Wake-up test method and system

Abstract

The embodiment of the invention provides a wake-up test method and system applied to a distributed voice wake-up test scene. According to the method, based on basic parameter information and a first condition in a distributed voice wake-up test scene, a first detection path corresponding to a first device is determined by adopting a first algorithm, and wake-up test information for executing a wake-up test on at least two second devices on the first detection path is acquired; the wake-up test information is used for determining a wake-up test result. By using the method, the false wake-up rate and the probability of false detection and false recording can be effectively reduced on the basis of ensuring the communication efficiency.

Description

Technical Field

[0001] This application relates to the field of quality inspection, and more specifically, to a wake-up test method and system. Background Art

[0002] In the field of intelligent household appliance quality inspection, quality inspection is a crucial step to ensure the performance and user experience of intelligent household appliances. The voice wake-up success rate of household appliances is an important product / technical parameter. An accurate voice wake-up success rate is of great significance and value for household appliance manufacturers to evaluate product performance. Currently, most intelligent household appliance manufacturers test the voice wake-up success rate of intelligent household appliances through manual quality inspection. However, this manual quality inspection method cannot maintain high-accuracy repeated testing for a long time, increasing the probability of false wake-up, misdetection, and misrecording, and cannot meet the current needs of intelligent household appliance quality inspection. Summary of the Invention

[0003] Embodiments of this application provide a wake-up test method and system to improve the success rate of voice wake-up.

[0004] To solve the above technical problems and achieve the above objectives, this application provides the following technical solutions:

[0005] In a first aspect, an embodiment of this application provides a wake-up test method. This method can be applied to an electronic device in a distributed voice wake-up test scenario. Without special instructions, the "electronic device" in the embodiments of this application can refer to the electronic device itself, a component in the electronic device (such as a processor, a chip, or a chip system, etc.), or a logical module or software that can implement all or part of the functions of the electronic device. This method includes: obtaining basic parameter information in the distributed voice wake-up test scenario; based on the basic parameter information and a first condition, using a first algorithm to determine a first detection path corresponding to a first device, the first detection path is related to N test positions, N is an integer greater than or equal to 2, and the first condition is determined based on test requirements; obtaining wake-up test information, the wake-up test information is obtained by performing a wake-up test on at least two second devices on the first detection path, and the wake-up test information is used to determine the wake-up test result.

[0006] The electronic device in this application can be a first device such as a robot or other devices with detection functions. Therefore, by adopting the solution of this application, compared with the manual quality inspection test method, it can achieve the automatic quality inspection of the wake-up success rate of smart home appliances continuously and efficiently in the long term. Quality inspectors do not need to work under high load for a long time, which can reduce labor costs. In addition, the wake-up success rate is automatically recorded, which can minimize the probability of misdetection and misrecording. Furthermore, since this case introduces a solution of generating detection paths using different algorithms based on different test scenarios, it can perform tests in different scenarios according to different test requirements. When the detection path is the shortest, the probability of false wake-up, misdetection, and misrecording can be reduced. On the other hand, the technical solution provided in this application can be further applied to the wake-up test scenario of dynamic mobile home appliances, and can optimize the detection path in a timely manner according to the real-time position change of the dynamic home appliances. Compared with the distributed smart home appliance test method in the static scenario, it has a wider applicability and higher efficiency.

[0007] Combined with the first aspect, in a possible implementation manner of the first aspect, the basic parameter information includes the position information of all devices in the first test environment, and the first test environment is a test environment for static wide measurement of static devices or a test environment for static precise measurement of static devices in the distributed voice wake-up test scenario. All devices in the first test environment use the same test environment map.

[0008] Combined with the first aspect, in another possible implementation manner of the first aspect, the first condition includes one or more of the following: the number of steps, step length, and direction change probability set according to the test requirements; or, the access order of the N test positions; or, the minimization of the distance of the first detection path.

[0009] Combined with the first aspect, in another possible implementation manner of the first aspect, the first algorithm is a random walk algorithm. Based on the basic parameter information and the first condition, using the first algorithm to determine the first detection path corresponding to the first device includes: based on the basic parameter information and the first condition, using the random walk algorithm to determine the first detection path corresponding to the first device, and the first detection path passes through the N test positions.

[0010] In the distributed voice wake-up test, the number of steps, step size, and direction change probability of the random walk algorithm need to be set according to factors such as the size of the test area, device distribution, and voice wake-up range. By dynamically adjusting the parameters, a test path that covers comprehensively and meets the actual test requirements can be generated, improving the comprehensiveness of the test. The algorithm is easy to implement and has a low computational complexity, making it suitable for the test requirements of large-scale distributed systems. Compared with manual detection that relies on manual operations, it takes less time and has high efficiency. Since the test path can cover the voice wake-up test range of all second devices, missed detections can be avoided, and the test results are more reliable, thus achieving the purpose of effectively evaluating the performance of the voice wake-up system.

[0011] Combined with the first aspect, in another possible implementation manner of the first aspect, the first condition includes one or more of the following: the N test positions determined according to the test requirements; or, the access order of the N test positions; or, the distance between adjacent two of the N test positions is less than or equal to a first value; or, the distance between each of the N test positions and at least one noise simulation device is less than or equal to a second value; or, the distance from each of the at least two second devices to the nearest test position is the same; or, the distance of the first detection path is minimized.

[0012] Combined with the first aspect, in another possible implementation manner of the first aspect, the first algorithm is the simulated annealing algorithm. Based on the basic parameter information and the first condition, using the first algorithm to determine the first detection path corresponding to the first device includes: based on the basic parameter information and the first condition, using the simulated annealing algorithm to determine the first detection path corresponding to the first device, and the first detection path passes through the N test positions.

[0013] In the static fine measurement scenario, the simulated annealing algorithm is adopted through a scheme of global search, dynamic adjustment, and gradual optimization, and the first condition can be flexibly defined according to the test requirements to adapt to different test objectives. Compared with the way of manual quality inspection that is prone to falling into local optima and difficult to comprehensively cover complex networks, the simulated annealing algorithm can jump out of local optima, find the global optimum or a near-global optimum test path, and generate the test path quickly. It is suitable for large-scale distributed voice wake-up test systems, with low labor costs and not affected by uncertain factors such as the experience and state of quality inspection personnel, effectively meeting the test requirements, node coverage requirements, environmental simulation requirements, performance optimization, etc. in the distributed voice wake-up detection path, ensuring that the test process is more comprehensive, efficient, and adaptable.

[0014] In combination with the first aspect, in another possible implementation manner of the first aspect, the basic parameter information includes the dynamic position information of all devices in the second test environment, where the second test environment is a test environment for testing dynamic devices in a distributed voice wake-up test scenario. All devices in the second test environment use the same test environment map and update the position information in real time.

[0015] In combination with the first aspect, in another possible implementation manner of the first aspect, the first condition includes one or more of the following: preferentially selecting a point close to the N test positions based on test requirements; or, the access order of the N test positions; or, the priority of accessing path nodes when the path costs are the same.

[0016] In combination with the first aspect, in another possible implementation manner of the first aspect, the first algorithm is the A* path planning algorithm. Based on the basic parameter information and the first condition, using the first algorithm to determine the first detection path corresponding to the first device includes: according to the basic parameter information and the first condition, using the A* path planning algorithm to determine the first detection path corresponding to the first device, where the first detection path is related to the N test positions. According to the basic parameter information and the second condition, using the A* path planning algorithm to determine the detection path corresponding to each of the at least two second devices, and indicating the detection path corresponding to each of the at least two second devices to each of the at least two second devices. The detection path corresponding to each second device passes through the N test positions, and the detection path corresponding to each second device is used for each second device to perform wake-up tests at the N test positions on its corresponding detection path.

[0017] In combination with the first aspect, in another possible implementation manner of the first aspect, the second condition includes one or more of the following: preferentially selecting a point passing through the N test positions; or, the detection paths corresponding to each of the at least two second devices do not overlap at the N test positions; or, the access order of the N test positions; or, the priority of accessing path nodes when the path costs are the same.

[0018] In a dynamic test scenario, using the A* path planning algorithm to generate a detection path can quickly achieve real-time path planning and obstacle avoidance, is suitable for large-scale distributed systems. Compared with manual detection, it takes less time, has higher efficiency, and lower cost. And the A* algorithm can cover a wide area of the network through heuristic search, can avoid missed detections, is not affected by subjective factors (the state and experience of the inspector), and can find high-quality test paths in complex networks through a global search, dynamic adjustment, and step-by-step optimization solution.

[0019] In combination with the first aspect, in another possible implementation manner of the first aspect, the method further includes: based on the second basic parameter information and the first condition, using the A* path planning algorithm to optimize the first detection path corresponding to the first device, and determining the second detection path corresponding to the first device, where the second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment; and / or based on the second basic parameter information and the second condition, using the A* path planning algorithm to optimize the detection paths corresponding to at least one second device, and determining the third detection path corresponding to each second device among the at least one second device, and instructing each second device among the at least one second device to its corresponding third detection path, where the second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment.

[0020] By adopting the above method, the path can be updated in real time to adapt to environmental changes (such as the movement or addition of obstacles), ensuring that the detection path is always optimal. In addition, by optimizing the path in real time, the A* algorithm can quickly adjust the path, reduce unnecessary movements, improve the execution efficiency of the distributed voice wake-up test task, and save time and energy.

[0021] In combination with the first aspect, in another possible implementation manner of the first aspect, the obtaining of the wake-up test information includes: obtaining the wake-up test information from at least one visual signal processing device in the first test environment and obtaining the wake-up test information from at least one acoustic processing device in the first test environment.

[0022] In combination with the first aspect, in another possible implementation manner of the first aspect, the obtaining of the wake-up test information includes: obtaining the wake-up test information determined by the first device in the first detection path in the second test environment and the wake-up test information determined by each second device among the at least two second devices in its corresponding detection path.

[0023] In combination with the first aspect, in another possible implementation manner of the first aspect, the method further includes: obtaining the first wake-up test information, where the first wake-up test information includes the wake-up test information obtained by the first device in the second detection path in the second test environment and the wake-up test information obtained by each second device among the at least one second device in its corresponding third detection path.

[0024] In a second aspect, the present application provides a distributed voice wake-up test system, which is applied to static fine measurement test scenarios and static wide measurement test scenarios, and includes a first device, at least one noise simulation device, at least one visual signal processing device, at least one acoustic signal processing device, and at least two second devices that use the same test environment map, where,

[0025] The first device is configured to obtain the basic parameter information in the distributed voice wake-up test scenario. The basic parameter information includes the location information of all devices in the first test environment. Based on the basic parameter information and the first condition, a first detection path corresponding to the first device is determined using a first algorithm. The first detection path is related to N test positions, where N is an integer greater than or equal to 2, and the first condition is determined based on the test requirements.

[0026] The at least one noise simulation device is configured to simulate different noises.

[0027] The at least one visual signal processing device and the at least one acoustic signal processing device are configured to obtain the wake-up test information obtained by performing wake-up tests on the at least two second devices at the N test positions, and report the wake-up test information to the first device.

[0028] The first device is configured to obtain the wake-up test information, and the wake-up test information is used to determine the wake-up test result.

[0029] Combined with the second aspect, in a possible implementation manner of the first aspect, the first algorithm is a random walk algorithm, and the first condition includes one or more of the following: the number of steps, step size, and direction change probability set according to the test requirements; or, the access order of the N test positions; or, the minimization of the distance of the first detection path.

[0030] Combined with the second aspect, in another possible implementation manner of the first aspect, the first algorithm is a simulated annealing algorithm, and the first condition includes one or more of the following: the N test positions determined according to the test requirements; or, the distance between adjacent two of the N test positions is less than or equal to a first value; or, the distance between each of the N test positions and at least one noise simulation device is less than or equal to a second value; or, the distance between each of the at least two second devices and the nearest test position is the same.

