Abnormality detection method and device for voice equipment and electronic equipment

Abstract

The embodiment of the invention provides an anomaly detection method and device for voice equipment and electronic equipment, relates to the technical field of audio detection, and is used for improving the anomaly detection accuracy of the voice equipment. The method comprises the following steps: acquiring an audio test signal; controlling the at least one loudspeaker to play the audio test signal in sequence, and controlling each of the at least one sound pick-up to collect the currently played audio test signal when each loudspeaker plays the audio test signal to obtain a collection result corresponding to each loudspeaker; wherein the acquisition result comprises at least one sub-acquisition result, and the at least one sub-acquisition result is in one-to-one correspondence with the at least one sound pick-up; any two loudspeakers in the at least one loudspeaker do not play the audio test signal at the same moment; and based on the collection result corresponding to the at least one loudspeaker, determining an abnormal detection result of the at least one loudspeaker and the at least one pickup.

Description

Technical Field

[0001] This application relates to the technical field of audio detection, and particularly to an abnormal detection method, device and electronic device for voice devices. Background Art

[0002] With the gradual development of monitoring technology, the audio monitoring system combining the monitoring system and the audio system is growing rapidly in applications in multiple industries, especially in fields such as security, smart home, medical and industrial. The progress of technology has made the audio monitoring system more and more intelligent and efficient, and can be combined with other technologies (such as artificial intelligence and Internet of Things).

[0003] The combination of the monitoring system and the audio system can enable the monitoring system to provide more security options and functions for the audio system, and the audio system to provide better sound quality and more entertainment for the monitoring system. And as the most basic and important pick-up and speaker in the audio monitoring system, whether the functions are normal is also a prerequisite for determining whether the audio monitoring can work properly.

[0004] In related technologies, usually the pick-up or speaker in the audio monitoring system is detected, and there is usually a problem of being easily interfered by environmental noise, and these noises will affect the accuracy of the audio monitoring system, resulting in low detection accuracy. Summary of the Invention

[0005] Embodiments of this application provide an abnormal detection method, device and electronic device for voice devices, so as to improve the accuracy of abnormal detection of voice devices.

[0006] In a first aspect, embodiments of this application provide an abnormal detection method for voice devices. The voice devices include at least one speaker and at least one pick-up, and the method includes:

[0007] Obtain an audio test signal;

[0008] Control the at least one speaker to play the audio test signal in sequence, and when each speaker plays, control each pick-up in the at least one pick-up to respectively collect the currently played audio test signal, and obtain a collection result corresponding to each speaker; wherein, the collection result includes at least one sub-collection result, and the at least one sub-collection result corresponds one-to-one with the at least one pick-up; any two speakers in the at least one speaker do not play the audio test signal at the same time;

[0009] Based on the collection results corresponding to the at least one speaker respectively, determine the abnormal detection results of the at least one speaker and the at least one pick-up.

[0010] In a possible implementation, the obtaining of the audio test signal includes: selecting multiple target audio signals from multiple pre-stored audio signals according to the frequency range supported by the at least one speaker; combining the multiple target audio signals in a randomly arranged manner to generate the audio test signal.

[0011] In a possible implementation, the combining the multiple target audio signals in a randomly arranged manner to generate the audio test signal includes: randomly selecting a group from multiple pre-stored groups of candidate playback parameters as the test playback parameters; adjusting the gains of the multiple target audio signals respectively according to the test playback parameters, and combining the adjusted multiple target audio signals in a randomly arranged manner to generate the audio test signal.

[0012] In a possible implementation, when the voice device includes one speaker and one microphone, the determining of the abnormality detection results of the at least one speaker and the at least one microphone based on the collection results respectively corresponding to the at least one speaker includes: when the collection results indicate that the audio test signal is not collected, determining that the speaker and / or the microphone is abnormal; when the collection results indicate that the audio test signal is collected, analyzing the collected audio test signal included in the collection results to obtain the collection signal parameters and the collection playback parameters; matching the collection signal parameters with the test signal parameters of the audio test signal, and matching the collection playback parameters with the test playback parameters of the audio test signal, and when the collection signal parameters or the collection playback parameters fail to match, determining that the speaker and / or the microphone is abnormal; when both the collection signal parameters and the collection playback parameters match successfully, determining that both the speaker and the microphone are not abnormal.

[0013] In a possible implementation, when the voice device includes one speaker and multiple pickups, determining the anomaly detection results of the at least one speaker and the at least one pickup based on the respective acquisition results of the at least one speaker includes: when any one of the sub-acquisition results included in the acquisition result indicates that the audio test signal is not acquired, determining that there is an anomaly in the speaker and / or the multiple pickups; when multiple sub-acquisition results included in the acquisition result all indicate that the audio test signal is acquired, respectively analyzing the acquired audio test signals included in each of the multiple sub-acquisition results to obtain the acquisition signal parameters and acquisition playback parameters corresponding to each sub-acquisition result; matching the acquisition signal parameters corresponding to each of the multiple sub-acquisition results with the test signal parameters of the audio test signal respectively, matching the acquisition playback parameters corresponding to each of the multiple sub-acquisition results with the test playback parameters of the audio test signal respectively, and when any one of the acquisition signal parameters or any one of the acquisition playback parameters fails to match, determining that there is an anomaly in the speaker and / or the multiple pickups; when the acquisition signal parameters and acquisition playback parameters corresponding to each of the multiple sub-acquisition results all match successfully, determining that neither the speaker nor the multiple pickups has an anomaly.

[0014] In a possible implementation, when any one of the sub-acquisition results included in the acquisition result indicates that the audio test signal is not acquired, determining that there is an anomaly in the speaker or the multiple pickups includes: when some of the sub-acquisition results included in the acquisition result indicate that the audio test signal is not acquired, determining that some of the multiple pickups have an anomaly; when multiple sub-acquisition results included in the acquisition result all indicate that the audio test signal is not acquired, determining that there is an anomaly in the speaker or the multiple pickups.

[0015] In a possible implementation, when the voice device includes multiple speakers and one pickup, determining the abnormality detection results of the at least one speaker and the at least one pickup based on the respective acquisition results corresponding to the at least one speaker includes: for any one of the acquisition results corresponding to the multiple speakers, perform the following operations: when any one of the acquisition results indicates that the audio test signal is not acquired, determine that the speaker corresponding to the any one of the acquisition results and / or the pickup is abnormal; when any one of the acquisition results indicates that the audio test signal is acquired, analyze the acquired audio test signal included in the any one of the acquisition results to obtain an acquisition signal parameter and an acquisition playback parameter; match the acquisition signal parameter with the test signal parameter of the audio test signal, and match the acquisition playback parameter with the test playback parameter of the audio test signal, and when the acquisition signal parameter or the acquisition playback parameter fails to match, determine that the speaker corresponding to the any one of the acquisition results and / or the pickup is abnormal; when both the acquisition signal parameter and the acquisition playback parameter match successfully, determine that neither the speaker nor the pickup corresponding to the any one of the acquisition results is abnormal.

[0016] In a possible implementation, when the voice device includes multiple speakers and multiple pickups, determining the abnormality detection results of the at least one speaker and the at least one pickup based on the respective acquisition results corresponding to the at least one speaker includes: for any one of the acquisition results corresponding to the multiple speakers, perform the following operations: when any one of the sub-acquisition results included in the any one of the acquisition results indicates that the audio test signal is not acquired, determine that the speaker corresponding to the any one of the acquisition results and / or the multiple pickups is abnormal; when all of the multiple sub-acquisition results included in the any one of the acquisition results indicate that the audio test signal is acquired, for the acquired audio test signals included in the multiple sub-acquisition results respectively, analyze each to obtain the acquisition signal parameter and the acquisition playback parameter corresponding to each sub-acquisition result; match the acquisition signal parameters corresponding to the multiple sub-acquisition results respectively with the test signal parameter of the audio test signal, match the acquisition playback parameters corresponding to the multiple sub-acquisition results respectively with the test playback parameter of the audio test signal, and when any one of the acquisition signal parameters or any one of the acquisition playback parameters fails to match, determine that the speaker corresponding to the any one of the acquisition results and / or the multiple pickups is abnormal; when the acquisition signal parameters and the acquisition playback parameters corresponding to the multiple sub-acquisition results all match successfully, determine that neither the speaker nor the multiple pickups corresponding to the any one of the acquisition results is abnormal.

