Far-field voice wake-up method and system of application, terminal and storage medium

Abstract

The invention discloses a far-field voice wake-up method and system for an application, a terminal and a storage medium, and the method comprises the steps: obtaining recording data collected by a target device according to an application wake-up instruction when the application wake-up instruction sent by a user is detected, and carrying out the enhanced noise reduction processing of the recording data, and obtaining target recording data; performing type confirmation on the target device to obtain device type information, and performing architecture decision evaluation according to the device type information to obtain a target wake-up architecture; and performing data processing on the target recording data according to the target wake-up architecture to obtain an application wake-up signal, and waking up a corresponding voice application according to the application wake-up signal. The optimal wake-up support structure can be quickly evaluated by introducing the decision module, and the reuse of the wake-up algorithm module is realized by customizing a unified universal interface.

Description

Technical Field

[0001] The present invention relates to the field of voice wake-up processing technology, and in particular to an applied far-field voice wake-up method, system, terminal and computer-readable storage medium. Background Art

[0002] With the rapid development of smart devices, far-field voice wake-up of applications has become a standard function of smart devices. There are two existing far-field voice wake-up solution architectures, namely the wake-up support architecture that follows the system and the wake-up support architecture that follows the APP. For the wake-up support architecture that follows the system, the recording data after enhanced noise reduction is obtained at the hardware abstraction layer of the system. The wake-up signal obtained after the wake-up algorithm module is passed to the voice APP through a shared socket (network programming interface) by the hardware abstraction layer. If there is an algorithm update, a system upgrade is usually required, resulting in insufficient flexibility. However, because it is a bottom-level solution, it has better real-time performance and the wake-up function effect is usually better. For the wake-up support architecture that follows the APP, that is, the APP obtains the noise-reduced enhanced data through the standard recording interface, and then calls the wake-up algorithm module through the custom jni interface. The wake-up signal is given to the APP through the custom callback interface. If there is an algorithm update, only the APP needs to be updated. This solution will have a faster adaptation speed, and the speed of subsequent product iterations and launches will also be faster.

[0003] At present, the two existing far-field voice wake-up solution architectures are usually two sets of program packages, which need to be customized into one of them according to the device and needs. This is not conducive to the evaluation of the solution, resulting in the inability to quickly switch to verify the effect, and thus unable to decide on the optimal adaptation method. In addition, due to the poor expansion function of the system-level support architecture, it is impossible to quickly and lightweight migrate from one device to another. The solution that follows the APP only has a wake-up algorithm, and the enhanced noise reduction algorithm depends on the device itself, which is prone to poor results.

[0004] Therefore, the prior art still needs to be improved and developed. Summary of the invention

[0005] The main purpose of the present invention is to provide an applied far-field voice wake-up method, system, terminal and storage medium, aiming to solve the problem that the prior art cannot quickly decide on the optimal adaptation method and has poor wake-up effect.

[0006] To achieve the above object, the present invention provides an application far-field voice wake-up method, and the application far-field voice wake-up method comprises the following steps:

[0007] When it is detected that the user issues an application wake-up instruction, the recording data collected by the target device is obtained according to the application wake-up instruction, and the recording data is enhanced and noise-reduced to obtain the target recording data;

[0008] Confirming the type of the target device to obtain device type information, and performing architecture decision evaluation based on the device type information to obtain a target wake-up architecture;

[0009] The target recording data is processed according to the target wake-up architecture to obtain an application wake-up signal, and the corresponding voice application is woken up according to the application wake-up signal.

[0010] Optionally, the far-field voice wake-up method of the application, wherein when it is detected that the user issues an application wake-up instruction, the recording data collected by the target device is obtained according to the application wake-up instruction, and the recording data is enhanced and noise-reduced to obtain the target recording data, specifically including:

[0011] When it is detected that the user issues an application wake-up instruction, the recording data collected by the array microphone of the target device is obtained according to the application wake-up instruction, wherein the recording data includes microphone data and echo data;

[0012] Acquire a data processing format of an enhanced noise reduction algorithm, and adjust the data format of the recording data according to the data processing format to obtain recording adjustment data;

[0013] The signal-to-noise ratio of the recording adjustment data is optimized according to the enhanced noise reduction algorithm to obtain target recording data.

