入/控制设备310的一个或多个输入控制器3092。其中，一个或多个输入控制器3092从其他输入/控制设备310接收电信号或者向其他输入/控制设备310发送电信号，其他输入/控制设备310可以包括物理按钮(按压按钮以及摇臂按钮等)、拨号盘、滑动开关、操纵杆以及点击滚轮。在一实施例中，输入控制器3092可以与以下任一个连接：键盘、红外端口、USB接口以及诸如鼠标的指示设备。
触摸屏312，所述触摸屏312是用户终端设备与用户之间的输入接口和输出接口，将可视输出显示给用户，可视输出可以包括图形、文本、图标以及视频等。
I/O子系统309中的显示控制器3091从触摸屏312接收电信号或者向触摸屏312发送电信号。触摸屏312检测触摸屏312上的接触，显示控制器3091将检测到的接触转换为与显示在触摸屏312上的用户界面对象的交互，即实现人机交互，显示在触摸屏312上的用户界面对象可以是运行游戏的图标、联网到相应网络的图标等。本实施例中，设备还可以包括光鼠，光鼠是不显示可视输出的触摸敏感表面，或者是由触摸屏形成的触摸敏感表面的延伸。
RF电路305，主要设置为建立手机与无线网络(即网络侧)的通信，实现手机与无线网络的数据接收和发送。例如收发短信息以及电子邮件等。本实施例中，RF电路305接收并发送RF信号，RF信号也称为电磁信号，RF电路305将电信号转换为电磁信号或将电磁信号转换为电信号，并且通过该电磁信号与通信网络以及其他设备进行通信。RF电路305可以包括用于执行这些功能的已知电路，RF电路305包括但不限于天线系统、RF收发机、一个或多个放大器、调谐器、一个或多个振荡器、数字信号处理器、编译码器(COder-DECoder，CODEC)芯片组、用户标识模块(Subscriber Identity Module，SIM)等等。
音频电路306，主要设置为从外设接口303接收音频数据，将该音频数据转换为电信号，并且将该电信号发送给扬声器311。
扬声器311，设置为将手机通过RF电路305从无线网络接收的语音信号，还原为声音并向用户播放该声音。
电源管理芯片308，设置为为CPU302、I/O子系统及外设接口所连接的硬件进行供电及电 源管理。
本申请实施例提供的终端设备，可以提高语音验证的准确性。
上述实施例中提供的语音处理装置、存储介质及终端设备可执行本公开任意实施例所提供的语音处理方法，具备执行该方法相应的功能模块和有益效果。未在上述实施例中描述的技术细节，可参见本公开任意实施例所提供的语音处理方法。
Claims (20)
- 一种语音处理方法，包括：获取第一待验证语音信息，并对所述第一待验证语音信息进行第一验证；如果所述第一验证失败，则启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式；接收所述第二验证的验证信息，根据所述验证信息判断所述第一待验证语音信息是否通过所述第二验证；如果所述第一待验证语音信息通过所述第二验证，则将所述第一待验证语音信息确定为预存语音信息，并将所述预存语音信息存储至识别库。
- 如权利要求1所述的方法，其中，所述将所述第一待验证语音信息确定为预存语音信息，并将所述预存语音信息存储至识别库之后，还包括：获取第二待验证语音信息，根据所述识别库中的预存语音信息对所述第二待验证语音信息进行验证；其中，所述预存语音信息为至少一个。
- 如权利要求2所述的方法，其中，所述根据所述识别库中的语音信息对第二待验证语音信息进行验证包括：将所述第二待验证语音信息和所述预存语音信息进行比对；如果任一所述预存语音信息和所述第二待验证语音信息的比对结果符合预设条件，则确定所述第二语音信息通过验证。
- 如权利要求3所述的方法，其中，所述将所述第一待验证语音信息确定为预存语音信息，并将所述预存语音信息存储至识别库包括：将所述第一待验证语音信息确定为预存语音信息；对所述预存语音信息进行预处理，得到所述预存语音信息的特征参数；其中，所述特征参数包括体现声音特质的特征参数；将所述预存语音信息的特征参数存储至识别库。
- 如权利要求4所述的方法，其中，所述将所述第二待验证语音信息和所述预存语音信息进行比对包括：对所述第二待验证语音信息进行预处理，得到所述第二待验证语音信息的特征参数；计算所述第二待验证语音信息的特征参数与所述预存语音信息的特征参数的欧氏距离；所述如果任一所述预存语音信息和所述第二待验证语音信息的比对结果符合预设条件，则确定所述第二待验证语音信息通过验证，包括：如果任一所述预存语音信息的特征参数和所述第二待验证语音信息的特征参数的欧氏距 离小于预设阈值，则确定所述第二语音信息通过验证。
- 如权利要求3所述的方法，在所述第一待验证语音信息通过所述第二验证之后，还包括：获取与所述第一待验证语音信息对应的失败类别信息；所述将所述第一待验证语音信息确定为预存语音信息，并将所述预存语音信息存储至识别库包括：将所述第一待验证语音信息确定为预存语音信息，并将所述预存语音信息和所述失败类别信息存储至识别库；在确定第二待验证语音信息通过验证之后，还包括：从所述识别库中获取与目标语音信息对应的失败类别信息，其中，所述目标语音信息为符合预设条件的比对结果对应的预设语音信息。