[0017] In a possible implementation, after determining the abnormal detection results of at least one speaker and at least one pickup, the method further includes: when the abnormal detection results indicate that at least one speaker or at least one pickup is abnormal, reporting the abnormal detection results.

[0018] In a second aspect, an embodiment of the present application provides an abnormal detection device for a voice device, where the voice device includes at least one speaker and at least one pickup, and the device includes:

[0019] An acquisition unit, configured to acquire an audio test signal;

[0020] A processing unit, configured to control the at least one speaker to sequentially play the audio test signal, and when each speaker plays, control each pickup among the at least one pickup to respectively collect the currently played audio test signal to obtain an acquisition result corresponding to each speaker; wherein the acquisition result includes at least one sub-acquisition result, and the at least one sub-acquisition result corresponds one-to-one to the at least one pickup; any two speakers among the at least one speaker do not play the audio test signal at the same time; based on the acquisition results corresponding to the at least one speaker respectively, determine the abnormal detection results of the at least one speaker and the at least one pickup.

[0021] In a possible implementation, the acquisition unit acquires an audio test signal, specifically by: selecting multiple target audio signals from multiple pre-stored audio signals according to the frequency range supported by the at least one speaker; combining the multiple target audio signals in a randomly arranged manner to generate the audio test signal.

[0022] In a possible implementation, the processing unit combines the multiple target audio signals in a randomly arranged manner to generate the audio test signal, specifically by: randomly selecting a group from multiple pre-stored candidate playback parameters as test playback parameters; adjusting the gains of the multiple target audio signals according to the test playback parameters, and combining the adjusted multiple target audio signals in a randomly arranged manner to generate the audio test signal.

[0023] In a possible implementation, when the voice device includes a speaker and a pickup, the processing unit determines the abnormality detection results of the at least one speaker and the at least one pickup based on the acquisition results respectively corresponding to the at least one speaker, specifically: when the acquisition results indicate that the audio test signal is not acquired, it is determined that there is an abnormality in the speaker and / or the pickup; when the acquisition results indicate that the audio test signal is acquired, the acquired audio test signal included in the acquisition results is analyzed to obtain the acquisition signal parameters and the acquisition playback parameters; the acquisition signal parameters are matched with the test signal parameters of the audio test signal, and the acquisition playback parameters are matched with the test playback parameters of the audio test signal, and when the acquisition signal parameters or the acquisition playback parameters fail to match, it is determined that there is an abnormality in the speaker and / or the pickup; when both the acquisition signal parameters and the acquisition playback parameters match successfully, it is determined that there is no abnormality in both the speaker and the pickup.

[0024] In a possible implementation, when the voice device includes a speaker and multiple pickups, the processing unit determines the abnormality detection results of the at least one speaker and the at least one pickup based on the acquisition results respectively corresponding to the at least one speaker, specifically: when any one of the sub-acquisition results included in the acquisition results indicates that the audio test signal is not acquired, it is determined that there is an abnormality in the speaker and / or the multiple pickups; when multiple sub-acquisition results included in the acquisition results all indicate that the audio test signal is acquired, for the acquired audio test signals included in each of the multiple sub-acquisition results, the acquisition signal parameters and the acquisition playback parameters corresponding to each sub-acquisition result are respectively analyzed; the acquisition signal parameters corresponding to each of the multiple sub-acquisition results are respectively matched with the test signal parameters of the audio test signal, the acquisition playback parameters corresponding to each of the multiple sub-acquisition results are respectively matched with the test playback parameters of the audio test signal, and when any one of the acquisition signal parameters or any one of the acquisition playback parameters fails to match, it is determined that there is an abnormality in the speaker and / or the multiple pickups; when the acquisition signal parameters and the acquisition playback parameters corresponding to each of the multiple sub-acquisition results all match successfully, it is determined that there is no abnormality in both the speaker and the multiple pickups.

[0025] In a possible implementation, when any sub - acquisition result included in the acquisition result indicates that the audio test signal is not acquired, the processing unit determines that the speaker or the multiple pick - up microphones is / are abnormal. Specifically, when some sub - acquisition results included in the acquisition result indicate that the audio test signal is not acquired, the processing unit determines that some of the multiple pick - up microphones are abnormal; when multiple sub - acquisition results included in the acquisition result all indicate that the audio test signal is not acquired, the processing unit determines that the speaker or the multiple pick - up microphones is / are abnormal.

[0026] In a possible implementation, when the voice device includes multiple speakers and one pick - up microphone, the processing unit determines the abnormality detection results of the at least one speaker and the at least one pick - up microphone based on the acquisition results respectively corresponding to the at least one speaker. Specifically, for any one of the acquisition results respectively corresponding to the multiple speakers, the following operations are performed: when the any one of the acquisition results indicates that the audio test signal is not acquired, it is determined that the speaker corresponding to the any one of the acquisition results and / or the pick - up microphone is / are abnormal; when the any one of the acquisition results indicates that the audio test signal is acquired, the acquired audio test signal included in the any one of the acquisition results is analyzed to obtain acquisition signal parameters and acquisition playback parameters; the acquisition signal parameters are matched with the test signal parameters of the audio test signal, and the acquisition playback parameters are matched with the test playback parameters of the audio test signal, and when the acquisition signal parameters or the acquisition playback parameters fail to match, it is determined that the speaker corresponding to the any one of the acquisition results and / or the pick - up microphone is / are abnormal; when both the acquisition signal parameters and the acquisition playback parameters match successfully, it is determined that neither the speaker nor the pick - up microphone corresponding to the any one of the acquisition results is abnormal.

[0027] In a possible implementation, when the voice device includes a plurality of speakers and a plurality of pickups, the processing unit determines the abnormality detection results of the at least one speaker and the at least one pickup, specifically: for any one of the acquisition results corresponding to the plurality of speakers, perform the following operations: when any one of the sub-acquisition results included in the any one of the acquisition results indicates that the audio test signal is not acquired, determine that there is an abnormality in the speaker corresponding to the any one of the acquisition results and / or the plurality of pickups; when all of the plurality of sub-acquisition results included in the any one of the acquisition results indicate that the audio test signal is acquired, for the acquired audio test signals included in the plurality of sub-acquisition results respectively, parse to obtain the acquisition signal parameters and acquisition playback parameters corresponding to each sub-acquisition result; match the acquisition signal parameters corresponding to the plurality of sub-acquisition results with the test signal parameters of the audio test signal respectively, match the acquisition playback parameters corresponding to the plurality of sub-acquisition results with the test playback parameters of the audio test signal respectively, and when any one of the acquisition signal parameters or any one of the acquisition playback parameters fails to match, determine that there is an abnormality in the speaker corresponding to the any one of the acquisition results and / or the plurality of pickups; when the acquisition signal parameters and acquisition playback parameters corresponding to the plurality of sub-acquisition results all match successfully, determine that there is no abnormality in the speaker corresponding to the any one of the acquisition results and the plurality of pickups.

[0028] In a possible implementation, after the processing unit determines the abnormality detection results of the at least one speaker and the at least one pickup, it is further configured to: when the abnormality detection results indicate that there is an abnormality in the at least one speaker or the at least one pickup, report the abnormality detection results.

[0029] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0030] A memory for storing computer instructions;

[0031] A processor connected to the memory for executing the computer instructions in the memory and implementing the method according to any one of the first aspects when executing the computer instructions.

[0032] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, including:

[0033] The computer-readable storage medium stores computer instructions, and when the computer instructions run on a computer, the computer is caused to execute the method according to any one of the first aspects.

[0034] In a fifth aspect, an embodiment of the present application further provides a computer program product, including computer instructions, which implement the method described in any one of the first aspects when the computer instructions are executed by a processor.