[0014] Optionally, the far-field voice wake-up method of the application, wherein the type of the target device is confirmed to obtain device type information, and an architecture decision evaluation is performed according to the device type information to obtain a target wake-up architecture, specifically including:

[0015] Confirming the type of the target device to obtain device type information, and determining whether the target device provides system source code according to the device type information to obtain a first determination result;

[0016] If the first determination result is that the target device provides a system source code, determining whether the target device has an enhanced noise reduction algorithm according to the first determination result to obtain a second determination result;

[0017] If the second determination result is that the target device has an enhanced noise reduction algorithm, determining whether the target device has a high sound quality requirement according to the second determination result to obtain a third determination result;

[0018] If the third determination result is that the target device has a high sound quality requirement, determining whether the target device supports system upgrade according to the third determination result to obtain a fourth determination result;

[0019] If the fourth determination result is that the target device supports system upgrade, determining whether the target device requires lightweight adaptation according to the fourth determination result to obtain a fifth determination result;

[0020] If the fifth determination result is that the target device does not need lightweight adaptation, the target wake-up architecture is determined to be a wake-up support architecture of the follower system.

[0021] Optionally, the far-field voice wake-up method of the application, wherein the type of the target device is confirmed to obtain device type information, and an architecture decision evaluation is performed according to the device type information to obtain a target wake-up architecture, and then further includes:

[0022] If the first determination result is that the target device does not provide the system source code, or the second determination result is that the target device does not have an enhanced noise reduction algorithm, or the third determination result is that the target device does not have high sound quality requirements, or the fourth determination result is that the target device does not support system upgrades, or if the fifth determination result is that the target device requires lightweight adaptation, then the target wake-up architecture is determined to be a wake-up support architecture that follows the APP.

[0023] Optionally, the far-field voice wake-up method of the application, wherein the data processing of the target recording data according to the target wake-up architecture to obtain an application wake-up signal, and waking up the corresponding voice application according to the application wake-up signal, specifically includes:

[0024] If the target wake-up architecture is a wake-up support architecture of the following system, feature analysis is performed on the target recording data according to the wake-up algorithm;

[0025] If the target recording data meets the wake-up word feature, generating an application wake-up signal according to the target recording data;

[0026] The transmission interface of the application wake-up signal is determined according to the target wake-up architecture to obtain a network programming interface, and the application wake-up signal is sent to the corresponding voice application according to the network programming interface to wake up the voice application.

[0027] Optionally, the far-field voice wake-up method of the application, wherein the data processing of the target recording data according to the target wake-up architecture to obtain an application wake-up signal, and waking up the corresponding voice application according to the application wake-up signal, specifically includes:

[0028] If the target wake-up architecture is a wake-up support architecture following an APP, a corresponding data filling interface is formulated according to the target wake-up architecture, and a wake-up algorithm is extracted according to the data filling interface;

[0029] Performing feature analysis on the target recording data according to the wake-up algorithm, and generating an application wake-up signal according to the target recording data if the target recording data meets the wake-up word feature;

[0030] A corresponding wake-up callback interface is formulated according to the application wake-up signal, and the application wake-up signal is sent to the corresponding voice application according to the wake-up callback interface to wake up the voice application.

[0031] Optionally, the far-field voice wake-up method of the application, wherein the target recording data is processed according to the target wake-up architecture to obtain an application wake-up signal, and the corresponding voice application is woken up according to the application wake-up signal, and then further includes:

[0032] If no wake-up feedback of the voice application is received within a preset time, it is determined that the voice application wake-up has failed, and a re-wake-up instruction is generated, and the voice application is re-waken up according to the re-wake-up instruction.