[0031] In a third aspect, an embodiment of the present application provides a distributed voice wake-up test system applied to a dynamic test scenario, including a first device, at least one noise simulation device, and at least two second devices that use the same test environment map, and the test environment map is updated in real time for location. The system includes:

[0032] The first device is used to obtain the basic parameter information in the distributed voice wake-up test scenario. The basic parameter information includes the dynamic position information of all devices in the dynamic test scenario. Based on the basic parameter information and the first condition, the A* path planning algorithm is used to determine the first detection path corresponding to the first device. The first detection path is related to N test positions, where N is an integer greater than or equal to 2. The wake-up test is performed on at least two second devices on the first detection path to obtain the wake-up test information. Among them, the first condition is determined based on the test requirements, and the first condition includes one or more of the following: preferentially selecting points close to the N test positions based on the test requirements, and the access order of the N test positions.

[0033] The first device is used to determine the detection path corresponding to each of at least two second devices by using the A* path planning algorithm according to the basic parameter information and the second condition, and indicate the detection path corresponding to each second device to each of the at least two second devices. The detection path corresponding to each second device passes through the N test positions, and controls each second device to perform the wake-up test at the N test positions on its corresponding detection path.

[0034] Each of the second devices is used to perform the wake-up test at the N test positions on its corresponding detection path and send the wake-up test information to the first device.

[0035] Among them, the second condition includes one or more of the following: the detection paths corresponding to each of at least two second devices do not overlap at the N test positions, the access order of the N test positions, or the priority of path node access in the case of the same path cost.

[0036] The at least one noise simulation device is used to simulate different noises.

[0037] Combined with the third aspect, in another possible implementation manner of the first aspect, the first device is further used to optimize the first detection path corresponding to the first device by using the first algorithm based on the second basic parameter information and the first condition, determine the second detection path, and the first device performs the wake-up test on the second detection path. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment; and / or

[0038] Based on the second basic parameter information and the second condition, optimize the detection path corresponding to the at least one second device by using the first algorithm, determine the third detection path corresponding to each second device in the at least one second device, indicate the corresponding third detection path to each second device in the at least one second device, and control each second device in the at least one second device to perform a wake-up test on its corresponding third detection path;

[0039] Each second device in the at least one second device is configured to perform a wake-up test on the corresponding third detection path and report wake-up test information to the first device.

[0040] In a fourth aspect, an embodiment of the present application provides an electronic device, including: one or more processors; one or more memories; the one or more memories store one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device is caused to execute the method in the first aspect or any optional implementation manner in the first aspect.

[0041] In a fifth aspect, an embodiment of the present application provides a communication device, including a module for executing the method in the first aspect or any optional implementation manner in the first aspect.

[0042] In a sixth aspect, an embodiment of the present application provides a computer program product, including instructions, when the instructions run on a computer, the computer is caused to execute the method in the first aspect or any optional implementation manner in the first aspect.

[0043] In a seventh aspect, an embodiment of the present application provides a computer storage medium for storing a computer program, when the computer program runs on a computer, the computer is caused to execute the method in the first aspect or any optional implementation manner in the first aspect.

[0044] In an eighth aspect, an embodiment of the present application provides a chip system, the chip system includes a processor for supporting a device to implement the functions involved in the above aspects, for example, sending or processing data and / or information involved in the above method. In a possible design, the chip system further includes a memory for storing necessary program instructions and data of the device. The chip system may be composed of chips or may include chips and other discrete devices.

[0045] In a ninth aspect, an embodiment of the present application provides a chip, including one or more interface circuits and one or more processors; the interface circuit is configured to receive a signal from a memory of an electronic device and send the signal to the processor, the signal includes computer instructions stored in the memory; when the processor executes the computer instructions, the electronic device is caused to execute the communication method in the first aspect or any optional implementation manner in the first aspect. Description of the Drawings

[0046] Figure 1 Schematic diagram of a test environment architecture for performing distributed voice wake-up testing provided by an embodiment of the present application;

[0047] Figure 2 Based on an embodiment of the present application Figure 1 Schematic diagram of the process of a distributed voice wake-up testing method provided by the test architecture shown;

[0048] Figure 3 Schematic diagram of a path for determining a first detection path by a random walk algorithm provided by an embodiment of the present application;

[0049] Figure 4 Schematic diagram of a path for determining a first detection path by a simulated annealing algorithm provided by an embodiment of the present application;

[0050] Figure 5 Another schematic diagram of a test environment architecture for performing distributed voice wake-up testing provided by an embodiment of the present application;

[0051] Figure 6 Based on an embodiment of the present application Figure 5 Schematic diagram of the process of a distributed voice wake-up testing method provided by the test environment shown;

[0052] Figure 7 Another schematic diagram of a test environment architecture for performing distributed voice wake-up testing provided by an embodiment of the present application;

[0053] Figure 8 Based on an embodiment of the present application Figure 7 Schematic diagram of the process of a distributed voice wake-up testing method provided by the test environment shown;

[0054] Figures 8A to 8F Schematic diagram of the process of determining a first detection path by using the A* path planning algorithm provided by an embodiment of the present application;

[0055] Figures 9 to 9J Another schematic diagram of the process of determining a first detection path by using the A* path planning algorithm provided by an embodiment of the present application;

[0056] Figure 10 Based on an embodiment of the present application Figure 7 Schematic diagram of the detection path determined by the test environment shown;

[0057] Figure 11 Another schematic diagram of a test environment architecture for performing distributed voice wake-up testing provided by an embodiment of the present application;

[0058] Figure 12 This is a schematic flowchart of a distributed voice wake-up test method provided by an embodiment of the present application based on Figure 11 the test environment shown;

[0059] Figure 13 This is a schematic diagram of a distributed voice wake-up test device provided by an embodiment of the present application;

[0060] Figure 14 This is a schematic diagram of another distributed voice wake-up test device provided by an embodiment of the present application. Detailed implementation manners

[0061] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. The terms used in the following embodiments are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and claims of the present application, the singular forms "a", "an", "the", "above", "said", "this" are also intended to include the forms such as "one or more", unless the context clearly indicates otherwise. It should also be understood that in the embodiments of the present application, "one or more" means one, two or more than two; " / ", which describes the relationship between associated objects, indicates that three relationships may exist; for example, A and / or B may indicate: A exists alone, A and B exist simultaneously, and B exists alone, where A and B may be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0062] Reference to "an embodiment" or "some embodiments" in this specification means that specific features, structures, or characteristics described in connection with the embodiment are included in one or more embodiments of the present application. Thus, the phrases "in an embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments" and the like that appear in different places in this specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in another way. The terms "including", "comprising", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in another way.

[0063] The multiple involved in the embodiments of the present application means greater than or equal to two. It should be noted that in the description of the embodiments of the present application, the terms "first", "second", etc. are only used for the purpose of distinguishing descriptions and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying order.

[0064] The embodiments of the present application involve step numbers, which do not represent the inevitable order of the steps. They are only used for the purpose of distinguishing descriptions and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying an order.

[0065] In the embodiments provided by the present application, the electronic device can be in various forms. For example, a mobile phone, a tablet computer (Pad), a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a vehicle-mounted terminal device, a wireless terminal in self-driving, a wireless terminal in remote medical, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, a wearable terminal device, and so on. The control terminal can sometimes also be referred to as a terminal device, a user equipment (UE), an access terminal device, a vehicle-mounted terminal, an industrial control terminal, a UE unit, a UE station, a mobile station, a mobile device, a remote station, a remote terminal device, a mobile device, a UE terminal device, a terminal device, a wireless communication device, a UE agent, or a UE device, etc. The terminal can also be a fixed terminal or a mobile terminal.

[0066] Before introducing the solution of the present application, first, relevant concepts involved in this article are explained:

[0067] The static wide test scenario is used to comprehensively evaluate the performance of the device under test for voice wake-up in different conditions, find the common problems of the test device, so that the device performance can be optimized according to the test results of the static wide test, thereby achieving the purpose of improving the performance of the test device.

[0068] The static precise test scenario is used to conduct in-depth tests on specific conditions to discover the performance bottleneck of the device under test for voice wake-up in a specific scenario, and conduct performance evaluation according to the test results of the voice wake-up test to achieve the purpose of solving specific problems.

[0069] The dynamic test scenario refers to simulating user-state behaviors in the actual usage scenario to test the performance of the device under test in the voice wake-up scenario. Compared with the static wide test scenario and the static precise test scenario, the dynamic test scenario pays more attention to the performance under dynamic factors such as the movement of the device under test and device interaction.

[0070] The random walk algorithm, also known as the random stroll algorithm or random walking algorithm, its core concept refers to a movement trajectory composed of a series of random steps, that is, it describes the process of an object moving randomly in space.

[0071] The simulated annealing algorithm is a probability-based optimization algorithm, often used to search for approximate global optimal solutions in a large solution space. The simulated annealing algorithm can plan the optimal path from the starting point to the ending point based on the constraint conditions.

[0072] The A* path planning algorithm is a global path planning algorithm. It finds the shortest path by comprehensively considering the heuristic evaluation function and the actual cost function and searching for nodes in the search state space.

[0073] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.

[0074] Currently, in the field of quality inspection of intelligent devices (such as home appliance devices), the voice wake-up rate is a very important product / technical parameter. An accurate wake-up rate is of great significance and value for manufacturers to evaluate the product performance of intelligent devices and horizontally evaluate products of similar competitors. At the same time, if there are multiple products of the same type in a scenario, a wake-up word may wake up multiple products of the same type simultaneously, thus reducing the user experience. To better serve users, the concept of "distributed wake-up" is introduced. Distributed wake-up means that if there are multiple devices of the same type in the same scenario, when the user issues a wake-up word, only the device closest to the user will be woken up. However, there are many drawbacks to distributed wake-up. First, in this method, through manual quality inspection, it is very difficult for quality inspectors to maintain a high accuracy rate in repeated tests for a long time. Second, this method is time-consuming and laborious. Many quality inspection personnel need to be hired to conduct automated quality inspections on various products on each production line, and it is very difficult for the manual quality inspection plan to maintain non-stop long-term operation for 24 hours. Therefore, a large number of personnel need to work in relays to meet the continuous quality inspection requirements of production line products.

[0075] The present application provides a wake-up test method, which can be applicable to various scenarios of distributed voice wake-up detection. An electronic device determines a first detection path corresponding to a first device based on basic parameter information and a first condition in a distributed voice wake-up test scenario, and the first condition is determined based on test requirements. The electronic device obtains wake-up test information, which is obtained by performing a wake-up test on at least two second devices on the first detection path, and the wake-up test information is used to determine a wake-up test result. Using the electronic device to perform a distributed voice wake-up test can meet the continuous quality inspection requirements of the product line for products, and can improve the voice wake-up success rate of the device and reduce the probability of false detection and misrecording.

[0076] To ensure the accuracy of the voice wake-up test, a professional test environment needs to be built. This environment should have low noise, stable acoustic characteristics, and be able to simulate real usage scenarios. The test environment needs to be under wireless network coverage. In addition, the test environment also needs to have adjustable volume and sound quality to evaluate the voice wake-up effect under different conditions.

[0077] Figure 1 It is a schematic diagram of the first test environment architecture for distributed voice wake-up testing. All devices in the first test environment are equipped with positioning modules and use the map of the same test environment, and the location information (such as coordinate information) of all devices in the test environment is marked in the environmental map. For ease of understanding, the example of this application takes two second devices (Device No. 1 and Device No. 2) as an example for illustration. As Figure 1 shown, according to the acoustic-related standards, a noise simulation device 1 is placed at a vertical position 2 meters above the midpoint of the perpendicular bisector of Device No. 1 and Device No. 2, and a noise simulation device 2 is placed at a vertical position 2 meters below the midpoint of the perpendicular bisector of Device No. 1 and Device No. 2. A visual signal processing device 1, a visual signal processing device 2, an acoustic signal processing device 1, and an acoustic signal processing device 2 are respectively arranged near Device No. 1 and Device No. 2. Among them, the types of Device No. 1 and Device No. 2 are the same. Figure 1 The layout of the devices in the test environment shown, the number and types of devices included in the test scenario are only examples cited to understand the technical solution of this application, and this application includes but is not limited to the above examples. The devices shown below will be introduced separately: Figure 1 shown will be introduced:

[0078] The visual signal processing device 1 and the visual signal processing device 2 are used to capture, analyze, and process visual signals. For example, the visual signal processing device can be a camera.

[0079] The acoustic signal processing device 1 and the acoustic signal processing device 2 are used to analyze and process acoustic signals. For example, the acoustic signal processing device can be a high-precision microphone or a recording device.