[0035] The beneficial effects of the present application are as follows:

[0036] An embodiment of the present application provides an abnormal detection method, device, and electronic device for a voice device. The method includes: obtaining an audio test signal; controlling the at least one speaker to sequentially play the audio test signal, and when each speaker plays, controlling each of the at least one pickups to respectively collect the currently played audio test signal to obtain a collection result corresponding to each speaker; wherein the collection result includes at least one sub-collection result, and the at least one sub-collection result corresponds one-to-one to the at least one pickup; any two speakers among the at least one speaker do not play the audio test signal at the same time; determining an abnormal detection result of the at least one speaker and the at least one pickup based on the collection results respectively corresponding to the at least one speaker.

[0037] Based on the above solution, in the abnormal detection method for a voice device provided by the embodiment of the present application, mutual self-checking between the speaker and the pickup can be realized through the speaker and pickup of the device itself, which can avoid the situation of excessive noise when performing abnormal detection through additional external speakers and pickups in the related art, thereby improving the accuracy of abnormal detection. In addition, this method can also cover configuration scenarios of single speaker single pickup, multiple speakers single pickup, single speaker multiple pickups, and multiple speakers multiple pickups, so it can meet the self-checking needs of most voice devices and has a wider coverage compared to the self-checking solutions in the related art.

[0038] Other features and advantages of the present application will be described in the subsequent specification, and part of them will become obvious from the specification, or will be understood by implementing the present application. The objectives and other advantages of the present application can be achieved and obtained through the structures specifically pointed out in the written specification, claims, and drawings. Description of the Drawings

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the following drawings are only some embodiments of the present application.

[0040] Figure 1 It is a schematic diagram of the system architecture applied to the abnormal detection method for a voice device provided by the embodiment of the present application;

[0041] Figure 2An exemplary flowchart of an anomaly detection method for a voice device provided by an embodiment of the present application;

[0042] Figure 3 A schematic diagram of the anomaly detection result judgment process provided by an embodiment of the present application;

[0043] Figure 4 A schematic diagram of the anomaly detection result judgment process provided by an embodiment of the present application;

[0044] Figure 5 A schematic diagram of the anomaly detection result judgment process provided by an embodiment of the present application;

[0045] Figure 6 A schematic diagram of the anomaly detection result judgment process provided by an embodiment of the present application;

[0046] Figure 7 A schematic diagram of an anomaly detection device for a voice device provided by an embodiment of the present application;

[0047] Figure 8 A schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0048] To facilitate understanding of the technical solutions provided by the embodiments of the present application, the following introduces the professional terms involved in the embodiments of the present application.

[0049] (1) Speaker: A device such as a speaker that plays an audio signal.

[0050] (2) Pickup: A device such as a microphone that collects an audio signal.

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions of the present application with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are some, but not all, of the embodiments of the technical solutions of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments recorded in this application document without creative efforts shall fall within the scope of protection of the technical solutions of the present application.

[0052] The terms "first" and "second" in the embodiments of this application are used to distinguish different objects, rather than to describe a specific order. In addition, the term "including" and any variations thereof are intended to cover non-exclusive protection. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or devices. The "plurality" in this application may mean at least two, for example, it may be two, three, or more, and the embodiments of this application do not make restrictions.

[0053] In addition, the term "and / or" in this article is merely a relational description of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article, without special explanation, generally represents an "or" relationship between the associated objects before and after.

[0054] Refer to Figure 1 FIG. is a schematic diagram of the system architecture to which the abnormal detection method for a voice device provided by the embodiments of this application is applied. The system 100 may include: a signal preparer 110, at least one speaker 120, at least one pickup 130, and a judge 140.

[0055] Among them, the signal preparer 110 is used to obtain an audio test signal, and send the test signal parameters and test playback parameters corresponding to the audio test signal to the judge 140. At least one speaker 120 is used to sequentially play the audio test signal, and at least one pickup 130 is used to collect the currently played audio test signal when each speaker plays, and obtain the collection result corresponding to each speaker. Among them, the collection result may at least include at least one sub-collection result, and at least one sub-collection result corresponds one-to-one to at least one pickup 130. Any two speakers 120 among at least one speaker 120 do not play the audio test signal at the same time.

[0056] The judge 140 is used to determine the abnormal detection results of at least one speaker 120 and at least one pickup 130 based on the collection results corresponding to at least one speaker 120 respectively.

[0057] Optionally, the system may further include an alarm 150, which is used to report the abnormal detection result when the abnormal detection result indicates that at least one speaker 120 and / or at least one pickup 130 is abnormal. For example, a maintenance signal may be sent to the corresponding staff.

[0058] It should be noted that the above Figure 1The structure shown is only an example and is not limited to this embodiment of the present application.

[0059] See also Figure 2 An exemplary flow chart of an abnormality detection method for a voice device provided in an embodiment of the present application, which can be applied to Figure 1 The system shown may include the following processes:

[0060] S201, obtaining an audio test signal.

[0061] In a possible implementation, after the system starts the self-check function, the signal preparer of the system first obtains the audio test signal. In a specific implementation, the signal preparer may select multiple target audio signals from multiple pre-stored audio signals according to the frequency range supported by at least one speaker. Then, the multiple target audio signals are combined in a random arrangement to generate the audio test signal.

[0062] In some embodiments, the signal preparer combines multiple target audio signals in a random arrangement to generate an audio test signal. It can also randomly select one group from multiple pre-stored candidate playback parameters as a test playback parameter, and then adjust the gains of the multiple target audio signals according to the test playback parameters, and combine the adjusted multiple target audio signals in a random arrangement to generate an audio test signal.

[0063] After generating the audio test signal, the signal preparer may also send the test signal parameters and the test playback parameters corresponding to the audio test signal to the determiner for use in subsequent signal determination.

[0064] In some embodiments, the pre-stored multiple audio signals may be system-provided or user-uploaded. The multiple audio signals may be single-frequency audio signals of different frequencies, fixed decibels, and fixed durations covering the entire frequency band. The signal preparer may pre-set serial numbers corresponding to different frequencies. For example, the frequencies of the audio signals include: 1000Hz, 2000Hz, 3000Hz, 4000Hz, 5000Hz, 6000Hz, 7000Hz, 8000Hz, 9000Hz, and 10000Hz. The serial number corresponding to each frequency may be determined in the order of 0-9 from small to large. That is, 0 corresponds to 1000Hz, 1 corresponds to 2000Hz, and so on.

[0065] After the system starts the self-check function, the signal preparer can generate a sequence of random serial numbers according to the frequency range supported by at least one speaker, then select audio signals as target audio signals according to the frequencies corresponding to each serial number in the sequence, and then combine the target audio signals in the order of each serial number to generate an audio test signal. It should be understood that the number of serial numbers included in the random serial number sequence can be set according to the actual situation, for example, it can be 10, 5, etc., and the present application does not limit this.

[0066] For example, assuming that the generated random serial number is 8154609372, the target audio signals can be combined in the order of 9000Hz -> 2000Hz -> 6000Hz -> 5000Hz -> 7000Hz -> 1000Hz -> 10000Hz -> 4000Hz -> 8000Hz -> 3000Hz, with each signal duration of 10 seconds and a maximum amplitude of 8192, to generate a long signal composed of ten single-frequency signals with a total duration of 100 seconds and a maximum amplitude of 8192 as the audio test signal.

[0067] In another example, the signal preparer can also randomly select a group from multiple pre-stored candidate playback parameters as the test playback parameter. For example, the pre-stored candidate playback parameters can include: reducing by 20dB, reducing by 15dB, reducing by 10dB, reducing by 5dB, increasing by 5dB, increasing by 10dB, increasing by 15dB, etc., then the randomly selected test playback parameter can be reducing by 10dB. At this time, the gain of each of the above-selected target audio signals can be adjusted according to the test playback parameter, and the adjusted 10 target audio signals can be combined in the order of the random serial number sequence to generate an audio test signal.

[0068] Then, the signal preparer can send both the test signal parameters corresponding to the audio test signal and the selected test playback parameter to the judge. Among them, the test signal parameters can include signal basic information such as the frequencies of each target audio signal included in the audio test signal, the arrangement order of the target audio signals, the amplitudes of each target audio signal, and the durations of each target audio signal.