[0033] Optionally, in the far-field voice wake-up method of the application, the far-field voice wake-up system of the application includes:

[0034] The data processing module is used to obtain the recording data collected by the target device according to the application wake-up instruction when detecting that the user issues an application wake-up instruction, and perform enhanced noise reduction processing on the recording data to obtain target recording data;

[0035] An architecture evaluation module is used to confirm the type of the target device to obtain device type information, and to perform architecture decision evaluation based on the device type information to obtain a target wake-up architecture;

[0036] The application wake-up module is used to process the target recording data according to the target wake-up architecture, obtain an application wake-up signal, and wake up the corresponding voice application according to the application wake-up signal.

[0037] In addition, to achieve the above-mentioned purpose, the present invention also provides a terminal, wherein the terminal includes: a memory, a processor, and a far-field voice wake-up program of an application stored in the memory and runnable on the processor, and when the far-field voice wake-up program of the application is executed by the processor, the steps of the far-field voice wake-up method of the application as described above are implemented.

[0038] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a far-field voice wake-up program of an application, and when the far-field voice wake-up program of the application is executed by a processor, the steps of the far-field voice wake-up method of the application as described above are implemented.

[0039] In the present invention, when it is detected that the user issues an application wake-up instruction, the recording data collected by the target device is obtained according to the application wake-up instruction, and the recording data is enhanced and noise-reduced to obtain the target recording data; the type of the target device is confirmed to obtain device type information, and an architecture decision evaluation is performed based on the device type information to obtain a target wake-up architecture; the target recording data is processed according to the target wake-up architecture to obtain an application wake-up signal, and the corresponding voice application is awakened according to the application wake-up signal. The present invention can quickly evaluate the best wake-up support structure by introducing a decision module, and realizes the reuse of wake-up algorithm modules by customizing a unified universal interface. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a flow chart of a preferred embodiment of the far-field voice wake-up method applied by the present invention;

[0041] Figure 2 Schematic diagram of the overall architecture of the far-field voice wake-up method applied by the present invention;

[0042] Figure 3 It is a flow chart of architecture evaluation performed by a decision module of a preferred embodiment of the far-field voice wake-up method applied by the present invention;

[0043] Figure 4 It is a structural diagram of a preferred embodiment of a far-field voice wake-up system applied in the present invention;

[0044] Figure 5 It is a schematic diagram of the operating environment of a preferred embodiment of the terminal of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0046] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0047] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0048] The far-field voice wake-up method of the application described in the preferred embodiment of the present invention is as follows: Figure 1 As shown, the far-field voice wake-up method of the application includes the following steps:

[0049] Step S10: When it is detected that the user issues an application wake-up instruction, the recording data collected by the target device is obtained according to the application wake-up instruction, and the recording data is enhanced and noise reduced to obtain the target recording data.

[0050] The step S10 comprises:

[0051] Step S11: when it is detected that the user issues an application wake-up instruction, the recording data collected by the array microphone of the target device is obtained according to the application wake-up instruction, wherein the recording data includes microphone data and echo data;

[0052] Step S12, obtaining a data processing format of the enhanced noise reduction algorithm, and adjusting the data format of the recording data according to the data processing format to obtain recording adjustment data;

[0053] Step S13: Optimizing the signal-to-noise ratio of the recording adjustment data according to the enhanced noise reduction algorithm to obtain target recording data.

[0054] Specifically, the two existing far-field voice wake-up solution architectures are usually two sets of program packages, which need to be customized into one of them according to the device and needs, which is not conducive to the evaluation of the solution, resulting in the inability to quickly switch to verify the effect, and thus unable to decide on the optimal adaptation method. In addition, due to the poor expansion function of the system-level support architecture, it is impossible to quickly migrate from one device to another in a lightweight manner. The solution that follows the APP only has a wake-up algorithm, and the enhanced noise reduction algorithm depends on the device itself, which is prone to poor results. In an embodiment of the present invention, an application of a far-field voice wake-up method is proposed to solve the above-mentioned problems. The specific processing process is as follows: Figure 2As shown, when it is detected that the user issues an application wake-up instruction, the recording data collected by the array microphone (for example, 2mic and 4mic) of the target device is obtained according to the application wake-up instruction, wherein the recording data includes microphone data and echo data.