[0080] Noise simulation devices 1 and 2 are used to simulate different noises (such as white noise, human voices, music).

[0081] The first device determines the detection path and performs the wake-up test. Therefore, the first device needs to have the following capabilities: 1. A lidar for mapping, navigation, and obstacle avoidance of the first device; 2. A motion module for driving the movement of the first device; 3. An industrial computer for the core control and communication of the first device; 4. A lifting rod. In order to simulate sound sources emitted by users at various heights, a push-pull rod is required to drive an artificial mouth to reach different heights to simulate the user's voice; 5. A power supply module for supplying power to various devices on the first device; 6. An artificial mouth for simulating the user's voice. For example, the first device can be a robot or other devices that can perform voice wake-up tests.

[0082] If the test operation in the first test environment is controlled by the first device, the first device can also be used to receive the visual test information and acoustic test information reported by the visual signal processing device and the acoustic signal processing device, and send the test information to the control platform of the control terminal. In addition, the first device can obtain the position information of each device in the first test environment and mark it on the detection environment map to establish an electronic map of the entire closed-loop experimental environment.

[0083] In order to more finely distinguish the scenarios of distributed voice wake-up tests, the distributed voice wake-up test soundscapes can be divided into dynamic tests, static wide tests, and static fine tests. Among them, the dynamic test refers to the distributed voice wake-up test in which the first device and at least two second devices move within a certain range. The static wide test refers to the distributed voice wake-up test in which the first device performs distributed voice wake-up tests on the immovable second device from multiple angles and multiple positions. The static fine test refers to the distributed voice wake-up test on the immovable second device through several fixed angles and distances. The following will separately describe these three test scenarios.

[0084] The embodiment of the present application provides a wake-up test method, which can be applied to Figure 1 the test environment shown in Figure 2 As shown, the wake-up test method includes the following steps:

[0085] 200. The first device obtains the basic parameter information in the distributed voice wake-up test scenario.

[0086] Exemplarily, the basic parameter information may include the first test environment (such as Figure 1The location information of all devices within the shown test environment), where the first test environment is a test environment for conducting static wide - range testing and static precise testing. The devices within the first test environment may include a first device, at least two second devices, at least one visual signal processing device, at least one acoustic signal processing device, and at least one noise simulation device.

[0087] For example, Figure 1 After the shown test environment is successfully set up, the first device obtains the location information (such as coordinate information) of each device in the test environment based on the positioning modules on each device in the test environment, such as the location information of the first device, the location information of at least two second devices, the location information of at least one visual signal processing device, the location information of at least one acoustic signal processing device, and the location information of at least one noise simulation device. The first device marks the obtained location information of each device in the test environment on the detection environment map to establish an electronic map of the entire closed - loop experimental environment.

[0088] 201. Based on the basic parameter information and the first condition, use the first algorithm to determine the first detection path corresponding to the first device. The first detection path is related to N test positions, where N is an integer greater than or equal to 2.

[0089] Exemplarily, the first condition is determined based on test requirements. The test requirements may include path coverage requirements, node coverage requirements, environment simulation requirements, and performance evaluation requirements, etc. For example, the path coverage requirement may be to determine the test path to cover different nodes (such as test positions) in the distributed voice wake - up test system by randomly selecting paths; the node coverage can be to ensure that each node has the opportunity to be accessed to verify the success rate of the distributed voice wake - up test; the environment simulation requirement can be to introduce noise interference in the detection path to test the robustness of the system in a complex environment; the performance evaluation requirement can be to test the response time under different paths to evaluate the system performance. The above - mentioned test requirements are only examples for understanding the technical solution of the present application, and the present application includes but is not limited to this.

[0090] Exemplarily, the fact that the first detection path is related to N test positions may include that the first detection path needs to pass through N test positions. Among them, the role of the test position is to simulate the position of the played sound source. The first detection path is a path for the first device to move in the test environment to cover the voice wake - up test range of all second devices and ensure that the wake - up function can be correctly triggered at different positions.

[0091] Exemplarily, if it is necessary to conduct a distributed voice wake - up test on at least two immovable second devices from multiple angles and multiple positions, a static wide - range test scenario can be adopted.

[0092] For example, in the static wide-area testing scenario, the distributed voice wake-up method is adopted to perform voice wake-up tests on static devices such as Device No. 1 and Device No. 2. The first device can use the random walk algorithm to determine the first detection path corresponding to the first device according to the first condition. The first detection path passes through N test positions. Among them, the random walk algorithm can be used to simulate the movement path of the first device between different positions.

[0093] Exemplarily, the first condition may include the number of steps, step size, and direction change probability set according to the test requirements, the access order of the N test positions, and the minimization of the distance of the first detection path. The number of steps, step size, and direction change probability are the key parameters of the random walk algorithm, which directly affect the coverage range of the detection path and the test effect of the wake-up test. Among them:

[0094] The number of steps determines the length of the first detection path and the duration of the distributed voice wake-up test. When the test area is large, the number of steps should be set larger to ensure coverage of the entire area. When the test area is small, the number of steps should be appropriately reduced. When multiple second devices are densely distributed, the number of steps can be appropriately reduced. When multiple second devices are sparsely distributed, the frequency should be increased. For example, for a medium-sized room (such as 10m * 10m), the number of steps can be set to 50 - 100 steps. For a larger room or a complex scenario, the number of steps can be set to 100 - 200 steps.

[0095] The step size determines the distance that the first device moves each time and needs to be set according to the test scenario and the distribution of multiple second devices. For example, if the area of the test scenario is large, the step size can be set to a larger distance to quickly cover the entire area; if the area of the test scenario is small, the step size can be set to a smaller distance to improve the test accuracy. When multiple second devices are relatively densely distributed, the step size can be set to a smaller distance to ensure that the first detection path can pass near each second device; similarly, when multiple second devices are relatively sparsely distributed, the step size can be set to a larger distance. The step size can also be dynamically adjusted. For example, it can be adjusted according to the distance between the current position and multiple second devices. The step size decreases when approaching multiple second devices and increases when moving away from multiple second devices.

[0096] The direction change probability determines the randomness and coverage of the random walk path. The setting of the direction change probability specifically depends on the distribution of the test target and multiple second devices. For example, if the test target is to comprehensively cover the test area, the direction change probability can be set relatively high to increase the randomness of the path. If the test target is to test a specific path (such as a straight-line movement), the direction change probability can be set relatively low. When the distribution of multiple second devices is uneven, the direction change probability can be dynamically adjusted to guide the path towards the area where multiple second devices are concentrated. The direction change probability can adopt a fixed value, such as 0.2, that is, there is a 20% probability of changing the direction each time of movement. The direction change probability can also adopt a dynamic probability, that is, the direction change probability is dynamically adjusted according to the distance between the current position and the second device, increasing when approaching the second device and decreasing when moving away from the second device.

[0097] For example, as Figure 3 shown, taking the current position of the first device as the starting point, according to the current position and the distances from multiple second devices (Device No. 1 and Device No. 2), the step size and the direction change probability are dynamically adjusted, and the test position 1 is determined and recorded based on the adjusted step size and direction change probability. Taking the test position 1 as the starting point, according to the distance between the test position 1 and the second device, the step size and the direction change probability are dynamically adjusted, and the test position 2 is determined and recorded based on the adjusted step size and direction change probability. Taking the test position 2 as the starting point, according to the distance between the test position 2 and the second device, the step size and the direction change probability are dynamically adjusted, and the test position 3 is determined and recorded based on the adjusted step size and direction change probability. And so on, until after determining and recording the test position 15, the voice wake-up test range of all current second devices is covered, and the generation of the first detection path is completed. The first detection path covers 15 test positions.

[0098] In the distributed voice wake-up test, the number of steps, the step size, and the direction change probability of the random walk algorithm need to be set according to factors such as the size of the test area, the device distribution, and the voice wake-up range. By dynamically adjusting the parameters, a test path that covers comprehensively and meets the actual test requirements can be generated, improving the comprehensiveness of the test. The algorithm is easy to implement, has a low computational complexity, and is suitable for the test requirements of large-scale distributed systems. Compared with manual detection which relies on manual operations, it takes less time and has high efficiency. Since the first detection path determined by using the random walk algorithm can cover the voice wake-up test range of all second devices, it can avoid missed detections, and the test results are more reliable, thus achieving the purpose of effectively evaluating the performance of the voice wake-up system.

[0099] Exemplarily, when it is necessary to test the success rate of distributed voice wake-up for multiple immovable second devices through several fixed angles and distances, a static precise measurement scenario can be adopted.

[0100] For example, in the static precision measurement scenario, the distributed voice wake-up method is adopted to perform voice wake-up tests on multiple second devices such as Device No. 1 and Device No. 2. The first algorithm can be the simulated annealing algorithm, which can be used to simulate the movement path of the first device between different positions.

[0101] For example, the distributed voice wake-up method for home appliances is adopted to perform voice wake-up tests on multiple second devices such as Device No. 1 and Device No. 2. The voice wake-up tests need to be carried out at different test positions to verify the voice wake-up success rates of Device No. 1 and Device No. 2. Therefore, based on the basic parameter information and the first condition, the simulated annealing algorithm can be used to determine the first detection path corresponding to the first device, and the first detection path needs to pass through N test positions.

[0102] Exemplarily, the first device can determine the first condition according to the actual test requirements. For example, the first condition can include one or more of the following: N test positions determined according to the test requirements; or, the access order of the N test positions; or, the distance between adjacent two of the N test positions is less than or equal to the first value; or, the distance between each of the N test positions and at least one noise simulation device is less than or equal to the second value; or, the distance of the first detection path is minimized; or, the distance between each of the at least two second devices and the nearest test position is the same.

[0103] For example, assume N = 2, that is, 2 test positions are required. The position where the first device 20 is located is the starting position (x1, y1). A target position (x2, y2) such as Test Position 1 is randomly generated based on the starting position. The target position (x2, y2) satisfies the first condition. Then (x2, y2) = (x1, y1) + Δ, where Δ is a randomly generated movement vector. The energy difference is calculated by the following formula:

[0104] ΔE = E(x2, y2) - E(x1, y1) (1)

[0105] Among them, ΔE is the energy difference, indicating the energy change from the starting position (x1, y1) to the target position (x2, y2). For example, the penalty constraint conditions required for the test can be introduced into formula (2):

[0106] E = base_energy + k·d(x) (2)

[0107] Among them, base_energy is the basic energy value, which can represent the default energy when constraints are not considered and is usually used to quantify the basic cost of a path, such as path length; d(x) is the distance between the current solution and the constraint boundary, which is used to measure whether the current solution violates the constraint conditions (such as the distance between the test position (x2, y2) and the noise simulation device exceeds the second value), then d(x) is negative, and k is the penalty weight coefficient, which is used to adjust the penalty intensity for constraint violation. When d(x) is negative, k·d(x) will increase E, thereby punishing the solution that violates the constraint. The larger the value of K, the stronger the penalty for constraint violation.

[0108] Among them, if ΔE < 0, it means that the new path (the distance from (x1, y1) to (x2, y2)) is better, then the new solution is accepted; if ΔE ≥ 0, it means that the new path is worse, then the new solution is accepted with probability P, and the probability calculation is shown in formula (3):

[0109]

[0110] Among them, ΔE is the energy difference, that is, the calculation result of formula 1; T is the current temperature, which is a parameter controlling randomness. The initial value is relatively high and gradually decreases with iteration. When ΔE ≥ 0 (the new solution is worse), the inferior solution is accepted with probability P. The higher the temperature T, the greater the probability of accepting the inferior solution.

[0111] Each iteration starts from the last position of the current path and randomly generates a new target position. For example, starting from (x2, y2), a new target position (x3, y3), that is, the test position 3, is randomly generated. Based on the constraint conditions, the energy difference is calculated using formula (1), and whether to accept the new target position is determined through the energy function (formula 2) and probability judgment (formula 3). If accepted, the new target position is added to the path; if rejected, it is regenerated.

[0112] After each test position is generated, the temperature T is decreased once according to formula (4). Therefore, the current temperature T is updated using formula (4) for each calculation:

[0113] T = T × cooling_rate (4)

[0114] Among them, cooling_rate is used to represent the cooling rate, which is used to control the speed of temperature decrease, and its value range is between (0, 1).