[0069] It should be noted that when the signal preparer pre-sets the serial numbers corresponding to different frequencies, the serial numbers can also be set in the order of 0-9 from largest to smallest according to the frequencies, or the serial numbers can also be set randomly, and the present application does not limit this.

[0070] Based on the above solution, since the multiple pre-stored audio signals are audio signals with different frequencies, covering the entire frequency band with a fixed decibel and a fixed duration, various scenarios can be covered, and the detection range is relatively wide. Moreover, in this application, the interference of other signals in the scenario can be prevented by randomly generating a sequence of serial numbers, and then making the target audio signal form a long signal according to the randomly generated sequence of serial numbers, thereby improving the accuracy of detection.

[0071] S202, control at least one speaker to play the audio test signal in sequence, and when each speaker plays, control each of the at least one pickups to respectively collect the currently played audio test signal, and obtain the collection result corresponding to each speaker.

[0072] Among them, the collection result includes at least one sub-collection result, and the at least one sub-collection result corresponds one-to-one to the at least one pickup; any two speakers among the at least one speaker do not play the audio test signal at the same time.

[0073] For example, assume that the voice device includes 2 speakers: speaker A1 and speaker A2, and 2 pickups: pickup B1 and pickup B2. Then when speaker A1 plays the audio test signal, speaker A2 does not play. Taking the playback order as speaker A1 -> speaker A2 as an example, when speaker A1 plays the audio test signal, pickup B1 and pickup B2 collect the audio test signal played by speaker A1 in real time. Pickup B1 can obtain the sub-collection result a11, and pickup B2 can obtain the sub-collection result a12. Then the sub-collection result a11 and the sub-collection result a12 can be combined into the collection result a1 corresponding to speaker A1. Similarly, when speaker A2 plays the audio test signal, pickup B1 and pickup B2 collect the audio test signal played by speaker A2 in real time. Pickup B1 can obtain the sub-collection result a21, and pickup B2 can obtain the sub-collection result a22. Then the sub-collection result a21 and the sub-collection result a22 can be combined into the collection result a2 corresponding to speaker a2.

[0074] In some embodiments, each pickup can send the obtained sub-collection result to the discriminator each time the sub-collection result is obtained, so that the discriminator analyzes and judges each sub-collection result according to the test signal parameters and test playback parameters of the test audio signal, and then determines the abnormal detection results of at least one speaker and at least one pickup. The specific analysis and judgment method will be elaborated in detail in S203.

[0075] S203, based on the collection results corresponding to each of the at least one speaker, determine the abnormal detection results of at least one speaker and at least one pickup.

[0076] Based on the above solution, in the abnormal detection method for voice devices provided in the embodiments of the present application, self-check between the speaker and the pickup can be achieved through the device's own speaker and pickup, thereby avoiding the situation of excessive noise when performing abnormal detection through additional external speakers and pickups in the related art, and further improving the accuracy of abnormal detection.

[0077] When the determiner determines the abnormal detection results of at least one speaker and at least one pickup based on the acquisition results corresponding to each of the at least one speaker, it can be divided into the following four cases:

[0078] Case 1: The voice device includes one speaker and one pickup.

[0079] See Figure 3 It is a schematic diagram of the abnormal detection result determination process provided in the embodiments of the present application. Assume that the one speaker and one pickup included in the voice device are: Speaker 1 and Pickup 1 respectively. After the self-check function is enabled, the signal preparer first generates an audio test signal, sends the test signal parameters and test playback parameters to the determiner, and then can control Speaker 1 to play the audio test signal, while controlling Pickup 1 to collect the audio test signal played by Speaker 1 in real time as the acquisition result corresponding to Speaker 1, and send the acquisition result to the determiner.

[0080] When the acquisition result indicates that no audio test signal is collected, since the failure to collect the audio test signal may be that Speaker 1 did not play the audio test signal, or that Speaker 1 played the audio test signal but Pickup 1 could not collect it, or that Speaker 1 did not play the audio test signal and at the same time Pickup 1 could not collect it either, the determiner can determine that the abnormal detection result is that Speaker 1 is abnormal, or Pickup 1 is abnormal, or both Speaker 1 and Pickup 1 are abnormal.

[0081] When the acquisition result indicates that the audio test signal is collected, the collected audio test signal included in the acquisition result can be analyzed to obtain the acquisition signal parameters and acquisition playback parameters.

[0082] Specifically, when analyzing the acquisition result, the signal magnitude of the collected audio test signal can be determined and used as the acquisition playback parameter. The signal frequency, the arrangement order of the signal frequencies, and the amplitude of the signal included in the collected audio test signal can also be determined and used as the acquisition signal parameters.

[0083] Then, match the collected signal parameters with the test signal parameters of the audio test signal, and match the collected playback parameters with the test playback parameters of the audio test signal. When the collected signal parameters or the collected playback parameters fail to match, determine that the abnormal detection result indicates an abnormality in the speaker or the pickup.

[0084] Specifically, when the pickup 1 collects the audio test signal, but the collected signal parameters do not match the test signal parameters of the audio test signal, that is, the frequency of the collected signal does not match the signal frequency of the audio test signal, and / or the amplitude of the collected signal does not match the signal amplitude of the audio test signal, it can be determined that the abnormal detection result indicates an abnormality in the pickup 1. When the pickup 1 collects the audio test signal, but the collected playback parameters do not match the test playback parameters of the audio test signal, that is, the size of the collected signal does not match the size of the audio test signal, it can be determined that the abnormal detection result indicates an abnormality in the pickup 1 and / or the speaker 1.

[0085] Conversely, when both the collected signal parameters and the collected playback parameters are successfully matched, it can be determined that the abnormal detection result indicates that neither the speaker nor the pickup is abnormal.

[0086] Exemplarily, the failure of the collected signal parameters to match may be that for one or more signals in the audio test signal, the frequency of the collected signal is different from the actual frequency of that signal. For example, if the frequency of a certain signal in the audio test signal is 1000 Hz, but the frequency of the collected signal of that signal is 2000 Hz, it can be determined that the collected signal parameters fail to match.

[0087] Exemplarily, the failure of the collected playback parameters to match may be that the size of the collected signal is different from the test playback parameters. For example, the size of the audio test signal before adjustment is 50 dB, the test playback parameter is to reduce by 10 dB, but the size of the collected audio test signal is 55 dB, which is not equal to 40 dB. At this time, it can be determined that the collected playback parameters fail to match.

[0088] When it is determined that the abnormal detection result indicates an abnormality in the speaker and / or the pickup, the judge can report the abnormal detection result. For example, it can be reported to the alarm, so that the alarm can notify the relevant staff to repair the voice device by sending a maintenance signal.

[0089] Optionally, when reporting the abnormal detection result, the reported information may include relevant information such as the cause of the abnormality and the device where the abnormality occurs. For example, the cause of the abnormality may be that the pickup 1 fails to collect the audio test signal, and the device where the abnormality occurs can be identified by an identifier. For example, the speaker is 1, the pickup is 2, and when the specific abnormal device cannot be determined, it can be identified by 3.

[0090] Case 2: The voice device includes a speaker and multiple pickups.

[0091] See Figure 4 It is a schematic diagram of the abnormal detection result judgment process provided by the embodiment of the present application. Assume that the one speaker and multiple pickups included in the voice device are respectively: Speaker 1 and Pickups 1 - Pickup S, where S > 1. After the self-check function is enabled, the signal preparer first generates an audio test signal, and sends the test signal parameters and test playback parameters to the judge, and then can control Speaker 1 to play the audio test signal, and at the same time control Pickups 1 - Pickup S to collect the audio test signal played by Speaker 1 in real time, obtain the respective sub-collection results 1 - sub-collection result S corresponding to Pickups 1 - Pickup S, combine the sub-collection results 1 - sub-collection result S into the collection result corresponding to Speaker 1, and send this collection result to the judge.

[0092] When any one of the sub-collection results included in the collection result indicates that the audio test signal is not collected, the judge can determine that one or more pickups do not collect the audio test signal, so it can be determined that the abnormal detection result is that there is an abnormality in Speaker 1 and / or Pickups 1 - Pickup S.