- 如权利要求1至5任一项所述的方法，在所述第一待验证语音信息通过所述第二验证之后，还包括：获取与所述第一待验证语音信息对应的失败类别信息；将所述第一待验证语音信息和所述失败类别信息发送至后台服务器。
- 根据权利要求1-7任一项所述的方法，其中，所述第二验证包括指纹验证、密码验证、图形验证、人脸识别验证和虹膜识别验证中的至少一种。
- 根据权利要求6所述的方法，其中，在所述从所述识别库中获取与目标语音信息对应的失败类别信息之后，还包括：根据所述目标语音信息对应的失败类别信息推送与所述失败类别信息对应的服务信息。
- 一种语音处理装置，包括：第一语音验证模块，设置为获取第一待验证语音信息，并对所述第一待验证语音信息进行第一验证；验证启动模块，设置为在所述第一验证失败时，启动第二验证；其中，所述第二验证包括除语音验证以外的验证方式；验证模块，设置为接收所述第二验证的验证信息，根据所述验证信息判断所述第一待验证语音信息是否通过所述第二验证；语音存储模块，设置为在所述第一待验证语音信息通过所述第二验证时，将所述第一待验证语音信息确定为预存语音信息，并将所述预存语音信息存储至识别库。
- 根据权利要求10所述的装置，还包括：第二语音验证模块，设置为在所述将所述第 一待验证语音信息确定为预存语音信息，并将所述预存语音信息存储至识别库之后：获取第二待验证语音信息，根据所述识别库中的预存语音信息对所述第二待验证语音信息进行验证；其中，所述预存语音信息为至少一个。
- 根据权利要求11所述的装置，其中，所述验证模块包括：比对单元，设置为将所述第二待验证语音信息和所述预存语音信息进行比对；验证单元，设置为如果任一所述预存语音信息和所述第二待验证语音信息的比对结果符合预设条件，则确定所述第二语音信息通过验证。
- 根据权利要求12所述的装置，其中，所述语音存储模块包括：确定单元，设置为将所述第一待验证语音信息确定为预存语音信息；预处理单元，设置为对所述预存语音信息进行预处理，得到所述预存语音信息的特征参数；其中，所述特征参数包括体现声音特质的特征参数；存储单元，设置为将所述预存语音信息的特征参数存储至识别库。
- 根据权利要求13所述的装置，其中，所述比对单元包括：预处理子单元，设置为对所述第二待验证语音信息进行预处理，得到所述第二待验证语音信息的特征参数；计算子单元，设置为计算所述第二待验证语音信息的特征参数与所述预存语音信息的特征参数的欧氏距离；所述验证单元是设置为：如果任一所述预存语音信息的特征参数和所述第二待验证语音信息的特征参数的欧氏距离小于预设阈值，则确定所述第二语音信息通过验证。
- 根据权利要求12所述的装置，还包括：类别信息获取模块，设置为在所述第一待验证语音信息通过所述第二验证之后，获取与所述第一待验证语音信息对应的失败类别信息；语音存储模块是设置为将所述第一待验证语音信息确定为预存语音信息，并将所述预存语音信息和所述失败类别信息存储至识别库；所述装置还包括：类别信息确定模块，设置为在确定第二待验证语音信息通过验证之后，还包括：从所述识别库中获取与目标语音信息对应的失败类别信息，其中，所述目标语音信息为符合预设条件的比对结果对应的预设语音信息。
- 根据权利要求10-14任一项所述的装置，还包括：类别信息获取模块，设置为在所述第一待验证语音信息通过所述第二验证之后，获取与所述第一待验证语音信息对应的失败类别信息；后台发送模块，设置为将所述第一待验证语音信息和所述失败类别信息发送至后台服务器。
- 根据权利要求10-16任一项所述的装置，其中，所述第二验证包括指纹验证、密码验证、图形验证、人脸识别验证和虹膜识别验证中的至少一种。
- 根据权利要求15所述的装置，还包括：信息推送模块，设置为：在所述从所述识别库中获取与目标语音信息对应的失败类别信息之后，根据所述目标语音信息对应的失败类别信息推送与所述失败类别信息对应的服务信息。
- 一种计算机可读存储介质，存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1-9中任一项所述的语音处理方法。
- 一种终端设备，包括存储器，处理器及存储在存储器上并可在处理器运行的计算机程序，所述处理器执行所述计算机程序时实现如权利要求1-9任一项所述的语音处理方法。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711339174.0A CN109960910B (zh) | 2017-12-14 | 2017-12-14 | 语音处理方法、装置、存储介质及终端设备 |
| CN201711339174.0 | 2017-12-14 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2019114507A1 true WO2019114507A1 (zh) | 2019-06-20 |
Family
ID=66818905