[0055] Afterwards, the recording data is packaged into data corresponding to the data format required by the enhanced noise reduction algorithm through alsa (i.e., the sound card driver). Specifically, the data processing format of the enhanced noise reduction algorithm is obtained, and the data format of the recording data is formatted according to the data processing format to obtain recording adjustment data. Taking a 2mic array microphone as an example, the recording adjustment data usually includes 2-channel microphone data and 2-channel recollection data, and the microphone data and the recollection data are consistent in time; then, the signal-to-noise ratio of the recording adjustment data is optimized according to the enhanced noise reduction algorithm to obtain target recording data. After being processed by the enhanced noise reduction algorithm, effective voice data is highlighted, and other external noises and audio and video resource noises played by the device itself are mostly eliminated. The signal-to-noise ratio of the obtained target recording data is usually less than 10db, and the purpose of performing enhanced noise reduction algorithm processing is to meet the subsequent wake-up algorithm.

[0056] Step S20: confirm the type of the target device to obtain device type information, and perform architecture decision evaluation based on the device type information to obtain a target wake-up architecture.

[0057] The step S20 comprises:

[0058] Step S21, confirming the type of the target device to obtain device type information, and judging whether the target device provides system source code according to the device type information to obtain a first judgment result;

[0059] Step S22: If the first determination result is that the target device provides a system source code, determine whether the target device has an enhanced noise reduction algorithm according to the first determination result to obtain a second determination result;

[0060] Step S23: If the second determination result is that the target device has an enhanced noise reduction algorithm, determine whether the target device has a high sound quality requirement according to the second determination result to obtain a third determination result;

[0061] Step S24: If the third determination result is that the target device has a high sound quality requirement, determining whether the target device supports system upgrade according to the third determination result to obtain a fourth determination result;

[0062] Step S25: If the fourth determination result is that the target device supports system upgrade, determine whether the target device needs lightweight adaptation according to the fourth determination result to obtain a fifth determination result;

[0063] Step S26: If the fifth determination result is that the target device does not require lightweight adaptation, then determining the target wake-up architecture to be a wake-up support architecture of the following system.

[0064] Specifically, after obtaining the target recording data, it is necessary to decide on the optimal wake-up support architecture. In the embodiment of the present invention, the appropriate wake-up support architecture is quickly evaluated by combining the device type, quality requirements, and product iteration requirements. The specific processing process is as follows: Figure 3 As shown, the type of the target device is confirmed by the decision module to obtain device type information (for example, types such as television, mobile phone and tablet), and whether the target device provides system source code is determined according to the device type information to obtain a first determination result; if the first determination result is that the target device provides system source code, then whether the target device has an enhanced noise reduction algorithm is determined according to the first determination result to obtain a second determination result; if the second determination result is that the target device has an enhanced noise reduction algorithm, then whether the target device has high sound quality requirements is determined according to the second determination result to obtain a third determination result; if the third determination result is that the target device has high sound quality requirements, then whether the target device supports system upgrades is determined according to the third determination result to obtain a fourth determination result; if the fourth determination result is that the target device supports system upgrades, then whether the target device needs lightweight adaptation is determined according to the fourth determination result to obtain a fifth determination result; if the fifth determination result is that the target device does not need lightweight adaptation, then the target wake-up architecture is determined to be a wake-up support architecture that follows the system.

[0065] Furthermore, if the first judgment result is that the target device does not provide the system source code, or the second judgment result is that the target device does not have an enhanced noise reduction algorithm, or the third judgment result is that the target device does not have high sound quality requirements, or the fourth judgment result is that the target device does not support system upgrades, or if the fifth judgment result is that the target device requires lightweight adaptation, then the target wake-up architecture is determined to be a wake-up support architecture that follows the APP.