[0093] Specifically, when some of the sub-collection results included in the collection result indicate that the audio test signal is not collected, since some pickups can collect the audio test signal, it can be determined that the abnormal detection result is that some of Pickups 1 - Pickup S are abnormal, that is, the pickups corresponding to some of the sub-collection results are abnormal.

[0094] For example, assume that sub-collection result 2 and sub-collection result 3 indicate that the audio test signal is not collected, and other sub-collection results all indicate that the audio test signal is collected, and sub-collection result 2 is collected by Pickup 2, and sub-collection result 3 is collected by Pickup 3, then it can be determined that the abnormal detection result is that Pickup 2 and Pickup 3 are abnormal.

[0095] When all of the sub-collection results included in the collection result indicate that the audio test signal is not collected, since the non-collection of the audio test signal may be that Speaker 1 does not play the audio test signal, or that Speaker 1 plays the audio test signal but Pickups 1 - Pickup S are all unable to collect it, or that Speaker 1 does not play the audio test signal and at the same time Pickups 1 - Pickup S are all unable to collect it, so it can be determined that the abnormal detection result is that Speaker 1 is abnormal, or Pickups 1 - Pickup S are all abnormal, or Speaker 1 and Pickups 1 - Pickup S are all abnormal.

[0096] When multiple sub - acquisition results included in the acquisition result all indicate that an audio test signal has been acquired, the discriminator can respectively analyze the acquired audio test signals included in each of the multiple sub - acquisition results to obtain the acquisition signal parameters and acquisition playback parameters corresponding to each sub - acquisition result. Among them, the process of analyzing each sub - acquisition result can refer to the relevant description in Case 1 and will not be elaborated here.

[0097] Then, the acquisition signal parameters corresponding to each of the multiple sub - acquisition results are respectively matched with the test signal parameters of the audio test signal, and the acquisition playback parameters corresponding to each of the multiple sub - acquisition results are respectively matched with the test playback parameters of the audio test signal. When any one of the acquisition signal parameters or any one of the acquisition playback parameters fails to match, it is determined that there is an abnormality in the speaker and / or multiple pick - up microphones.

[0098] Specifically, when matching the acquisition signal parameters and the acquisition playback parameters respectively, if the acquisition playback parameters corresponding to each of the multiple sub - acquisition results are inconsistent, that is, the signal magnitudes collected by each pick - up microphone are inconsistent, it can be determined that the abnormal detection result is that there is an abnormality in some of the pick - up microphones. If the acquisition playback parameters corresponding to each of the multiple sub - acquisition results are consistent, but the acquisition playback parameters do not match the test playback parameters, it can be determined that the abnormal detection result is that there is an abnormality in Speaker 1, or that there is an abnormality in Pick - up Microphone 1 - Pick - up Microphone S, or that there is an abnormality in both Speaker 1 and Pick - up Microphone 1 - Pick - up Microphone S. Further, if the acquisition playback parameters corresponding to each of the multiple sub - acquisition results are consistent, and the acquisition playback parameters match the test playback parameters, but the acquisition signal parameters do not match the test signal parameters, the discriminator can determine that the abnormal detection result is that there is an abnormality in at least one of Pick - up Microphone 1 - Pick - up Microphone S.

[0099] For example, assume that the voice device includes Speaker 1, Pick - up Microphone 1, Pick - up Microphone 2, and Pick - up Microphone 3. The test playback parameter is to increase by 5 dB, the magnitude of the audio test signal before adjustment is 50 dB, and the random sequence number of the frequency is 8153. Among them, the frequency corresponding to each number takes the example in the embodiment given in S201.

[0100] If the signal magnitude collected by Pick - up Microphone 1 is 50 dB, the signal magnitude collected by Pick - up Microphone 2 is 45 dB, and the signal magnitude collected by Pick - up Microphone 3 is 45 dB, then since the collected signal magnitudes are inconsistent, it can be determined that there is an abnormality in at least one of Pick - up Microphone 1 - Pick - up Microphone 3.

[0101] If the signal magnitudes collected by pickup 1, pickup 2, and pickup 3 are all 50 dB, then, since the collected signal magnitudes are the same but the collection playback parameters do not match the test playback parameters, it can be determined that there may be an abnormality in speaker 1, or there may be abnormalities in pickups 1 - 3, or there may be abnormalities in speaker 1, pickup 1, pickup 2, and pickup 3.

[0102] If the signal magnitudes collected by each pickup are the same, and the collection playback parameters match the test playback parameters, but the frequency sequence numbers determined by pickup 1 are 8133, the frequency sequence numbers determined by pickup 2 are 8153, and the frequency sequence numbers determined by pickup 3 are 8153, then it can be determined that there is an abnormality in pickup 1 due to the mismatch between the collection signal parameters and the test signal parameters of pickup 1.

[0103] Conversely, when the collection signal parameters and the collection playback parameters corresponding to each of the multiple sub - collection results are successfully matched, the discriminator can determine that the abnormality detection result is that there are no abnormalities in the speaker and the multiple pickups.

[0104] Optionally, when the abnormality detection result is that any device has an abnormality, the abnormality detection result can be reported, and the abnormality detection process can be ended. In this case, after the alarm is reported to the relevant staff, the relevant staff can perform manual maintenance on all the speakers and all the pickups included in the voice device, thereby avoiding the situation where some devices have abnormalities but are not detected, and then improving the user experience.

[0105] Optionally, when reporting the abnormality detection result, if the reported information includes relevant information such as the cause of the abnormality and the devices with abnormalities, then in this case, after the alarm is reported to the relevant staff, the relevant staff can also perform manual maintenance on the devices involved in the reported information, thereby performing more precise maintenance and reducing the workload of the relevant personnel.

[0106] Case 3: The voice device includes multiple speakers and one pickup.

[0107] See Figure 5Schematic diagram of the abnormal detection result judgment process provided by the embodiment of the present application. Assume that a speaker and multiple pickups included in the voice device are respectively: Speaker 1 - Speaker T and Pickup 1, where T > 1. After the self - test function is enabled, the signal preparer first generates an audio test signal, and sends the test signal parameters and test playback parameters to the judge. Then, it can first control Speaker 1 to play the audio test signal, and at the same time control Pickup 1 to collect the audio test signal played by Speaker 1 in real - time, obtain the sub - collection result 1 corresponding to Pickup 1, and use the sub - collection result 1 as the collection result 1 corresponding to Speaker 1, and send the collection result 1 to the judge. Then, in sequence, according to the order of Speaker 2 - Speaker T, the above - mentioned process of collecting the audio test signal played by Speaker 1 is executed in turn, so as to obtain the collection results 2 - collection results T corresponding to Speaker 2 - Speaker T respectively, and send the collection results 2 - collection results T to the judge respectively.

[0108] In some embodiments, the judge can perform the following operations for each of the collection results 1 - collection results T:

[0109] When any one of the collection results indicates that the audio test signal is not collected, it is determined that the abnormal detection result is that the speaker and / or pickup corresponding to any one of the collection results is abnormal. For example, if any one of the collection results is collection result 1, when collection result 1 indicates that the audio test signal is not collected, it can be determined that Speaker 1 and / or Pickup 1 is abnormal.

[0110] When any one of the collection results indicates that the audio test signal is collected, the collected audio test signal included in any one of the collection results is analyzed to obtain the collection signal parameters and collection playback parameters. Then, the judge matches the collection signal parameters with the test signal parameters of the audio test signal, and matches the collection playback parameters with the test playback parameters of the audio test signal, and when the collection signal parameters or the collection playback parameters fail to match, it is determined that the abnormal detection result is that the speaker and / or Pickup 1 corresponding to any one of the collection results is abnormal.

[0111] On the contrary, when both the collection signal parameters and the collection playback parameters match successfully, the judge can determine that the abnormal detection result is that the speaker and pickup corresponding to any one of the collection results are not abnormal.

[0112] Among them, the specific matching process of each collection result is the same as that in Case 1, so the relevant description in Case 1 can be referred to and will not be elaborated here.