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2018/116587 WO2019114507A1 (zh) | 2017-12-14 | 2018-11-21 | 语音处理方法、装置、存储介质及终端设备 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109960910B (zh) |
| WO (1) | WO2019114507A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112351047A (zh) * | 2021-01-07 | 2021-02-09 | 北京远鉴信息技术有限公司 | 基于双引擎的声纹身份认证方法、装置、设备及存储介质 |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113254901B (zh) * | 2021-06-11 | 2021-10-01 | 长扬科技（北京）有限公司 | 一种数据安全访问方法和装置 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104915582A (zh) * | 2015-05-28 | 2015-09-16 | 努比亚技术有限公司 | 解锁方法及装置 |
| CN105512535A (zh) * | 2016-01-08 | 2016-04-20 | 广东德生科技股份有限公司 | 一种用户认证的方法及装置 |
| CN106446631A (zh) * | 2016-09-20 | 2017-02-22 | 深圳市金立通信设备有限公司 | 一种查看锁屏通知信息的方法及终端 |
| CN106961418A (zh) * | 2017-02-08 | 2017-07-18 | 北京捷通华声科技股份有限公司 | 身份认证方法和身份认证系统 |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004305330A (ja) * | 2003-04-03 | 2004-11-04 | Nec Corp | 移動体通信端末装置及び音声情報発生装置 |
| US8467320B2 (en) * | 2005-10-06 | 2013-06-18 | Telecommunication Systems, Inc. | Voice over internet protocol (VoIP) multi-user conferencing |
| CN102306493A (zh) * | 2011-08-18 | 2012-01-04 | 鸿富锦精密工业（深圳）有限公司 | 终端机、语音辨识系统及其语音辨识方法 |
| CN105376196A (zh) * | 2014-08-19 | 2016-03-02 | 深圳市科瑞电子有限公司 | 一种基于人体生物特征的系统登陆方法、装置及终端 |
| CN104538034B (zh) * | 2014-12-31 | 2018-08-28 | 深圳雷柏科技股份有限公司 | 一种语音识别方法及系统 |
| CN106453859B (zh) * | 2016-09-23 | 2019-11-15 | 维沃移动通信有限公司 | 一种语音控制方法及移动终端 |
| CN106503513A (zh) * | 2016-09-23 | 2017-03-15 | 北京小米移动软件有限公司 | 声纹识别方法及装置 |
| CN106653033A (zh) * | 2016-10-28 | 2017-05-10 | 努比亚技术有限公司 | 一种语音解锁装置、终端及方法 |
| CN107068167A (zh) * | 2017-03-13 | 2017-08-18 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | 融合多种端到端神经网络结构的说话人感冒症状识别方法 |
-
2017
- 2017-12-14 CN CN201711339174.0A patent/CN109960910B/zh active Active
-
2018
- 2018-11-21 WO PCT/CN2018/116587 patent/WO2019114507A1/zh active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104915582A (zh) * | 2015-05-28 | 2015-09-16 | 努比亚技术有限公司 | 解锁方法及装置 |
| CN105512535A (zh) * | 2016-01-08 | 2016-04-20 | 广东德生科技股份有限公司 | 一种用户认证的方法及装置 |
| CN106446631A (zh) * | 2016-09-20 | 2017-02-22 | 深圳市金立通信设备有限公司 | 一种查看锁屏通知信息的方法及终端 |