[0066] Step S30: Process the target recording data according to the target wake-up architecture to obtain an application wake-up signal, and wake up the corresponding voice application according to the application wake-up signal.

[0067] Specifically, after determining the target wake-up architecture, the target recording data is processed according to the target wake-up architecture. In the embodiment of the present invention, different processing methods are used for different target wake-up architectures, such as Figure 2 As shown, if the target wake-up architecture is a wake-up support architecture that follows the system, the target recording data is directly processed according to the wake-up algorithm at the bottom layer. Specifically, the target recording data is feature analyzed according to the wake-up algorithm, and by comparing the features of a specific wake-up word, if the target recording data meets the wake-up word features, an application wake-up signal is generated according to the target recording data; then, the transmission method of the wake-up signal is determined by the result cached by the decision module, thereby avoiding the functional redundancy of repeatedly sending the wake-up signal, and the result cached by the decision module is a state. For example, if the state is STATUS_APP, it indicates that the current wake-up signal will be sent by the customized wake-up callback interface; if the state is STATUS_SYS, it indicates that the wake-up signal will be sent by the socket interface (i.e., the network programming interface); specifically, the transmission interface of the application wake-up signal is determined according to the target wake-up architecture to obtain a socket interface, and the application wake-up signal is sent to the corresponding voice application according to the network programming interface to wake up the voice application.

[0068] If the target wake-up architecture is a wake-up support architecture that follows the APP, and the corresponding target recording data is in the voice application, then the target recording data needs to be extracted before processing the target recording data. Specifically, a corresponding data filling interface is formulated according to the target wake-up architecture, and a wake-up algorithm is extracted according to the data filling interface, and feature analysis is performed on the target recording data according to the wake-up algorithm. If the target recording data meets the wake-up word feature, an application wake-up signal is generated according to the target recording data; thereafter, a corresponding wake-up callback interface is formulated according to the application wake-up signal, and the application wake-up signal is sent to the corresponding voice application according to the wake-up callback interface to wake up the voice application.

[0069] Furthermore, after the application wake-up signal is sent to the corresponding voice application, it is necessary to determine whether the voice application is successfully awakened. If a wake-up feedback from the voice application is received within a preset time, for example, the voice application will respond with a human voice, such as "I'm here", or render a specific animation, such as a text prompt box with the words "I'm here", then it is determined that the voice application has failed to wake up; if no wake-up feedback from the voice application is received within the preset time, then it is determined that the voice application has failed to wake up, and a re-wake-up instruction is generated, and the voice application is re-awakened according to the re-wake-up instruction.

[0070] Furthermore, if Figure 3As shown, based on the above-mentioned far-field voice wake-up method of the application, the present invention also provides a far-field voice wake-up system of the application, wherein the far-field voice wake-up system of the application includes:

[0071] The data processing module 51 is used to obtain the recording data collected by the target device according to the application wake-up instruction when it is detected that the user issues an application wake-up instruction, and perform enhanced noise reduction processing on the recording data to obtain target recording data;

[0072] The architecture evaluation module 52 is used to confirm the type of the target device to obtain device type information, and to perform architecture decision evaluation based on the device type information to obtain a target wake-up architecture;

[0073] The application wake-up module 53 is used to process the target recording data according to the target wake-up architecture, obtain an application wake-up signal, and wake up the corresponding voice application according to the application wake-up signal.

[0074] Furthermore, if Figure 4 As shown, based on the far-field voice wake-up method of the above application, the present invention also provides a terminal accordingly, and the terminal includes a processor 10, a memory 20 and a display 30. Figure 4 Only some components of the terminal are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0075] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory of the terminal. In other embodiments, the memory 20 may also be an external storage device of the terminal, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the terminal. Furthermore, the memory 20 may also include both an internal storage unit of the terminal and an external storage device. The memory 20 is used to store application software and various types of data installed on the terminal, such as the program code of the installation terminal. The memory 20 can also be used to temporarily store data that has been output or is to be output. In one embodiment, the memory 20 stores an applied far-field voice wake-up program 40, which can be executed by the processor 10, thereby realizing the far-field voice wake-up method applied in the present application.