[0113] Optionally, when the detector determines that the abnormality detection result is that there is an abnormality in the speaker and / or the pickup for any one of the acquisition results, it can report the abnormality detection result before performing the above analysis and judgment operation on the next acquisition result, so that the relevant staff can manually repair the speaker and the pickup corresponding to any one of the acquisition results. In this case, since the pickup has been repaired, there is no abnormality, so it can be ensured that when performing the above analysis and judgment operation on the subsequent acquisition results, only the judgment of whether there is an abnormality in the speaker is required to complete the abnormality detection, thereby reducing the calculation amount of the detector and improving the judgment efficiency.

[0114] For example, assume that for acquisition result 1, the abnormality detection result is that there is an abnormality in the speaker and / or the pickup. Then the abnormality detection result can be reported so that the relevant staff can manually repair the speaker and the pickup 1 corresponding to any one of the acquisition results. Then when analyzing and judging acquisition results 2 - acquisition result T, it can be defaulted that all pickups 1 do not have abnormalities, that is, it is only necessary to judge whether signals are collected for acquisition results 2 - acquisition result T and whether the acquisition and playback parameters match the test playback parameters.

[0115] It should be noted that Figure 5 The situation shown only is the situation where the pickup 1 has been repaired before analyzing and judging the acquisition result corresponding to the speaker T. If the pickup 1 has not been repaired before analyzing and judging the acquisition result corresponding to the speaker T, it is processed in the same way as analyzing and judging the acquisition result corresponding to the speaker 1.

[0116] In addition, the timing of each speaker playing the audio test signal can be set according to the actual situation. For example, it can be when the detector finishes analyzing and judging the audio test signal played by the previous speaker, or it can also be to immediately control the next speaker to play the audio test signal after the previous speaker finishes playing the audio test signal. This application does not make any limitations on this.

[0117] Case 4: The voice device includes multiple speakers and multiple pickups.

[0118] See Figure 6Schematic diagram of the abnormal detection result judgment process provided by the embodiment of the present application. Assume that the multiple speakers and multiple pickups included in the voice device are respectively: Speaker 1 - Speaker T and Pickup 1 - Pickup S, where both S and T are greater than 1. In this case, for each speaker, the playback and acquisition of the audio test signal are performed in the manner of Case 2 respectively. That is, after the self - test function is enabled, the signal preparer first generates an audio test signal, and sends the test signal parameters and test playback parameters to the judgment device. Then, it can first control Speaker 1 to play the audio test signal, and at the same time control Pickup 1 - Pickup S to collect the audio test signal played by Speaker 1 in real time, obtaining the respective sub - acquisition results 1 - sub - acquisition results S corresponding to Pickup 1 - Pickup S, combining the sub - acquisition results 1 - sub - acquisition results S into the acquisition result 1 corresponding to Speaker 1, and sending the acquisition result 1 to the judgment device. Then, in sequence, according to the order of Speaker 2 - Speaker T, the above - mentioned process of collecting the audio test signal played by Speaker 1 is executed in turn, so as to obtain the respective acquisition results 2 - acquisition results T corresponding to Speaker 2 - Speaker T, and send the acquisition results 2 - acquisition results T to the judgment device respectively. Each acquisition result includes sub - acquisition results 1 - sub - acquisition results S.

[0119] For each of the acquisition results 1 - acquisition results T, the following operations are performed respectively:

[0120] When any one of the sub - acquisition results included in any one of the acquisition results indicates that the audio test signal is not collected, it is determined that there is an abnormality in the speaker and / or multiple pickups corresponding to any one of the acquisition results.

[0121] When all of the multiple sub - acquisition results included in any one of the acquisition results indicate that the audio test signal is collected, for the collected audio test signals included in each of the multiple sub - acquisition results, the acquisition signal parameters and acquisition playback parameters corresponding to each sub - acquisition result are respectively analyzed;

[0122] The acquisition signal parameters corresponding to each of the multiple sub - acquisition results are respectively matched with the test signal parameters of the audio test signal, and the acquisition playback parameters corresponding to each of the multiple sub - acquisition results are respectively matched with the test playback parameters of the audio test signal. When any one of the acquisition signal parameters or any one of the acquisition playback parameters fails to match, it is determined that there is an abnormality in the speaker and / or multiple pickups corresponding to any one of the acquisition results.

[0123] On the contrary, when the acquisition signal parameters and acquisition playback parameters corresponding to each of the multiple sub - acquisition results are all successfully matched, it is determined that there is no abnormality in the speaker and multiple pickups corresponding to any one of the acquisition results.

[0124] Among them, the specific matching process of each acquisition result is the same as that in Case 2. Therefore, the relevant descriptions in Case 2 can be referred to and will not be elaborated here.

[0125] Optionally, when the detector determines that there is an abnormality in the speaker and / or the pickup for any one acquisition result, it can report the abnormality detection result before performing the above analysis and judgment operation on the next acquisition result, so that the relevant staff can manually repair the speaker and pickups 1 - pickup S corresponding to any one acquisition result. In this case, since pickups 1 - pickup S have all been repaired, there is no abnormality, so it can be ensured that when performing the above analysis and judgment operation on the subsequent acquisition results, only the abnormality of the speaker corresponding to the acquisition result needs to be judged to complete the abnormality detection, thereby reducing the calculation amount of the detector and improving the judgment efficiency.

[0126] Figure 6 The situation shown only is the situation where pickups 1 - pickup S have been repaired before analyzing and judging the acquisition result corresponding to speaker T. If pickups 1 - pickup S have not been repaired before analyzing and judging the acquisition result corresponding to speaker T, it will be processed in the same way as analyzing and judging the acquisition result corresponding to speaker 1.

[0127] Based on the judgments of the above four cases, the configuration scenarios of single speaker and single pickup, multiple speakers and single pickup, single speaker and multiple pickups, and multiple speakers and multiple pickups can be covered. Therefore, the self - inspection requirements of most voice devices can be met, and the coverage is relatively wide compared with the self - inspection schemes in the related technologies.

[0128] Based on the same concept of the above method, refer to Figure 7 , an abnormality detection device 700 for a voice device provided in an embodiment of the present application. The device 700 can execute each step in the above method. To avoid repetition, it will not be elaborated here. The device 700 includes an acquisition unit 701 and a processing unit 702. In one scenario:

[0129] The acquisition unit 701 is used to acquire an audio test signal;

[0130] A processing unit 702 is configured to control the at least one speaker to sequentially play the audio test signal, and when each speaker is playing, control each of the at least one pickups to respectively collect the currently played audio test signal to obtain a collection result corresponding to each speaker; wherein, the collection result includes at least one sub-collection result, and the at least one sub-collection results correspond one-to-one to the at least one pickups; any two speakers among the at least one speakers do not play the audio test signal at the same time; based on the collection results respectively corresponding to the at least one speakers, determine the abnormality detection results of the at least one speakers and the at least one pickups.

[0131] In a possible implementation manner, the obtaining unit 701 obtains the audio test signal, and specifically is configured to: select multiple target audio signals from multiple pre-stored audio signals according to the frequency range supported by the at least one speaker; combine the multiple target audio signals in a randomly arranged manner to generate the audio test signal.

[0132] In a possible implementation manner, the processing unit 702 combines the multiple target audio signals in a randomly arranged manner to generate the audio test signal, and specifically is configured to: randomly select a group from multiple pre-stored candidate playback parameters as the test playback parameters; adjust the gains of the multiple target audio signals according to the test playback parameters, and combine the adjusted multiple target audio signals in a randomly arranged manner to generate the audio test signal.

[0133] In a possible implementation manner, when the voice device includes one speaker and one pickup, the processing unit 702 determines the abnormality detection results of the at least one speaker and the at least one pickup based on the collection results respectively corresponding to the at least one speaker, and specifically is configured to: when the collection result indicates that the audio test signal is not collected, determine that the speaker and / or the pickup is abnormal; when the collection result indicates that the audio test signal is collected, analyze the collected audio test signal included in the collection result to obtain the collection signal parameters and the collection playback parameters; match the collection signal parameters with the test signal parameters of the audio test signal, and match the collection playback parameters with the test playback parameters of the audio test signal, and when the collection signal parameters or the collection playback parameters fail to match, determine that the speaker and / or the pickup is abnormal; when both the collection signal parameters and the collection playback parameters match successfully, determine that both the speaker and the pickup are not abnormal.