[0115] Through the above iterative process, until two test positions that meet the constraint conditions, such as (x2, y2) and (x3, y3), are obtained, and finally a first test path including two test positions is cumulatively formed. As Figure 4 shown is the path schematic diagram of the first detection path determined by the simulated annealing algorithm.

[0116] The process of obtaining the first detection path using the simulated annealing algorithm is introduced below through specific examples.

[0117] Assume the starting position (x1, y1) = (0, 0), generate a new position (x2, y2) = (1, 0). If the noise simulation device is at position (3, 0), assume the first condition is that the distance between the test position and the noise simulation device must be greater than or equal to 2. base_energy is the Manhattan distance (the path length from the starting point to the new position), the penalty coefficient k = 2, the initial temperature T0 = 100, and the cooling rate cooling_rate = 0.95. The distance from the starting position (0, 0) to the noise simulation device is 3 (meeting the first condition). d(x) = 3 - 2 = 1 (no penalty). E = base_energy + k·d(x), that is, E(x1, y1) = 0×(base_energy) + 2·1 = 2. The distance from the new position (1, 0) to the noise simulation device is 2 (meeting the first condition), d(x) = 2 - 2 = 0 (no penalty). E(x2, y2) = 1×(base_energy) + 2*0 = 1. The energy difference ΔE = E(x2, y2) - E(x1, y1) = 1 - 2 = -1, that is, ΔE < 0. Since ΔE < 0, directly accept the new position. Update the temperature T = T×cooling_rate = 100*0.95 = 95. After each iteration, the temperature T is updated according to formula (4), and the temperature gradually decreases, and the probability of accepting a worse solution decreases.

[0118] Further, for the second iteration, T0 = 95, the starting position (x2, y 2 ) = (1, 0), E(x2, y 2 ) = 1, the new position (x3, y 3 ) = (2, 0). The distance from the new position (2, 0) to the noise simulation device is 1 (violating the first condition), d(x) = 1 - 2 = -1 (negative number, imposing a penalty). E(x3, y3) = 2(base_energy) + 2*(-1) = 0, ΔE = E(x3, y3) - E(x2, y2) = 0 - 1 = -1. Since ΔE < 0, also accept the new position. Although the first condition is violated, the total energy is lower. Update the temperature T = T×cooling_rate = 95*0.95 = 90.25. The first detection path can be (0, 0)->(1, 0)->(2, 0).

[0119] Exemplarily, assume the current temperature T0 = 95, randomly generate a new position (x4, y 4 ) = (2, 0) from (1, 0). Assume when ΔE = 1, the new solution is worse and the acceptance probability needs to be further calculated. Accept the new position with probability P≈0.9895, update the new path to (0,0)->(1,0)->(2,0), and update the temperature T = T×cooling_rate = 95*0.95 = 90.25.

[0120] In the static precise measurement scenario, the simulated annealing algorithm is adopted through a scheme of global search, dynamic adjustment, and gradual optimization. Moreover, the first condition can be flexibly defined according to the test requirements to adapt to different test objectives. Compared with manual quality inspection, which is prone to falling into local optima and difficult to comprehensively cover complex networks, the simulated annealing algorithm can jump out of local optima, find the global optimum or a near-global optimum test path, generate the test path quickly, is suitable for large-scale distributed voice wake-up test systems, has low labor costs, is not affected by uncertain factors such as the experience and state of quality inspection personnel, and can effectively meet the test requirements, node coverage requirements, environment simulation requirements, performance optimization, etc. in the distributed voice wake-up detection path, ensuring that the test process is more comprehensive, efficient, and adaptable.

[0121] 202. The first device performs wake-up tests at N test positions on the first detection path.

[0122] Exemplarily, the first device moves on the first detection path and performs a voice wake-up test every time it moves to a test position. For example, as Figure 3 shown, in the static wide measurement scenario, the first device moves to test position 1. First, calculate the distances d1 and d2 between test position 1 and device No. 1 and device No. 2. At the same time, calculate the angles between the orientation of the first device and device No. 1 and device No. 2. If it is determined that the distance d1 between device No. 1 and the first device is the shortest, use noise simulation device 1 and noise simulation device 2 to play the corpus, obtain test data through visual signal processing device 1 and visual signal processing device 2, acoustic signal processing device 1 and acoustic signal processing device 2, and logs, etc., and send the test data to the control platform. The control platform comprehensively determines the wake-up situation of device No. 1 and device No. 2 based on the obtained test data. If only device No. 1 is woken up, record the voice wake-up success rate result of this time as "success". Otherwise, record all other wake-up results as "failure", and the control platform completes the recording of the test results at test position 1.

[0123] Assume that the distance d1 between the test position 1 and the first device is the same as the distance d2 between the test position 1 and the second device, and the distance is the shortest. Further, the discrimination principle of "same distance, based on orientation" needs to be adopted, that is, further determine the magnitudes of the angle A between the orientation of the first device and the first device, and the angle B between the orientation of the first device and the second device. For example, if it is determined that the angle A between the first device and the first device is the smallest, and only the first device is awakened during the wake-up test, then the result of the voice wake-up success rate of this time is recorded as "success", and all other wake-up results are recorded as "failures". The test results are uploaded to the control terminal to complete the recording of the test results at the test position 1.

[0124] According to the above operation logic, complete the test operations at all test positions (it is also possible to perform multiple rounds of iterative testing according to test requirements).

[0125] 203. The first device obtains wake-up test information.

[0126] Exemplarily, the wake-up test information is obtained by the first device performing wake-up tests on at least two second devices at the N test positions, and the wake-up test information is used to determine the wake-up test results.

[0127] For example, the wake-up test information can be obtained in the following way: The first device receives the wake-up test information from at least one visual signal processing device and at least one acoustic processing device in the first test environment.

[0128] Exemplarily, the first device receives the visual information data generated by the visual signal processing device 1 and the visual signal processing device 2, and receives the acoustic information data generated by the acoustic signal processing device 1 and the acoustic signal processing device 2. After completing the test operations at the N target positions, the first device reports the obtained test information to the control terminal, and the control terminal generates a test report, which details the distributed wake-up test experiment situation of the first device and the second device and the results of the distributed voice wake-up test.

[0129] Adopting the above solution, the first detection path is determined based on the random walk algorithm and the simulated annealing algorithm, so that the first device performs wake-up tests at the N target positions on the first detection path. It can reduce labor costs, automatically record the wake-up success rate, and minimize the probability of misdetection and misrecording. In addition, since the solution of generating the detection path using different algorithms based on different test scenarios is introduced, it is possible to perform tests in different scenarios according to different test requirements, and it can adapt to distributed voice wake-up test scenarios of different scales, thereby improving the detection efficiency.

[0130] Another wake-up test method is provided in the embodiments of the present application, and this method can be applied to Figure 5 the test environment shown. Figure 5 The test environment shown is the same asFigure 1 The test environment shown is similar and can also be applied to the test scenarios of static wide - scale testing and static precise testing. The difference is that, Figure 5 the test environment shown also includes a control terminal, which is used to help the operator complete the management, control, and information recording of various experimental devices in the test scenario. In this scenario, the control terminal controls the first device to perform the wake - up test. It should be particularly noted that, Figure 5 the layout of the devices in the test environment shown and the quantity and types of devices included in the test scenario are only examples cited for understanding the technical solution of this application. This application includes but is not limited to the above examples.

[0131] Figure 6 This is a schematic flowchart of a distributed voice wake - up test method provided by an embodiment of this application based on Figure 5 the test environment shown. As shown in Figure 6 the figure, this wake - up test method includes the following steps:

[0132] 601. The control terminal obtains the basic parameter information in the distributed voice wake - up test scenario.

[0133] Exemplarily, the basic parameter information may include the location information of all devices in the first test environment (such as the test environment shown in Figure 5 the figure). The first test environment is a test environment for static wide - scale testing and static precise testing. The devices in the first test environment may include a control terminal, a first device, at least two second devices, at least one visual signal processing device, at least one acoustic signal processing device, and at least one noise simulation device. The specific way to obtain the basic parameter information may refer to the way the first device obtains the basic parameter information in the example shown in Figure 2 the figure, which will not be elaborated here.

[0134] 602. The control terminal, based on the basic parameter information and the first condition, uses the first algorithm to determine the first detection path corresponding to the first device, and instructs the first device of the first detection path. The first detection path is related to N test positions, where N is an integer greater than or equal to 2, and the first condition is determined based on the test requirements.

[0135] Exemplarily, if a distributed voice wake - up test needs to be performed on multiple immovable second devices from multiple angles and multiple positions, a static wide - scale test scenario can be adopted. The first algorithm may be a random walk algorithm. The control terminal, based on the basic parameter information and the first condition, uses the random walk algorithm to determine the first detection path corresponding to the first device. The specific determination method may refer to Figure 2In the illustrated embodiment, a method for obtaining a first detection path corresponding to a first device by using a random walk algorithm. Among them, the first condition may include the number of steps, step size, and direction change probability set according to test requirements, the access order of the N test positions, and the minimization of the distance of the first detection path. The introduction of the number of steps, step size, and direction change probability can be referred to Figure 2 the introduction in the illustrated example.

[0136] Exemplarily, when it is necessary to test the success rate of distributed voice wake-up of multiple immovable second devices through several fixed angles and distances, a static fine measurement scenario can be adopted. The first algorithm can be a simulated annealing algorithm, which can be used to simulate the movement path of the first device between different positions. The first condition may include one or more of the following: N test positions determined according to test requirements; or, the access order of the N test positions; or, the distance between adjacent two of the N test positions is less than or equal to a first value; or, the distance between each of the N test positions and at least one noise simulation device is less than or equal to a second value; or, the minimization of the distance of the first detection path; or, the distance between each of at least two second devices and the nearest test position is the same. The control terminal determines the first detection path corresponding to the first device by using the simulated annealing algorithm based on the basic parameter information and the first condition. The specific determination method can be referred to Figure 2 the method for determining the first detection path corresponding to the first device by using the simulated annealing algorithm in the illustrated embodiment.

[0137] 603. The control terminal controls the first device to perform a wake-up test on at least two second devices at the N test positions of the first detection path.

[0138] The specific manner of performing the wake-up test can be referred to Figure 2 the method shown.

[0139] 604. The control terminal receives the wake-up test information, determines the wake-up test result based on the wake-up test information, and generates a test report.

[0140] Exemplarily, the control terminal receives the visual information data generated by the visual signal processing device 1 and the visual signal processing device 2, and receives the acoustic information data generated by the acoustic signal processing device 1 and the acoustic signal processing device 2. After completing the test operations at the N test positions, a test report is generated, which details the distributed wake-up test experiment of Device No. 1 and Device No. 2 and the results of the distributed voice wake-up test.

[0141] Adopting the above solution, the first detection path is determined based on the random walk algorithm and the simulated annealing algorithm, enabling the first device to perform wake-up tests at N test positions on the first detection path, determining the wake-up test results based on the wake-up test information, and generating a test report. This can reduce labor costs and automatically record the wake-up success rate, minimizing the probability of misdetection and misrecording. In addition, since a solution of generating detection paths using different algorithms based on different test scenarios is introduced, it is possible to conduct tests in different scenarios according to different test requirements, reducing the false wake-up rate, the probability of misdetection and misrecording while minimizing the detection path. Compared with the method of the first device managing distributed voice wake-up tests, the control terminal has a longer battery life for managing distributed voice wake-up tests and can execute test tasks more stably in the long term.

[0142] Figure 7 is another second detection environment built for performing distributed voice wake-up. Figure 7 The detection environment shown can detect multiple dynamic second devices, and the number of multiple second devices in this application is not limited. For ease of understanding, two second devices (Device No. 3 and Device No. 4) are taken as examples for illustration. As Figure 7 shown, the detection environment includes a noise simulation device 1, a noise simulation device 2, a first device, Device No. 3, and Device No. 4. Among them, for the introduction of the noise simulation device, the first device, and the second device, please refer to the Figure 1 example shown, which will not be elaborated here. Device No. 3 and Device No. 4 in the detection environment are of the same type and have a remote log transmission function, which remains enabled during the test. The first device, Device No. 3, and Device No. 4 all have an obstacle avoidance function, so there will be no collision during the actual movement process. Figure 7 The layout of the devices in the test environment shown, the number and types of devices included in the test scenario, are only examples given for understanding the technical solution of this application, and this application includes but is not limited to the above examples.