[0076] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run the program code or process data stored in the memory 20, such as executing the far-field voice wake-up method of the application.

[0077] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 30 is used to display information on the terminal and to display a visual user interface. The components of the terminal communicate with each other via a system bus.

[0078] In one embodiment, when the processor 10 executes the far-field voice wake-up program 40 applied in the memory 20, the following steps are implemented:

[0079] When it is detected that the user issues an application wake-up instruction, the recording data collected by the target device is obtained according to the application wake-up instruction, and the recording data is enhanced and noise-reduced to obtain the target recording data;

[0080] Confirming the type of the target device to obtain device type information, and performing architecture decision evaluation based on the device type information to obtain a target wake-up architecture;

[0081] The target recording data is processed according to the target wake-up architecture to obtain an application wake-up signal, and the corresponding voice application is woken up according to the application wake-up signal.

[0082] Among them, when it is detected that the user issues an application wake-up instruction, the recording data collected by the target device is obtained according to the application wake-up instruction, and the recording data is enhanced and noise-reduced to obtain the target recording data, which specifically includes:

[0083] When it is detected that the user issues an application wake-up instruction, the recording data collected by the array microphone of the target device is obtained according to the application wake-up instruction, wherein the recording data includes microphone data and echo data;

[0084] Acquire a data processing format of an enhanced noise reduction algorithm, and adjust the data format of the recording data according to the data processing format to obtain recording adjustment data;

[0085] The signal-to-noise ratio of the recording adjustment data is optimized according to the enhanced noise reduction algorithm to obtain target recording data.

[0086] The step of confirming the type of the target device to obtain device type information, and performing architecture decision evaluation based on the device type information to obtain a target wake-up architecture specifically includes:

[0087] Confirming the type of the target device to obtain device type information, and determining whether the target device provides system source code according to the device type information to obtain a first determination result;

[0088] If the first determination result is that the target device provides a system source code, determining whether the target device has an enhanced noise reduction algorithm according to the first determination result to obtain a second determination result;

[0089] If the second determination result is that the target device has an enhanced noise reduction algorithm, determining whether the target device has a high sound quality requirement according to the second determination result to obtain a third determination result;

[0090] If the third determination result is that the target device has a high sound quality requirement, determining whether the target device supports system upgrade according to the third determination result to obtain a fourth determination result;

[0091] If the fourth determination result is that the target device supports system upgrade, determining whether the target device requires lightweight adaptation according to the fourth determination result to obtain a fifth determination result;

[0092] If the fifth determination result is that the target device does not need lightweight adaptation, the target wake-up architecture is determined to be a wake-up support architecture of the follower system.

[0093] The method further comprises: confirming the type of the target device to obtain device type information, and performing architecture decision evaluation according to the device type information to obtain a target wake-up architecture. Then, the method further comprises:

[0094] If the first determination result is that the target device does not provide the system source code, or the second determination result is that the target device does not have an enhanced noise reduction algorithm, or the third determination result is that the target device does not have high sound quality requirements, or the fourth determination result is that the target device does not support system upgrades, or if the fifth determination result is that the target device requires lightweight adaptation, then the target wake-up architecture is determined to be a wake-up support architecture that follows the APP.

[0095] The processing of the target recording data according to the target wake-up architecture to obtain an application wake-up signal, and waking up the corresponding voice application according to the application wake-up signal specifically includes:

[0096] If the target wake-up architecture is a wake-up support architecture of the following system, feature analysis is performed on the target recording data according to the wake-up algorithm;

[0097] If the target recording data meets the wake-up word feature, generating an application wake-up signal according to the target recording data;

[0098] The transmission interface of the application wake-up signal is determined according to the target wake-up architecture to obtain a network programming interface, and the application wake-up signal is sent to the corresponding voice application according to the network programming interface to wake up the voice application.