[0134] In a possible implementation, when the voice device includes a speaker and multiple pickups, the processing unit 702 determines the anomaly detection results of the at least one speaker and the at least one pickup based on the acquisition results respectively corresponding to the at least one speaker, specifically: when any one of the sub-acquisition results included in the acquisition results indicates that the audio test signal is not acquired, it is determined that there is an anomaly in the speaker and / or the multiple pickups; when multiple sub-acquisition results included in the acquisition results all indicate that the audio test signal is acquired, for the acquired audio test signals respectively included in the multiple sub-acquisition results, the acquisition signal parameters and acquisition playback parameters corresponding to each sub-acquisition result are respectively obtained through parsing; the acquisition signal parameters respectively corresponding to the multiple sub-acquisition results are respectively matched with the test signal parameters of the audio test signal, the acquisition playback parameters respectively corresponding to the multiple sub-acquisition results are respectively matched with the test playback parameters of the audio test signal, and when any one of the acquisition signal parameters or any one of the acquisition playback parameters fails to match, it is determined that there is an anomaly in the speaker and / or the multiple pickups; when the acquisition signal parameters and acquisition playback parameters respectively corresponding to the multiple sub-acquisition results all match successfully, it is determined that there is no anomaly in the speaker and the multiple pickups.

[0135] In a possible implementation, when any one of the sub-acquisition results included in the acquisition results indicates that the audio test signal is not acquired, the processing unit 702 determines that there is an anomaly in the speaker or the multiple pickups, specifically: when some of the sub-acquisition results included in the acquisition results indicate that the audio test signal is not acquired, it is determined that some of the multiple pickups are abnormal; when multiple sub-acquisition results included in the acquisition results all indicate that the audio test signal is not acquired, it is determined that there is an anomaly in the speaker or the multiple pickups.

[0136] In a possible implementation, when the voice device includes a plurality of speakers and a microphone, the processing unit 702 determines the abnormality detection results of the at least one speaker and the at least one microphone based on the acquisition results respectively corresponding to the at least one speaker, and specifically: for any one of the acquisition results respectively corresponding to the plurality of speakers, perform the following operations: when any one of the acquisition results indicates that the audio test signal is not acquired, determine that there is an abnormality in the speaker corresponding to any one of the acquisition results and / or the microphone; when any one of the acquisition results indicates that the audio test signal is acquired, analyze the acquired audio test signal included in any one of the acquisition results to obtain the acquisition signal parameters and the acquisition playback parameters; match the acquisition signal parameters with the test signal parameters of the audio test signal, and match the acquisition playback parameters with the test playback parameters of the audio test signal, and when the acquisition signal parameters or the acquisition playback parameters fail to match, determine that there is an abnormality in the speaker corresponding to any one of the acquisition results and / or the microphone; when both the acquisition signal parameters and the acquisition playback parameters match successfully, determine that neither the speaker corresponding to any one of the acquisition results nor the microphone has an abnormality.

[0137] In a possible implementation, when the voice device includes a plurality of speakers and a plurality of microphones, the processing unit 702 determines the abnormality detection results of the at least one speaker and the at least one microphone based on the acquisition results respectively corresponding to the at least one speaker, and specifically: for any one of the acquisition results respectively corresponding to the plurality of speakers, perform the following operations: when any one of the sub-acquisition results included in any one of the acquisition results indicates that the audio test signal is not acquired, determine that there is an abnormality in the speaker corresponding to any one of the acquisition results and / or the plurality of microphones; when all of the multiple sub-acquisition results included in any one of the acquisition results indicate that the audio test signal is acquired, for the acquired audio test signals respectively included in the multiple sub-acquisition results, analyze them respectively to obtain the acquisition signal parameters and the acquisition playback parameters corresponding to each sub-acquisition result; match the acquisition signal parameters respectively corresponding to the multiple sub-acquisition results with the test signal parameters of the audio test signal, match the acquisition playback parameters respectively corresponding to the multiple sub-acquisition results with the test playback parameters of the audio test signal, and when any one of the acquisition signal parameters or any one of the acquisition playback parameters fails to match, determine that there is an abnormality in the speaker corresponding to any one of the acquisition results and / or the plurality of microphones; when the acquisition signal parameters and the acquisition playback parameters respectively corresponding to the multiple sub-acquisition results all match successfully, determine that neither the speaker corresponding to any one of the acquisition results nor the plurality of microphones has an abnormality.

[0138] In a possible implementation, after the processing unit 702 determines the anomaly detection results of at least one speaker and at least one pickup, it is further configured to: when the anomaly detection results indicate that there is an anomaly in at least one of the at least one speaker or at least one pickup, report the anomaly detection results.

[0139] Based on the same concept of the above method, refer to Figure 8 , which is a schematic structural diagram of an electronic device provided in an embodiment of the present application. The electronic device includes at least one processor 802, and a memory 801 connected or coupled to the at least one processor 802. In addition, the electronic device may further include a communication interface 803. The electronic device can interact with other devices through the communication interface 803.

[0140] Exemplarily, the communication interface 803 may be a transceiver, a circuit, a bus, a module, a pin, or other types of communication interfaces. When the electronic device is a chip-like device or circuit, the communication interface 803 in the electronic device may also be an input / output circuit, which can input information (or receive information) and output information (or send information). The processor is an integrated processor, a microprocessor, an integrated circuit, or a logic circuit. The processor can determine the output information based on the input information.

[0141] The coupling in the embodiments of the present application is an indirect coupling or communication connection between devices, units, or modules, which can be electrical, mechanical, or other forms, and is used for information interaction between devices, units, or modules. The processor 802 may cooperate with the memory 801 and the communication interface 803. In the present application, the specific connection medium between the above-mentioned processor 802, memory 801, and communication interface 803 is not limited.

[0142] Optionally, refer to Figure 8 , the processor 802, the memory 801, and the communication interface 803 are interconnected with each other through a bus. The bus may be a peripheral component interconnect (PCI) bus, an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, Figure 8 only a thick line is used to represent it in

[0143] In the embodiments of the present application, the memory 801, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. The memory 801 may include at least one type of storage medium. For example, it may include flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic memory, magnetic disk, optical disc, and so on. The memory 801 is any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 801 in the embodiments of the present application can also be a circuit or any other device capable of implementing a storage function, for storing instructions, computer programs, and / or data.

[0144] In the embodiments of the present application, the processor 802 can be a general-purpose processor, such as a central processing unit (CPU), a digital signal processor, an application-specific integrated circuit, a field-programmable gate array, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the abnormal detection method for voice devices disclosed in combination with the embodiments of the present application can be directly embodied as being executed by a hardware processor, or can be executed by a combination of hardware and software modules in the processor.

[0145] By designing and programming the processor 802, the code corresponding to the abnormal detection method for voice devices introduced in the foregoing embodiments can be solidified into the chip, so that the chip can execute the steps of the foregoing abnormal detection method for voice devices when running. How to design and program the processor 802 is a well-known technology to those skilled in the art and will not be elaborated here.

[0146] In one or more embodiments, the memory 801 stores instructions executable by at least one processor 802. The at least one processor 802 can implement the steps of any of the above methods by invoking the instructions or computer programs stored in the memory 801.

[0147] The embodiments of the present application also provide a computer-readable storage medium, on which computer instructions are stored. When the computer instructions run on a computer, the computer is caused to execute the steps of any of the above methods.

[0148] Based on the same inventive concept, the embodiments of the present application also provide a computer program product, which includes: computer program code. When the computer program code runs on a computer, the computer is caused to execute the anomaly detection method for a voice device as described in any of the foregoing discussions. Since the principle of solving problems by the above computer program product is similar to that of the anomaly detection method for a voice device, the implementation of the above computer program product can refer to the implementation of the method, and the repeated parts will not be described again.