[0143] In order to more accurately determine the success rate of voice wake-up testing, at least two second devices (such as Device No. 3 and Device No. 4) need to use the same electronic map, and each second device needs to be equipped with a positioning device so that the first device can obtain the real-time coordinates of all second devices in the test environment at the current moment in real time, facilitating the first device to perform position relationship calculation and path planning. Since the second devices are dynamic and their positions are not fixed, fixed cameras and microphones cannot be used to collect information data. Furthermore, the second devices cannot use image information and acoustic information to determine whether the devices are woken up by voice. In order to more accurately complete the voice wake-up success rate test for the second devices, the method of remote Android Debug Bridge (adb) needs to be introduced into the second devices, and the log information of the second devices is obtained through the remote adb method and sent to the first device. It should be particularly noted that for the second devices developed based on the Android system, the remote adb function can be directly enabled. The device under test (such as the second device) can send the log data of the second device to the control terminal by adding a wireless network function to the device through a wireless network USB adapter (WiFi-usb).

[0144] Figure 8 The method provided by the embodiment of the present application is based on Figure 7 The flow diagram of a distributed voice wake-up testing method provided for the shown test environment. As Figure 8 shown, it includes:

[0145] 800. The first device obtains the basic information in the distributed voice wake-up test scenario.

[0146] Exemplarily, the basic parameter information may include the real-time position information of all devices in the second test environment, and the second test environment is a test environment for testing at least two dynamic second devices, such as Figure 7 the shown test environment. The basic parameter information may include the real-time position information of the first device, the real-time position information of at least two second devices, and the real-time position information of at least one noise simulation device.

[0147] For example, Figure 7 after the shown test environment is successfully built, the first device obtains the real-time position information (such as coordinate information) of each device in the test environment based on the positioning module on each device in the test environment. The real-time position information can change dynamically as the device moves, so it can also be called dynamic position information, such as the dynamic position information of the first device, the dynamic position information of at least two second devices, and the dynamic position information of at least one noise simulation device. The first device marks the obtained dynamic position information of each device in the test environment on the detection environment map.

[0148] 801. The first device determines the first detection path corresponding to the first device based on the basic parameter information and the first condition, using the A* path planning algorithm. The first detection path is related to N test positions, where N is an integer greater than or equal to 2.

[0149] Exemplarily, the first condition may include one or more of the access order of the test positions, the priority of access when the path costs are the same, or the first detection path preferentially selecting points close to the test positions. Among them, the priority of access when the path costs are the same is, for example, a point that moves horizontally by a third value or vertically by a fourth value when the path costs are the same. For example, assuming N = 4, that is, there are 4 test positions, the access order of the test positions can be: access in the order of test position 1, test position 2, test position 3, and test position 4. The first detection path preferentially selects points close to the test positions, such as points within the range covered by a radius of a fourth value centered on the test position or points at a straight-line distance of a fifth value from the test position.

[0150] Exemplarily, the first detection path corresponding to the first device can be determined in the following way:

[0151] As Figure 8A shown, a grid map model is constructed according to the situation of obstacles in the distributed voice wake-up test scenario. Based on the grid map model, the starting node (x s , y s ) of the first path, that is, the location of the first device, and the target node (x i , y i ), that is, the first location, which is a point that meets the test requirements and corresponds to test position 1, are determined, and all obstacles (noise simulation devices and multiple movable second devices) are marked in the grid map model.

[0152] 1. The A* path planning algorithm is used to determine the first detection path in the following way.

[0153] First, an Open table and a Close table for the A* path planning algorithm are created. The starting node of the first detection path is placed in the Close table, and the nodes adjacent to the starting node of the first detection path and not coinciding with the obstacles are placed in the Open table. The path cost of the starting node is calculated and stored in the path cost table. All the nodes in the Open table are sorted according to the path cost, and the node with the minimum cost is selected as the next parent node and stored in the Close table. According to the path cost table, the node with the lowest path cost is sequentially found in the process of reaching the target node from the starting node. Among them, the Close table is used to store the nodes that have been expanded as parent nodes, and the Open table is used to store the nodes from which the parent node is selected next time.

[0154] Calculate the path cost of the starting node through the cost function f(x) = g(x) + h(x), where f(x) is the cost estimate from the starting node to the target node, g(x) is the actual cost from the starting node to the intermediate node, and h(x) is the estimated distance from the intermediate node to the target node. As Figure 8B shown, the upper gray area and the right gray area are the paths with the lowest cost. The priority of path node access can be constrained by the first condition when the path costs are the same. For example, the first condition can be to move horizontally first (select the right gray area) or move vertically first (select the upper gray area) when the path costs are the same.

[0155] Assume that the distance of moving one grid horizontally is 10 and the distance of moving one grid diagonally is 14. The method of calculating the path cost of the starting node is as Figure 8C shown. Traverse the 8 nodes (shown as the dark gray grids in the figure) in the neighborhood of the grid where the starting point (x s , y s ) is located, which is the first device 20. Calculate the path cost of each point in the neighborhood through the cost function (already marked in the grid), and take the point with the minimum traversed path cost (14 + 70 = 84) as the parent node and put it into the Close table and mark it as black, as Figure 8D shown.

[0156] Traverse the neighborhood nodes of the point with the minimum traversed path cost (14 + 70 = 84). Similarly, calculate the path cost of each point in the neighborhood through the cost function, and take the point with the minimum path cost (24 + 60 = 84) as the parent node and put it into the close table and mark it as black, as Figure 8E shown.

[0157] Each round of traversal is carried out according to this principle until the target node (x i , y i ) is traversed. As Figure 8F shown, the black area is the first detection path of the first device 20 in the distributed voice wake-up test scenario.

[0158] Exemplarily, during the detection process, due to the setting of obstacles, there is no point with the minimum path cost in the open table 1 in the neighborhood of the point with the current minimum path cost. Then, the first detection path can be determined in the following way:

[0159] As Figure 9 shown, construct a grid map model according to the situation of obstacles in the distributed voice wake-up test scenario, and determine the starting node (x s , y s ) of the first path, which is the location of the first device, and the target node (x i , yi ) and obstacles.

[0160] Create Open Table 1 and Close Table 1 for the A* path planning algorithm. Put the starting node of the first detected path into the Close Table 1, put the nodes adjacent to the starting node of the first detected path and not coinciding with the obstacles into the Open Table 1, calculate the path cost of the starting node, and store it in the Path Cost Table 1. Sort all the nodes in the Open Table 1 according to the path cost, and select the node with the minimum path cost as the next parent node and store it in the Close Table 1. According to the Path Cost Table, sequentially find the node with the lowest path cost during the process of reaching the target node (x s , y s ) from the starting node (x i , y i ). Among them, the Close Table 1 is used to store the nodes that have been expanded as parent nodes, and the Open Table 1 is used to store the nodes from which the parent node is selected next time.

[0161] 1. Traverse the five nodes in the neighborhood near the starting node (x s , y s ), and calculate the path cost of each point in the neighborhood through the cost function (already written in each grid). The point with the lowest path cost in this round of traversal is 10 + 130 = 144. Put this point into the close table 1 and mark it as black, as Figure 9A shown.

[0162] 2. Traverse Figure 9A the neighborhood of the point with the lowest path cost (10 + 130 = 144) in, and also calculate the path cost of each point in the neighborhood through the cost function, as Figure 9B shown. The point with the lowest path cost in this round is 28 + 110 = 138. Put this point into the close table 1 and mark it as black.

[0163] 3. Each round of traversal is carried out according to this principle. Without special circumstances, it will not be elaborated too much. The results of several rounds of traversal are directly given below, as Figures 9C to 9E shown.

[0164] 4. Figure 9E The point with the lowest path cost in the previous round selected in is 62 + 70 = 132. However, after traversing the neighborhood of this point, it is found that there is no point with the lowest path cost in the Open Table 1 in the neighborhood of this point. So at this time, the point with the lowest cost selected from the Open Table 1 is no longer the neighborhood of the point 62 + 70 = 132 (in each round, the point with the lowest cost is selected from the Open Table 1 and put into the colse table 1, but in the previous few rounds, the point with the lowest cost happened to be in the neighborhood of the point with the lowest cost in the previous round). As Figure 9FAs shown, the minimum point selected from the open list 1 in this round is 38 + 100 = 138. At this time, there are two minimum points (38 + 100). The traversal order can be set in the algorithm according to the first condition (preferably select the node to be traversed first). Here, it is assumed that the distances of each neighborhood are calculated clockwise from the neighborhood directly to the right of this neighborhood. Then, the point directly above 28 + 110 = 138 is the minimum node traversed later. To facilitate comparing the differences between these two points, these two points can be selected and put into the close list 1 at the same time, and the neighborhoods of the two points are calculated simultaneously. In this round of traversal, the path cost in the open list 1 is updated. The point (48 + 90) is the point where the distance in this round of traversal is less than the distance in the previous traversal, and the data is updated. For the updated point, the parent node will also change to the minimum point in the previous round.

[0165] 5. Select the point with the minimum distance cost from the openlis after the previous round of traversal. It happens to be the point 48 + 90 = 138 updated in the previous round, as Figure 9G shown.

[0166] 6. As Figure 9H shown, the point with the minimum cost in the previous round was 58 + 80 = 138. However, after traversing the neighborhood of this point, it is found that there is no point with the minimum cost in the open list 1 in the neighborhood of this point. So at this time, the point with the minimum cost selected from the open list 1 is no longer the neighborhood of the point 58 + 80 = 138, but the point with the minimum distance cost in the entire open list 1, which is 24 + 120 = 144. In this round, I still selected two points (24 + 120). These two points are put into the close list 1, and the neighborhoods of the two points are calculated simultaneously. Here, it is assumed that the distances of each neighborhood are calculated clockwise from the neighborhood directly to the right of this neighborhood. The point with the minimum cost distance is traversed, the path cost in the open list 1 is updated, and the data is updated. The determination method is as described above and will not be elaborated.

[0167] 7. After traversing, the target node is finally found, as Figure 9I shown. According to the record of the parent node, backtracking from the target point to the starting point, the shortest path is found, as Figure 9J shown.

[0168] Taking the starting node as (x i , y i ), that is, starting from the first position 1, and the target node as (x j , y j ), that is, taking the first position 2 as the target node, using the same method as above, the detection path from the first position 1 to the first position 2 is obtained. Taking the starting node as (x j , y j ), that is, starting from the first position 2, and the target node as (x k , y k)That is, taking the first position 3 as the target node, in the same manner as above, obtain the detection path from the first position 2 to the first position 3. With the starting node being (x k , y k )That is, taking the first position 3 as the starting point and the target node being (x l , y l )That is, taking the first position 4 as the target node, in the same manner as above, obtain the detection path from the first position 3 to the first position 4. Thus, obtain the first detection path corresponding to the first detection device, and the first detection path needs to pass through the first position 1 corresponding to the first test position, the first position 2 corresponding to the second test position, the first position 3 corresponding to the third test position, and the first position 4 corresponding to the fourth test position.

[0169] 802. The first device, according to the position information of the at least two second devices and the second condition, uses the A* path planning algorithm to determine the detection path corresponding to each of the at least two second devices. The detection path corresponding to each second device includes the N target positions, and instruct each of the at least two second devices to its corresponding detection path.

[0170] Exemplarily, the second condition may include one or more of the following: it is necessary to preferentially select points covering the test positions (such as the detection path corresponding to each second device has to pass through N test positions), the detection paths corresponding to each of the at least two second devices do not overlap at the N detection positions, and the access order of the N test positions (for example, access in the order of test position 1, test position 2, test position 3, and test position 4).

[0171] For example, the at least two second devices may include device three and device four. The first device determines the detection paths corresponding to device three and device four respectively. The specific determination method may refer to the introduction in step 801.

[0172] It should be particularly noted that when using the A* path planning algorithm to determine the detection paths corresponding to the at least two second devices, it is necessary to avoid the intersection of the detection paths. Therefore, during the movement of the device, it is necessary to detect the position changes of other devices in real time and adjust the path in a timely manner. If the conflict cannot be avoided, it is dynamically adjusted according to the preset priority (such as device three waiting). It is also possible to use the spatio-temporal A* path planning method, such that device three avoids the time window when device four occupies the same area in the path planning. It is also possible to calculate the speed adjustment required for obstacle avoidance based on the real-time speed and direction of other devices on the basis of the first detection path generated by the A* path planning algorithm.