[0099] The processing of the target recording data according to the target wake-up architecture to obtain an application wake-up signal, and waking up the corresponding voice application according to the application wake-up signal specifically includes:

[0100] If the target wake-up architecture is a wake-up support architecture following an APP, a corresponding data filling interface is formulated according to the target wake-up architecture, and a wake-up algorithm is extracted according to the data filling interface;

[0101] Performing feature analysis on the target recording data according to the wake-up algorithm, and generating an application wake-up signal according to the target recording data if the target recording data meets the wake-up word feature;

[0102] A corresponding wake-up callback interface is formulated according to the application wake-up signal, and the application wake-up signal is sent to the corresponding voice application according to the wake-up callback interface to wake up the voice application.

[0103] The target recording data is processed according to the target wake-up architecture to obtain an application wake-up signal, and a corresponding voice application is awakened according to the application wake-up signal, and then the following is further included:

[0104] If no wake-up feedback of the voice application is received within a preset time, it is determined that the voice application wake-up has failed, and a re-wake-up instruction is generated, and the voice application is re-waken up according to the re-wake-up instruction.

[0105] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a far-field voice wake-up program for an application, and when the far-field voice wake-up program for the application is executed by a processor, the steps of the far-field voice wake-up method for the application as described above are implemented.

[0106] In summary, the present invention provides a far-field voice wake-up method, system, terminal and storage medium for an application, the method comprising: when it is detected that a user issues an application wake-up instruction, the recording data collected by the target device is obtained according to the application wake-up instruction, and the recording data is enhanced and noise-reduced to obtain the target recording data; the type of the target device is confirmed to obtain device type information, and an architecture decision evaluation is performed based on the device type information to obtain a target wake-up architecture; data processing is performed on the target recording data according to the target wake-up architecture to obtain an application wake-up signal, and the corresponding voice application is woken up according to the application wake-up signal. The present invention can quickly evaluate the optimal wake-up support structure by introducing a decision module, and realizes the reuse of wake-up algorithm modules by customizing a unified universal interface.

[0107] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or terminal including the element.

[0108] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the program can be stored in a computer-readable storage medium that can be read by a computer, and the program can include the processes of the above-mentioned method embodiments when executed. The computer-readable storage medium can be a memory, a disk, an optical disk, etc.

[0109] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A far-field voice wake-up method for application, characterized in that: The far-field voice wake-up method of the application includes: When it is detected that the user issues an application wake-up instruction, the recording data collected by the target device is obtained according to the application wake-up instruction, and the recording data is enhanced and noise-reduced to obtain the target recording data; Confirming the type of the target device to obtain device type information, and performing architecture decision evaluation based on the device type information to obtain a target wake-up architecture; The target recording data is processed according to the target wake-up architecture to obtain an application wake-up signal, and the corresponding voice application is woken up according to the application wake-up signal.

2. The far-field voice wake-up method of claim 1, characterized in that: When it is detected that the user issues an application wake-up instruction, the recording data collected by the target device is obtained according to the application wake-up instruction, and the recording data is enhanced and noise-reduced to obtain the target recording data, which specifically includes: When it is detected that the user issues an application wake-up instruction, the recording data collected by the array microphone of the target device is obtained according to the application wake-up instruction, wherein the recording data includes microphone data and echo data; Acquire a data processing format of an enhanced noise reduction algorithm, and adjust the data format of the recording data according to the data processing format to obtain recording adjustment data; The signal-to-noise ratio of the recording adjustment data is optimized according to the enhanced noise reduction algorithm to obtain target recording data.