[0149] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: various media such as ROM, RAM, magnetic disk, or optical disc that can store program code.

[0150] Although the specific implementation manners of the present application have been described above, those skilled in the art should understand that these are only examples, and the protection scope of the present application is defined by the appended claims. Without departing from the principle and essence of the present application, those skilled in the art can make various changes or modifications to these implementation manners, but these changes and modifications all fall within the protection scope of the present application. Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concept. Therefore, the appended claims are intended to be interpreted to include the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0151] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. An abnormal detection method for a voice device, characterized in that, The voice device includes at least one speaker and at least one pickup, and the method includes: Obtain an audio test signal; Control the at least one speaker to sequentially play the audio test signal, and when each speaker plays, control each of the at least one pickups to respectively collect the currently played audio test signal to obtain a collection result corresponding to each speaker; wherein, the collection result includes at least one sub-collection result, and the at least one sub-collection result corresponds one-to-one with the at least one pickup; any two speakers among the at least one speaker do not play the audio test signal at the same time; Based on the collection results corresponding to the at least one speaker respectively, determine the anomaly detection results of the at least one speaker and the at least one pickup.

2. The method according to claim 1, wherein The obtaining of the audio test signal includes: Select multiple target audio signals from multiple pre-stored audio signals according to the frequency range supported by the at least one speaker; Combine the multiple target audio signals in a randomly arranged manner to generate the audio test signal.

3. The method according to claim 2, characterized in that, The combining the multiple target audio signals in a randomly arranged manner to generate the audio test signal includes: Randomly select a group from multiple pre-stored candidate playback parameters as the test playback parameters; Adjust the gains of the multiple target audio signals according to the test playback parameters, and combine the adjusted multiple target audio signals in a randomly arranged manner to generate the audio test signal.

4. The method according to any one of claims 1-3, characterized in that, When the voice device includes one speaker and one pickup, the determining the anomaly detection results of the at least one speaker and the at least one pickup based on the collection results corresponding to the at least one speaker respectively includes: When the collection result indicates that the audio test signal is not collected, determine that there is an anomaly in the speaker and / or the pickup; When the collection result indicates that the audio test signal is collected, parse the collected audio test signal included in the collection result to obtain collection signal parameters and collection playback parameters; Match the collection signal parameters with the test signal parameters of the audio test signal, and match the collection playback parameters with the test playback parameters of the audio test signal, and when the collection signal parameters or the collection playback parameters fail to match, determine that there is an anomaly in the speaker and / or the pickup; When both the collection signal parameters and the collection playback parameters match successfully, determine that there is no anomaly in both the speaker and the pickup.

5. The method according to any one of claims 1 to 3, characterized in that When the voice device includes one speaker and multiple pickups, the determining the anomaly detection results of the at least one speaker and the at least one pickup based on the collection results corresponding to the at least one speaker respectively includes: When any one of the sub-collection results included in the collection result indicates that the audio test signal is not collected, determine that there is an anomaly in the speaker and / or the multiple pickups; When multiple sub - acquisition results included in the acquisition result all indicate that the audio test signal has been acquired, for the acquired audio test signals included in each of the multiple sub - acquisition results, parsing is respectively performed to obtain the acquisition signal parameters and acquisition playback parameters corresponding to each sub - acquisition result; The acquisition signal parameters corresponding to each of the multiple sub - acquisition results are respectively matched with the test signal parameters of the audio test signal, and the acquisition playback parameters corresponding to each of the multiple sub - acquisition results are respectively matched with the test playback parameters of the audio test signal. When any one of the acquisition signal parameters or any one of the acquisition playback parameters fails to match, it is determined that the loudspeaker and / or the multiple pickups are abnormal; When the acquisition signal parameters and acquisition playback parameters corresponding to each of the multiple sub - acquisition results all match successfully, it is determined that neither the loudspeaker nor the multiple pickups are abnormal.

6. The method according to claim 5, wherein When any one of the sub - acquisition results included in the acquisition result indicates that the audio test signal has not been acquired, determining that the loudspeaker or the multiple pickups are abnormal includes: When some of the sub - acquisition results included in the acquisition result indicate that the audio test signal has not been acquired, it is determined that some of the multiple pickups are abnormal; When multiple sub - acquisition results included in the acquisition result all indicate that the audio test signal has not been acquired, it is determined that the loudspeaker or the multiple pickups are abnormal.

7. The method according to any one of claims 1-3, characterized in that, When the voice device includes multiple loudspeakers and one pickup, the method for determining the abnormality detection result of the at least one loudspeaker and the at least one pickup based on the acquisition result corresponding to each of the at least one loudspeaker includes: For any one of the acquisition results corresponding to each of the multiple loudspeakers, the following operations are performed: When the any one of the acquisition results indicates that the audio test signal has not been acquired, it is determined that the loudspeaker corresponding to the any one of the acquisition results and / or the pickup are abnormal; When the any one of the acquisition results indicates that the audio test signal has been acquired, for the acquired audio test signal included in the any one of the acquisition results, parsing is performed to obtain the acquisition signal parameters and acquisition playback parameters; The acquisition signal parameters are matched with the test signal parameters of the audio test signal, and the acquisition playback parameters are matched with the test playback parameters of the audio test signal. When the acquisition signal parameters or the acquisition playback parameters fail to match, it is determined that the loudspeaker corresponding to the any one of the acquisition results and / or the pickup are abnormal; When the acquisition signal parameters and the acquisition playback parameters both match successfully, it is determined that neither the loudspeaker nor the pickup corresponding to the any one of the acquisition results are abnormal.

8. The method according to any one of claims 1 to 3, characterized in that, When the voice device includes multiple loudspeakers and multiple pickups, the method for determining the abnormality detection result of the at least one loudspeaker and the at least one pickup based on the acquisition result corresponding to each of the at least one loudspeaker includes: For any one of the acquisition results corresponding to each of the multiple loudspeakers, the following operations are performed: When any one of the sub - acquisition results included in any one of the acquisition results indicates that the audio test signal has not been acquired, it is determined that there is an abnormality in the speaker and / or the plurality of pick - up microphones corresponding to the any one of the acquisition results; When the plurality of sub - acquisition results included in any one of the acquisition results all indicate that the audio test signal has been acquired, for the acquired audio test signals included in each of the plurality of sub - acquisition results, the acquisition signal parameters and acquisition playback parameters corresponding to each sub - acquisition result are respectively analyzed; The acquisition signal parameters corresponding to each of the plurality of sub - acquisition results are respectively matched with the test signal parameters of the audio test signal, and the acquisition playback parameters corresponding to each of the plurality of sub - acquisition results are respectively matched with the test playback parameters of the audio test signal. When any one of the acquisition signal parameters or any one of the acquisition playback parameters fails to match, it is determined that there is an abnormality in the speaker and / or the plurality of pick - up microphones corresponding to the any one of the acquisition results; When the acquisition signal parameters and acquisition playback parameters corresponding to each of the plurality of sub - acquisition results are all successfully matched, it is determined that there is no abnormality in the speaker and the plurality of pick - up microphones corresponding to any one of the acquisition results.

9. An abnormal detection device for a voice device, characterized in that, The voice device includes at least one speaker and at least one pick - up microphone, and the apparatus includes: An acquisition unit, configured to acquire an audio test signal; A processing unit, configured to control the at least one speaker to sequentially play the audio test signal, and when each speaker plays, control each of the at least one pick - up microphones to respectively acquire the currently played audio test signal to obtain the acquisition result corresponding to each speaker; wherein, the acquisition result includes at least one sub - acquisition result, and the at least one sub - acquisition result corresponds one - to - one with the at least one pick - up microphone; any two speakers among the at least one speaker do not play the audio test signal at the same time; based on the acquisition results corresponding to each of the at least one speaker, determine the abnormality detection result of the at least one speaker and the at least one pick - up microphone.

10. An electronic device, characterized in that, Comprising: A memory, configured to store computer instructions; A processor, connected to the memory, configured to execute the computer instructions in the memory, and when executing the computer instructions, implement the method according to any one of claims 1 to 8.

Images