[0173] 803. The first device receives the wake-up test information of each of the at least two second devices.

[0174] 804. The first device reports wake-up test information to the control terminal device.

[0175] further, Figure 8 The illustrated embodiment may further include the following step 805 and / or step 806:

[0176] 805. Based on the second basic parameter information and the first condition, use the A* path planning algorithm to optimize the first detection path corresponding to the first device, determine the second detection path, and obtain the wake-up test information determined by the first device on the second detection path.

[0177] 806. Based on the second basic parameter information and the second condition, an A* path planning algorithm is used to optimize the corresponding detection path corresponding to the at least one second device, determine the third detection path corresponding to each second device in the at least one second device, indicate the corresponding third detection path to each second device in the at least one second device, and obtain the wake-up test information determined by each second device in the at least one second device on its corresponding third detection path.

[0178] Exemplarily, the second basic parameter information includes dynamic position information of a device whose position has changed in the second test environment.

[0179] Exemplarily, the method of optimizing the first detection path corresponding to the first device and the method of optimizing the detection path corresponding to at least one second device in step 805 and step 806 can refer to the method shown in step 801 and will not be described in detail here.

[0180] It should be noted that step 801 and step 802 can be executed in any order and can be executed simultaneously. Step 805 and step 806 can be executed in any order and can be executed simultaneously. Step 805 and step 806 can be executed before step 804 or after step 804. If they are executed after step 804, the first device also needs to report the updated wake-up test information to the control terminal device.

[0181] like Figure 10 As shown, the first device plays the wake-up language material at each test position, and the device No. 3 and the device No. 4 send their respective detection information to the first device. After the first device completes the test operation of all test positions, it sends the wake-up test information to the control terminal. After the test is over, the first device returns to its original position.

[0182] In a dynamic testing scenario, the A* path planning algorithm is used to generate a detection path, which can quickly achieve real-time path planning and obstacle avoidance, is suitable for large-scale distributed systems, and the A* algorithm can avoid missed detections by heuristically searching a wide area of the network, is not affected by subjective factors (the state and experience of the inspector), and can find high-quality test paths in complex networks through a global search, dynamic adjustment, and step-by-step optimization scheme.

[0183] Figure 11 A second detection environment built for another execution of distributed voice wake-up, and the second detection environment is used for a dynamic testing scenario. Figure 11 The detection environment shown and Figure 7 The detection environment shown is similar, except that Figure 11 The detection environment shown also includes a control terminal. The control terminal is used to control and manage the wake-up tests of the first device, the noise simulation device, and at least two second devices. In this scenario, the control terminal controls the first device to perform the wake-up test. It should be noted specifically that Figure 11 The layout of the devices in the test environment shown, the number and types of devices included in the test scenario are only examples cited to understand the technical solution of this application, and this application includes but is not limited to the above examples.

[0184] Figure 12 This is a schematic flowchart of a distributed voice wake-up test method provided by an embodiment of this application based on Figure 11 the test environment shown. As Figure 12 shown, it includes:

[0185] 1200. The control terminal obtains the basic parameter information in the distributed voice wake-up test scenario.

[0186] Exemplarily, the basic parameter information includes the dynamic position information of all devices in the second test environment. The second test environment is a test environment for testing dynamic devices. All devices in the second test environment use the same test environment map and update the position information in real time. Among them, all devices in the second test environment may include a control terminal, a first device, at least two second devices, and at least one noise simulation device. The specific acquisition method can refer to Figure 8 the way the first device obtains the basic parameter information in the example shown, and will not be elaborated here.

[0187] 1201. The control terminal, based on the basic parameter information and the first condition, uses the A* path planning algorithm to determine the first detection path corresponding to the first device, and instructs the first device of the first detection path corresponding to the first device. The first detection path is related to N test positions, where N is an integer greater than or equal to 2, and the first condition is determined based on the test requirements.

[0188] Exemplarily, the first condition may include one or more of the access order of the test positions, the access priority when the path costs are the same, or preferentially selecting points close to the test positions in the first detection path. Among them, the access priority when the path costs are the same is, for example, a point where the test position moves horizontally by a third value or vertically by a fourth value when the path costs are the same. For example, assuming N = 4, that is, there are 4 test positions, the access order of the test positions may be: access in the order of test position 1, test position 2, test position 3, and test position 4. The first detection path preferentially selects points close to the test positions, such as points within the coverage range with a radius of the fourth value centered on the test position or points with a straight-line distance of the fifth value from the test position.

[0189] Exemplarily, based on the basic parameter information and the first condition, the control terminal uses the A* path planning algorithm to determine the first detection path corresponding to the first device. The specific determination method can refer to Figure 8 the method for the first device to obtain the first detection path corresponding to the first device using the A* path planning algorithm in the embodiments shown. Details are not described herein again.

[0190] 1202. The control terminal, according to the basic parameter information and the second condition, uses the A* path planning algorithm to determine the detection path corresponding to each of at least two second devices, and instructs each of the at least two second devices of its corresponding detection path. The detection path corresponding to each second device passes through N test positions.

[0191] Exemplarily, the second condition may include one or more of the following: preferentially selecting points passing through the test positions (for example, the detection path corresponding to each second device has to pass through N test positions), the detection paths corresponding to each of the at least two second devices do not overlap at the N detection positions, the access order of the N test positions, and accessing in a preset access order (for example, accessing in the order of test position 1, test position 2, test position 3, and test position 4). The method for the control terminal to determine the detection paths corresponding to at least two second devices respectively can refer to the method for the first device to determine the first detection path in step 801. Details are not described herein again.

[0192] 1203. The control terminal controls the first device to perform a wake-up test on at least two second devices on the first detection path, and controls each second device to perform a wake-up test at the N test positions on its corresponding detection path.

[0193] 1204. The control terminal receives the wake-up test information from the first device and at least two second devices, determines the wake-up test result based on the wake-up test information, and generates a test report.

[0194] Further,Figure 12 The illustrated embodiment may further include the following step 1205 and / or step 1206. The execution order of steps 1205 and 1206 may be before 1204 or after 1204, and there is no excessive limitation on this. The execution order of steps 1205 and 1206 is not in sequence and may also be executed simultaneously.

[0195] 1205. The control terminal optimizes the first detection path corresponding to the first device by using a first algorithm based on the second basic parameter information and the first condition, determines the second detection path, and instructs the first device of the second detection path corresponding to the first device. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment.

[0196] 1206. The control terminal optimizes the detection paths corresponding to at least one second device by using a first algorithm based on the second basic parameter information and the second condition, determines the third detection path corresponding to each second device in at least one second device, and instructs each second device in at least one second device of its corresponding third detection path. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment.

[0197] It should be particularly noted that if steps 1205 and 1206 are executed after 1204, the control terminal device also needs to receive the wake-up test information from the first device and at least one second device, determine the wake-up test result based on the wake-up test information, and generate a test report.

[0198] The control terminal controls the first device to play the wake-up corpus on the first detection path. The first device sends the wake-up test information to the control terminal. Device No. 3 and Device No. 4 respectively send their respective detection information to the control terminal. The control terminal 10 records the relative position relationship between the first device, Device No. 3 and Device No. 4 according to the received wake-up test information, determines the detection results of Device No. 3 and Device No. 4. After completing the test operations at all test positions, the control terminal 10 generates a test report, which details the distributed wake-up experiment situation and its distributed wake-up rate of Device No. 3 and Device No. 4. After the test is over, the first device returns to its original position.

[0199] In the dynamic testing scenario, the A* path planning algorithm is used to generate the detection path, which is fast, suitable for large-scale distributed systems. Compared with manual detection, it takes less time, has higher efficiency and lower cost. Moreover, the A* algorithm can cover a wide area of the network through heuristic search, avoid missed detection, and is not affected by subjective factors (the state and experience of the inspector). Through the scheme of global search, dynamic adjustment and gradual optimization, it can find high-quality test paths in complex networks. And the first condition can be flexibly defined according to the test requirements to adapt to different test objectives, effectively meeting the test requirements, node coverage requirements, environment simulation requirements, performance optimization and other requirements in the distributed voice wake-up detection path, ensuring that the test process is more comprehensive, efficient and adaptable. Compared with the method of managing distributed voice wake-up testing by the first device, the control terminal has a longer battery life for managing distributed voice wake-up testing and can execute test tasks more stably in the long term.

[0200] The embodiment of the present application also provides a distributed voice wake-up testing system, which can be applied to Figure 1 the test environment shown in the figure, and is used for static precise testing scenarios and static wide testing scenarios, including the first device with the same test environment map, at least one noise simulation device, at least one visual signal processing device, at least one acoustic signal processing device, and at least two second devices. The testing system includes:

[0201] The first device is used to obtain the basic parameter information in the distributed voice wake-up testing scenario. The basic parameter information includes the position information of all devices in the first test environment. Based on the basic parameter information and the first condition, the first device uses the first algorithm to determine the first detection path corresponding to the first device. The first detection path is related to N test positions, where N is an integer greater than or equal to 2, and the first condition is determined based on the test requirements;

[0202] At least one noise simulation device is used to simulate different noises;

[0203] The first device is used to perform wake-up testing on the first detection path corresponding to the first device;

[0204] At least one visual signal processing device and at least one acoustic signal processing device are used to obtain and report the wake-up test information determined by performing wake-up testing on the first detection path;

[0205] The first device is used to receive the wake-up test information, and the wake-up test information is used to determine the wake-up test result.

[0206] Exemplarily, the first condition includes one or more of the number of steps, step length, and direction change probability set according to the test requirements; or the access order of N test positions; or the distance minimization of the first detection path. The first algorithm is a random walk algorithm.

[0207] Exemplarily, the first algorithm is a simulated annealing algorithm, and the first condition includes one or more of the following: N test positions determined according to test requirements; or, the distance between two adjacent test positions among the N test positions is less than or equal to a first value; or, the distance between each test position among the N test positions and at least one noise simulation device is less than or equal to a second value; or the distance from each of at least two second devices to the nearest test position is the same.

[0208] For the specific description and beneficial effects of the system, reference can be made to Figure 1 and Figure 2 the embodiments shown, which will not be elaborated here.

[0209] The embodiment of the present application further provides a distributed voice wake-up test system, which can be applied to Figure 5 the test environment shown, and is used for static fine measurement test scenarios and static wide measurement test scenarios, including a test environment map control terminal, a first device, at least one noise simulation device, at least one visual signal processing device, at least one acoustic signal processing device, and at least two second devices that use the same. The distributed voice wake-up test system includes:

[0210] A control terminal, configured to obtain basic parameter information in the distributed voice wake-up test scenario, where the basic parameter information includes the position information of all devices in the first test environment, and based on the basic parameter information and the first condition, determine the first detection path corresponding to the first device by using the first algorithm, and instruct the first device to execute a wake-up test on the first detection path corresponding to the first device; the first detection path is related to N test positions, N is an integer greater than or equal to 2, and the first condition is determined based on test requirements;

[0211] At least one noise simulation device, configured to simulate different noises;

[0212] At least one visual signal processing device and at least one acoustic signal processing device, configured to obtain and report the wake-up test information determined by the first device when executing the wake-up test on its corresponding first detection path;

[0213] The control terminal is configured to receive the wake-up test information, determine the wake-up test result based on the wake-up test information, and generate a distributed voice wake-up test report.

[0214] Exemplarily, the first algorithm is a random walk algorithm, and the first condition includes one or more of the number of steps, step length, and direction change probability set according to test requirements; or the access order of N test positions; or the minimization of the distance of the first detection path.

[0215] Exemplarily, the first algorithm is a simulated annealing algorithm, and the first condition includes one or more of the following: N test positions determined according to test requirements; or, the distance between two adjacent test positions among the N test positions is less than or equal to a first value; or, the distance between each test position among the N test positions and at least one noise simulation device is less than or equal to a second value; or, the distance between each of at least two second devices and the nearest test position is the same.

[0216] For the specific description and beneficial effects of this system, reference can be made to Figure 5 and Figure 6 the embodiments shown, which will not be elaborated here.