3. The far-field voice wake-up method of claim 1, characterized in that: The step of confirming the type of the target device to obtain device type information, and performing architecture decision evaluation based on the device type information to obtain a target wake-up architecture specifically includes: Confirming the type of the target device to obtain device type information, and determining whether the target device provides system source code according to the device type information to obtain a first determination result; If the first determination result is that the target device provides a system source code, determining whether the target device has an enhanced noise reduction algorithm according to the first determination result to obtain a second determination result; If the second determination result is that the target device has an enhanced noise reduction algorithm, determining whether the target device has a high sound quality requirement according to the second determination result to obtain a third determination result; If the third determination result is that the target device has a high sound quality requirement, determining whether the target device supports system upgrade according to the third determination result to obtain a fourth determination result; If the fourth determination result is that the target device supports system upgrade, determining whether the target device requires lightweight adaptation according to the fourth determination result to obtain a fifth determination result; If the fifth determination result is that the target device does not need lightweight adaptation, the target wake-up architecture is determined to be a wake-up support architecture of the follower system.

4. The far-field voice wake-up method of claim 3, characterized in that: The method further includes: confirming the type of the target device to obtain device type information, and performing architecture decision evaluation according to the device type information to obtain a target wake-up architecture. If the first determination result is that the target device does not provide the system source code, or the second determination result is that the target device does not have an enhanced noise reduction algorithm, or the third determination result is that the target device does not have high sound quality requirements, or the fourth determination result is that the target device does not support system upgrades, or if the fifth determination result is that the target device requires lightweight adaptation, then the target wake-up architecture is determined to be a wake-up support architecture that follows the APP.

5. The far-field voice wake-up method according to claim 4, characterized in that: The processing of the target recording data according to the target wake-up architecture to obtain an application wake-up signal, and waking up the corresponding voice application according to the application wake-up signal specifically includes: If the target wake-up architecture is a wake-up support architecture of the following system, feature analysis is performed on the target recording data according to the wake-up algorithm; If the target recording data meets the wake-up word feature, generating an application wake-up signal according to the target recording data; The transmission interface of the application wake-up signal is determined according to the target wake-up architecture to obtain a network programming interface, and the application wake-up signal is sent to the corresponding voice application according to the network programming interface to wake up the voice application.

6. The far-field voice wake-up method for application according to claim 4, characterized in that: The processing of the target recording data according to the target wake-up architecture to obtain an application wake-up signal, and waking up the corresponding voice application according to the application wake-up signal specifically includes: If the target wake-up architecture is a wake-up support architecture following an APP, a corresponding data filling interface is formulated according to the target wake-up architecture, and a wake-up algorithm is extracted according to the data filling interface; Performing feature analysis on the target recording data according to the wake-up algorithm, and generating an application wake-up signal according to the target recording data if the target recording data meets the wake-up word feature; A corresponding wake-up callback interface is formulated according to the application wake-up signal, and the application wake-up signal is sent to the corresponding voice application according to the wake-up callback interface to wake up the voice application.

7. The far-field voice wake-up method for application according to claim 1, characterized in that: The target recording data is processed according to the target wake-up architecture to obtain an application wake-up signal, and a corresponding voice application is awakened according to the application wake-up signal, and then the following further includes: If no wake-up feedback of the voice application is received within a preset time, it is determined that the voice application wake-up has failed, and a re-wake-up instruction is generated, and the voice application is re-waken up according to the re-wake-up instruction.

8. A far-field voice wake-up system for application, characterized in that: The far-field voice wake-up system of the application includes: The data processing module is used to obtain the recording data collected by the target device according to the application wake-up instruction when detecting that the user issues an application wake-up instruction, and perform enhanced noise reduction processing on the recording data to obtain target recording data; An architecture evaluation module is used to confirm the type of the target device to obtain device type information, and to perform architecture decision evaluation based on the device type information to obtain a target wake-up architecture; The application wake-up module is used to process the target recording data according to the target wake-up architecture, obtain an application wake-up signal, and wake up the corresponding voice application according to the application wake-up signal.

9. A terminal, characterized in that: The terminal includes: a memory, a processor, and a far-field voice wake-up program of an application stored in the memory and executable on the processor. When the far-field voice wake-up program of the application is executed by the processor, the steps of the far-field voice wake-up method of the application as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a far-field voice wake-up program for an application, and when the far-field voice wake-up program for an application is executed by a processor, the steps of the far-field voice wake-up method for an application as described in any one of claims 1 to 7 are implemented.

Images