| CN106961418A (zh) * | 2017-02-08 | 2017-07-18 | 北京捷通华声科技股份有限公司 | 身份认证方法和身份认证系统 |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112351047A (zh) * | 2021-01-07 | 2021-02-09 | 北京远鉴信息技术有限公司 | 基于双引擎的声纹身份认证方法、装置、设备及存储介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109960910A (zh) | 2019-07-02 |
| CN109960910B (zh) | 2021-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10079014B2 (en) | Name recognition system | |
| US9767805B2 (en) | Voice recognition method, voice controlling method, information processing method, and electronic apparatus | |
| US11508383B2 (en) | Method for operating content and electronic device for implementing same | |
| WO2018188586A1 (zh) | 一种用户注册方法、装置及电子设备 | |
| US10270736B2 (en) | Account adding method, terminal, server, and computer storage medium | |
| EP3309783B1 (en) | Communication method, and electronic device therefor | |
| US20140149122A1 (en) | Voice control device and voice control method | |
| KR20180074210A (ko) | 전자 장치 및 전자 장치의 음성 인식 방법 | |
| WO2016015687A1 (zh) | 声纹验证方法及装置 | |
| US11164022B2 (en) | Method for fingerprint enrollment, terminal, and non-transitory computer readable storage medium | |
| KR102653450B1 (ko) | 전자 장치의 입력 음성에 대한 응답 방법 및 그 전자 장치 | |
| WO2020088483A1 (zh) | 一种音频控制方法及电子设备 | |
| CN108766438A (zh) | 人机交互方法、装置、存储介质及智能终端 | |
| US20200380971A1 (en) | Method of activating voice assistant and electronic device with voice assistant | |
| WO2019114507A1 (zh) | 语音处理方法、装置、存储介质及终端设备 | |
| WO2017215186A1 (zh) | 一种安全登录方法和装置、存储介质 | |
| US20220392448A1 (en) | Device for processing user voice input | |
| CN111684521A (zh) | 用于说话者识别的处理语音信号方法及实现其的电子装置 | |
| KR102501083B1 (ko) | 음성 인식 방법 및 이를 사용하는 전자 장치 | |
| WO2017166133A1 (zh) | 一种终端的语音解锁方法及终端 | |
| WO2022199405A1 (zh) | 一种语音控制方法和装置 | |
| US20200204365A1 (en) | Apparatus, system and method for application-specific biometric processing in a computer system | |
| US11244676B2 (en) | Apparatus for processing user voice input | |
| WO2022017152A1 (zh) | 资源转移方法、装置、计算机设备和存储介质 | |
| TW202046061A (zh) | 啟動語音助理的方法及具有語音助理的電子裝置 |
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: 18888267 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 18888267 Country of ref document: EP Kind code of ref document: A1 |