[0217] The embodiment of the present application also provides a distributed voice wake-up test system, which can be applied to Figure 7 the test environment shown, and is used for dynamic test scenarios, including a first device, at least one noise simulation device, and at least two second devices that use the same test environment map. The test environment map is updated in real time for location, including:

[0218] The first device is used to obtain the basic parameter information in the distributed voice wake-up test scenario. The basic parameter information includes the dynamic position information of all devices in the dynamic test scenario; based on the basic parameter information and the first condition, the A* path planning is used to determine the first detection path corresponding to the first device. The first detection path is related to N test positions, where N is an integer greater than or equal to 2. On the first detection path, wake-up tests are performed on at least two second devices to obtain wake-up test information, and the wake-up test information is used to determine the wake-up test result. Among them, the first condition is determined based on test requirements, and the first condition includes one or more of preferentially selecting points close to N test positions based on test requirements, the access order of N test positions, or minimizing the distance of the first detection path;

[0219] The first device is also used to determine the detection path corresponding to each of at least two second devices according to the basic parameter information and the second condition, and use the A* path planning algorithm to indicate the detection path corresponding to each of at least two second devices to each of at least two second devices. The detection path corresponding to each second device passes through N test positions; where the second condition includes one or more of the following: the detection paths corresponding to each of at least two second devices do not overlap at the N test positions, the access order of the N test positions, the priority of path node access when the path costs are the same, and minimizing the distance of the detection path corresponding to each second device;

[0220] At least one noise simulation device is used to simulate different noises.

[0221] Each of at least two second devices is configured to perform a wake-up test at N test positions on its corresponding detection path and send wake-up test information to the first device;

[0222] Exemplarily, the first device is further configured to optimize the first detection path corresponding to the first device by using the A* path planning algorithm based on the second basic parameter information and the first condition, and determine a second detection path. The first device performs a wake-up test on the second detection path. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment; and / or,

[0223] The first device is further configured to optimize the detection paths corresponding to at least one second device by using the A* path planning algorithm based on the second basic parameter information and the second condition, determine the third detection path corresponding to each of at least one second device, indicate the corresponding third detection path to each of at least one second device, and control each of at least one second device to perform a wake-up test on its corresponding third detection path; The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment;

[0224] Each of at least one second device is configured to perform a wake-up test on its corresponding third detection path and report wake-up test information to the first device;

[0225] For the specific description and beneficial effects of this system, reference can be made to Figure 7 and Figure 8 the embodiments shown, which will not be elaborated herein.

[0226] An embodiment of the present application further provides a distributed voice wake-up test system, which can be applied to Figure 11 the test environment shown, and is used for dynamic test scenarios, including a control terminal, a first device, at least one noise simulation device, and at least two second devices that use the same test environment map. The test environment map is updated in real time for position. The test system includes:

[0227] A control terminal, configured to obtain basic parameter information in a distributed voice wake-up test scenario, where the basic parameter information includes dynamic position information of all devices in the dynamic test scenario; based on the basic parameter information and a first condition, use the A* path planning algorithm to determine a first detection path corresponding to a first device, and instruct the first device of the first detection path corresponding to the first device. The first detection path is related to N test positions, where N is an integer greater than or equal to 2. The control terminal controls the first device to perform a wake-up test on at least two second devices on the first detection path. Wherein, the first condition is determined based on the test requirements, and the first condition includes one or more of the following: preferentially select points close to the N test positions based on the test requirements, the access order of the N test positions, and the priority of accessing path nodes when the path costs are the same;

[0228] A first device, configured to perform a wake-up test on the first detection path corresponding to the first device and send wake-up test information to the control terminal;

[0229] The control terminal is further configured to use the A* path planning algorithm to determine a detection path corresponding to each of at least two second devices based on the basic parameter information and the first condition, and instruct each of the at least two second devices of the detection path corresponding to each second device. The detection path corresponding to each second device passes through N test positions, and controls each second device to perform a wake-up test at the N test positions on its corresponding detection path; wherein, the second condition includes one or more of the following: preferentially select points passing through the N test positions, the detection paths corresponding to each of the at least two second devices do not overlap at the N test positions, the access order of the N test positions, or the priority of accessing path nodes when the path costs are the same;

[0230] Each of at least two second devices is configured to perform a wake-up test at the N test positions on the detection path corresponding to each second device and send wake-up test information to the control terminal;

[0231] The control terminal is configured to determine a distributed voice wake-up test result based on the wake-up test information sent by the first device and the wake-up test information sent by each second device, and generate a distributed voice wake-up test report;

[0232] At least one noise simulation device, configured to simulate different noises.

[0233] Exemplarily, the control terminal: based on the second basic parameter information and the first condition, uses the A* path planning algorithm to optimize the first detection path corresponding to the first device, determines a second detection path, instructs the first device of the second detection path, and controls the first device to perform a wake-up test on the second detection path. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment.

[0234] Exemplarily, a first device is configured to perform a wake-up test on a second detection path and report the obtained wake-up test information on the second detection path to a control terminal.

[0235] Exemplarily, a control terminal: is configured to optimize the detection paths corresponding to at least one second device by using the A* path planning algorithm based on second basic parameter information and a second condition, determine the third detection path corresponding to each second device among the at least one second device, instruct each second device among the at least one second device of its corresponding third detection path, control each second device among the at least one second device to perform a wake-up test on its corresponding third detection path, and the second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment.

[0236] Exemplarily, each second device among the at least one second device is configured to perform a wake-up test on its corresponding third detection path and report the obtained wake-up test information on the third detection path to the control terminal.

[0237] Exemplarily, the control terminal is configured to determine a distributed voice wake-up test result according to the wake-up test obtained by the first device on the second detection path and the wake-up test information obtained by at least one second device on its corresponding third detection path, and generate a distributed voice wake-up test report.

[0238] For the specific description and beneficial effects of this system, reference can be made to Figure 11 and Figure 12 the embodiments shown, which will not be elaborated here.

[0239] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0240] To facilitate better implementation of the above solutions of the embodiments of this application, the following also provides related devices for implementing the above solutions.

[0241] Please refer to Figure 13, which is a schematic structural diagram of a communication device provided by an embodiment of the present application. The communication device 1300 may include a transceiver module 1301 (sometimes also referred to as a transceiver unit) and a processing module 1302 (sometimes also referred to as a processing unit). The transceiver module can implement the sending function and the receiving function. When the transceiver module implements the sending function, it can be referred to as a sending module (sometimes also referred to as a sending unit). When the transceiver module implements the receiving function, it can be referred to as a receiving module (sometimes also referred to as a receiving unit). The sending module and the receiving module can be the same functional module, and this functional module is called the transceiver module, which can implement the sending function and the receiving function; or, the sending module and the receiving module can be different functional modules, and the transceiver module is a general term for these functional modules.

[0242] In some possible implementation manners, the communication device 1300 provided in the embodiment of the present application further includes: a storage module (sometimes also referred to as a storage unit), which is used to store any data, computer instructions, and / or computer programs that may be involved in the embodiments of the present application.

[0243] The transceiver module 1301, the processing module 1302, and the storage module in the embodiment of the present application are used to enable the communication device 1300 to implement the functions of the terminal device in the above method embodiment, or to enable the communication device 1300 to implement the functions of the network device in the above method embodiment.

[0244] The following takes the communication device 1300 for implementing the function of the first device shown in the above method embodiment Figure 2 to illustrate each module in the communication device.

[0245] In some possible implementation manners, in the communication device 1300 provided in the embodiment of the present application, the transceiver module 1301 is used to obtain the basic parameter information in the distributed voice wake-up test scenario, and the processing module 1302 is used to determine the first detection path corresponding to the first device by using the random walk algorithm based on the basic parameter information and the first condition. The first detection path is related to N test positions, where N is an integer greater than or equal to 2, and the first condition is determined based on the test requirements; the transceiver module 1301 is used to obtain the wake-up test information, and the wake-up test information is obtained by performing a wake-up test on at least two second devices on the first detection path, and the wake-up test information is used to determine the wake-up test result. The first condition includes one or more of the number of steps, step size, and direction change probability set according to the test requirements; or, the access order of the N test positions; or the distance minimization of the first detection path.

[0246] In some possible implementation manners, in the communication device 1300 provided in the embodiments of the present application, a transceiver module 1301 is configured to obtain basic parameter information in the distributed voice wake-up test scenario, and a processing module 1302 is configured to determine a first detection path corresponding to a first device by using a simulated annealing algorithm based on the basic parameter information and a first condition. The first detection path is related to N test positions, where N is an integer greater than or equal to 2, and the first condition is determined based on test requirements; the transceiver module 1301 is configured to obtain wake-up test information, where the wake-up test information is obtained by performing a wake-up test on at least two second devices on the first detection path, and the wake-up test information is used to determine a wake-up test result. The first condition includes one or more of the following: the N test positions determined according to test requirements; or, the access order of the N test positions; or, the distance between two adjacent test positions among the N test positions is less than or equal to a first value; or, the distance between each test position among the N test positions and at least one noise simulation device is less than or equal to a second value; or, the distance between each of the at least two second devices and the nearest test position is the same; or, the distance of the first detection path is minimized.

[0247] The following takes the communication device 1300 for implementing the function of the first device shown in the above method embodiment Figure 8 to describe each module in the communication device.

[0248] In some possible implementation manners, in the communication device 1300 provided in the embodiments of the present application, a transceiver module 1301 is configured to obtain basic parameter information in the distributed voice wake-up test scenario; the processing module 1302 is further configured to determine a first detection path corresponding to the first device by using the A* path planning algorithm according to the basic parameter information and the first condition. The first detection path is related to the N test positions. The first condition includes one or more of the following: preferentially selecting a point close to the N test positions based on test requirements; or, the access order of the N test positions; or, the distance of the first detection path is minimized; or, the priority of path node access in the case of the same path cost.

[0249] Exemplarily, the processing module 1302 is further configured to, according to the basic parameter information and the second condition, use the A* path planning algorithm to determine the detection path corresponding to each of the at least two second devices. The detection path corresponding to each second device passes through the N test positions, and instruct each of the at least two second devices to indicate the detection path corresponding to each second device. The detection path corresponding to each second device is used for each second device to perform a wake-up test at the N test positions on its corresponding detection path. The second condition includes one or more of the following: the detection paths corresponding to each of the at least two second devices do not overlap at the N detection positions, the access order of the N test positions, the points that need to preferentially pass through the N test positions, and the priority of path node access when the path costs are the same.

[0250] The processing module 1302 is further configured to optimize the first detection path corresponding to the first device by using the A* path planning algorithm based on the second basic parameter information and the first condition, determine the second detection path, and perform a wake-up test on the second detection path. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment, and based on the second basic parameter information and the second condition, optimize the detection paths corresponding to at least one second device by using the A* path planning algorithm, determine the third detection path, and instruct each of the at least one second device to indicate its corresponding third detection path, and control each of the at least one second device to perform a wake-up test on its corresponding third detection path. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment.

[0251] The following takes the communication device 1300 for implementing the functions of the control terminal device shown in the above method embodiment Figure 12 to illustrate each module in the communication device.

[0252] In some possible implementation manners, in the communication device 1300 provided in the embodiment of the present application, the transceiver module 1301 is configured to obtain the basic parameter information in the distributed voice wake-up test scenario, and the processing module 1302 is configured to use the first algorithm to determine the first detection path corresponding to the first device based on the basic parameter information and the first condition, and instruct the first device to indicate the first detection path corresponding to the first device. The first detection path is related to N test positions, where N is an integer greater than or equal to 2, and the first condition is determined based on the test requirements. The processing module 1302 is configured to control the terminal to control the first device to perform a wake-up test on at least two second devices at the N test positions on the first detection path. The transceiver module 1301 is configured to receive the wake-up test information, and the processing module 1302 is configured to determine the wake-up test result based on the wake-up test information and generate a test report.

[0253] Exemplarily, the processing module 1302 is further configured to determine, according to the basic parameter information and the second condition, a detection path corresponding to each of at least two second devices by using the A* path planning algorithm, and indicate the corresponding detection path to each of the at least two second devices. The detection path corresponding to each second device passes through N test positions. The second condition may include one or more of the following: the points passing through the N test positions need to be preferentially selected (for example, the detection path corresponding to each second device needs to pass through N test positions), the detection paths corresponding to each of the at least two second devices do not overlap at the N detection positions, the access order of the N test positions, and the priority of path node access when the path costs are the same.

[0254] The processing module 1302 is further configured to optimize the first detection path corresponding to the first device by using the A* path planning algorithm based on the second basic parameter information and the first condition, determine the second detection path, indicate the second detection path to the first device, and control the first device to perform a wake-up test on the second detection path. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment. And based on the second basic parameter information and the second condition, the processing module 1302 is further configured to optimize the detection paths corresponding to at least one second device by using the A* path planning algorithm, indicate the corresponding third detection path to each of the at least one second devices, and control each of the at least one second devices to perform a wake-up test on its corresponding third detection path. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment.

[0255] The communication device 1300 for implementing the functions of controlling the terminal device in the above method embodiments may refer to Figure 6 the embodiments shown in Figure 12 and

[0256] It should be particularly emphasized that the entity device corresponding to the above transceiver module 1301 may be a transceiver, the entity device corresponding to the processing module 1302 may be a processor, and the entity device corresponding to the storage module may be a memory.

[0257] In the figure, 1400 is an example of the composition of another communication device provided by an embodiment of the present application. The communication device 1400 may be a first device or a control unit. FIG. 1400 shows a schematic diagram of a simplified structure of the communication device. The device includes a part 1401, a part 1402, and a part 1403. The part 1402 is usually referred to as a processor and is used for the first device or the control terminal to execute the methods in the above method embodiments. The part 1401 is mainly used for storing computer program codes and data. The part 1403 is usually referred to as a transceiver module, a transceiver, a transceiver circuit, or a transceiver, etc. The devices used to implement the receiving function in the part 1403 can be regarded as a receiver, and the devices used to implement the sending function can be regarded as a transmitter, that is, the part 1403 includes a receiver 1432 and a transmitter 1431. The receiver can also be referred to as a receiving module, a receiver, or a receiving circuit, etc., and the transmitter can be referred to as a transmitting module, a transmitter, or a transmitting circuit, etc.

[0258] The part 1401 and the part 1402 may include one or more single boards, and each single board may include one or more processors and one or more memories. The processor is used to read and execute the programs in the memory to implement baseband processing functions and the control of the base station. If there are multiple single boards, the single boards can be interconnected to enhance the processing ability. As an optional implementation manner, it can also be that multiple single boards share one or more processors, or multiple single boards share one or more memories, or multiple single boards simultaneously share one or more processors.

[0259] For example, in one implementation manner, the transceiver module of the part 1403 is used to execute the transceiver-related processes performed by the first device or the control terminal in the foregoing method embodiments. The processor of the part 1402 is used to execute the processing-related processes performed by the first device or the control terminal in the foregoing method embodiments.

[0260] An embodiment of the present application also provides a computer-readable storage medium storing program instructions that can be run by a processor, and the program instructions are used to implement the steps in any of the above wake-up test method embodiments.

[0261] In the above solution, the computer-readable storage medium stores data obtained through a visual signal processing device or an acoustic signal processing device, or is used to store the detection data of the first device or at least two second devices, so as to determine the wake-up situation of at least two second devices, thereby reducing the probability of misdetection and false recording, and further meeting the requirements of intelligent home appliance quality detection.

[0262] An embodiment of the present application also provides a computer program product including instructions that, when run on a computer, cause the computer to execute the methods described in any of the above embodiments.

[0263] In some embodiments, the functions or modules included in the apparatus provided by the embodiments of the present disclosure can be used to execute the methods described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments. For the sake of brevity, it will not be repeated here.

[0264] The above descriptions of the various embodiments tend to emphasize the differences between the various embodiments. The similarities or similarities between them can be referred to each other. For the sake of brevity, they will not be repeated here.

[0265] In several embodiments provided in the present application, it should be understood that the disclosed methods and apparatuses can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0266] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0267] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0268] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

[0269] If the technical solution of this application involves personal information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, the personal information processing rules are notified by obvious signs / information, and the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

[0270] In addition, an operating system is running on the above components. For example, an iOS operating system, an Android operating system, a Windows operating system, etc. Applications can be installed and run on the operating system. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the explanation and beneficial effects of the relevant contents in any of the above-mentioned communication devices can refer to the corresponding method embodiments provided above, and will not be repeated here.

[0271] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection between each other can be an indirect coupling or communication connection through some interfaces, devices, or modules, and can be in electrical, mechanical, or other forms.

[0272] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they may be located in one place, or may be distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0273] In addition, in each embodiment of the present application, the functional modules can be integrated into a processing module, or each module can exist physically alone, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0274] If the above integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the part that essentially contributes to the technical solution of the present application, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the processes of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories, random access memories, magnetic disks, or optical discs that can store program codes.

[0275] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of each embodiment of the present application.

Claims

1. A wake-up test method, characterized in that, An electronic device applied to a distributed voice wake-up test scenario, including: Obtain basic parameter information in the distributed voice wake-up test scenario; Based on the basic parameter information and the first condition, use a first algorithm to determine a first detection path corresponding to a first device. The first detection path is related to N test positions, where N is an integer greater than or equal to 2, and the first condition is determined based on test requirements; Obtain wake-up test information, which is obtained by performing wake-up tests on at least two second devices on the first detection path. The wake-up test information is used to determine the wake-up test result.

2. The method according to claim 1, wherein The basic parameter information includes the position information of all devices in a first test environment, which is a test environment for static wide testing of static devices or a test environment for static precise testing of static devices in a distributed voice wake-up test scenario. All devices in the first test environment use the same test environment map.

3. The method according to claim 2, wherein The first condition includes one or more of the following: The number of steps, step size, and direction change probability set according to test requirements; or, The access order of the N test positions; or, Minimization of the distance of the first detection path.

4. The method according to claim 3, wherein The first algorithm is a random walk algorithm. Based on the basic parameter information and the first condition, using the first algorithm to determine the first detection path corresponding to the first device includes: Based on the basic parameter information and the first condition, use the random walk algorithm to determine the first detection path corresponding to the first device. The first detection path passes through the N test positions.

5. The method according to claim 2, wherein The first condition includes one or more of the following: The N test positions determined according to test requirements; or, The access order of the N test positions; or, The distance between two adjacent test positions among the N test positions is less than or equal to a first value; or, The distance between each test position among the N test positions and at least one noise simulation device is less than or equal to a second value; or, The distance from each of the at least two second devices to the nearest test position is the same; or, Minimization of the distance of the first detection path.

6. The method according to claim 1, characterized in that, The basic parameter information includes the dynamic position information of all devices in a second test environment, which is a test environment for testing dynamic devices in a distributed voice wake-up test scenario. All devices in the second test environment use the same test environment map and update the position information in real time.

7. The method according to claim 6, wherein The first condition includes one or more of the following: Prioritize selecting points close to the N test positions based on test requirements; or, The access order of the N test positions; or, The priority of path node access when the path costs are the same.

8. The method according to claim 6, wherein The first algorithm is an A* path planning algorithm. Based on the basic parameter information and the first condition, using the first algorithm to determine the first detection path corresponding to the first device includes: According to the basic parameter information and the first condition, use the A* path planning algorithm to determine the first detection path corresponding to the first device, where the first detection path is related to the N test positions.

9. The method according to any one of claims 6 to 8, characterized in that The method further includes: According to the basic parameter information and the second condition, use the A* path planning algorithm to determine the detection path corresponding to each of the at least two second devices, and indicate to each of the at least two second devices the detection path corresponding to each second device. The detection path corresponding to each second device passes through the N test positions, and the detection path corresponding to each second device is used for each second device to perform a wake-up test at the N test positions on its corresponding detection path.

10. The method according to claim 9, wherein The second condition includes one or more of the following: It is necessary to preferentially select points passing through the N test positions; or, The detection paths corresponding to each of the at least two second devices do not overlap at the N test positions; Or, The access order of the N test positions; or, The priority of path node access in the case of the same path cost.

11. The method according to any one of claims 7 to 10, characterized in that The method further includes: Based on the second basic parameter information and the first condition, use the A* path planning algorithm to optimize the first detection path corresponding to the first device, and determine the second detection path corresponding to the first device. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment; and / or Based on the second basic parameter information and the second condition, use the A* path planning algorithm to optimize the detection paths corresponding to at least one second device, and determine the third detection path corresponding to each of the at least one second device. Indicate to each of the at least one second device the corresponding third detection path. The second basic parameter information includes the dynamic position information of the devices whose positions change in the second test environment.

12. A distributed voice wake-up test system, characterized in that, Applied to static fine measurement test scenarios and static wide measurement test scenarios, it includes a first device, at least one noise simulation device, at least one visual signal processing device, at least one acoustic signal processing device, and at least two second devices that use the same test environment map. Among them, The first device is used to obtain the basic parameter information in the distributed voice wake-up test scenario. The basic parameter information includes the position information of all devices in the first test environment. Based on the basic parameter information and the first condition, use the first algorithm to determine the first detection path corresponding to the first device. The first detection path is related to N test positions, where N is an integer greater than or equal to 2, and the first condition is determined based on test requirements; The at least one noise simulation device is used to simulate different noises; The at least one visual signal processing device and the at least one acoustic signal processing device are used to obtain the wake-up test information obtained by performing wake-up tests on the at least two second devices at the N test positions, and report the wake-up test information to the first device; The first device is used to obtain the wake-up test information, and the wake-up test information is used to determine the wake-up test result.

13. The system according to claim 12, wherein The first algorithm is the simulated annealing algorithm, and the first condition includes one or more of the following: the N test positions determined according to the test requirements; or, the access order of the N test positions; or, the distance between two adjacent test positions among the N test positions is less than or equal to a first value; or, the distance between each test position among the N test positions and at least one noise simulation device is less than or equal to a second value; or, the distance from each of the at least two second devices to the nearest test position is the same; or, the distance of the first detection path is minimized.

14. A distributed voice wake-up test system, characterized in that, Applied to the scenario of dynamic testing, including a first device, at least one noise simulation device, and at least two second devices that use the same test environment map, and the position of the test environment map is updated in real time. The system includes: The first device is configured to obtain the basic parameter information in the distributed voice wake-up test scenario, where the basic parameter information includes the dynamic position information of all devices in the dynamic test scenario; based on the basic parameter information and the first condition, use the A* path planning algorithm to determine the first detection path corresponding to the first device, the first detection path is related to N test positions, N is an integer greater than or equal to 2, and perform wake-up tests on at least two second devices on the first detection path to obtain wake-up test information. Among them, the first condition is determined based on the test requirements, and the first condition includes: preferentially selecting points close to the N test positions based on the test requirements, the access order of the N test positions, or one or more of the priority of accessing path nodes when the path costs are the same; The first device is configured to use the A* path planning algorithm to determine the detection path corresponding to each of the at least two second devices according to the basic parameter information and the second condition, and indicate the detection path corresponding to each second device to each of the at least two second devices. The detection path corresponding to each second device passes through the N test positions, and control each second device to perform wake-up tests at the N test positions on its corresponding detection path; Each of the second devices is configured to perform wake-up tests at the N test positions on the corresponding detection path and send wake-up test information to the first device; Among them, the second condition includes one or more of preferentially selecting points passing through the N test positions, the detection paths corresponding to each of the at least two second devices do not overlap at the N test positions, the access order of the N test positions, or the priority of accessing path nodes when the path costs are the same; The at least one noise simulation device is configured to simulate different noises.

15. The system according to claim 14, wherein, The first device is further configured to optimize the first detection path corresponding to the first device based on the second basic parameter information and the first condition by using the first algorithm to determine the second detection path, and the first device performs wake-up tests on the second detection path. The second basic parameter information includes the dynamic position information of the devices whose positions have changed in the second test environment; And / or, based on the second basic parameter information and the second condition, optimize the detection path corresponding to the at least one second device by using the first algorithm, determine the third detection path corresponding to each second device in the at least one second device, indicate the corresponding third detection path to each second device in the at least one second device, and control each second device in the at least one second device to perform a wake-up test on its corresponding third detection path; Each second device in the at least one second device is configured to perform a wake-up test on its corresponding third detection path and report wake-up test information to the first device.

Images