Industrial microphone

Abstract

The invention discloses an industrial microphone, and relates to the technical field of audio acquisition and processing, and the industrial microphone can comprise a hardware layer, a resource virtualization layer and a software virtualization layer. The hardware layer comprises a microphone array, a processing unit, a storage unit, a communication unit and a power management unit; the resource virtualization layer interacts with hardware resources of the hardware layer; the software virtualization layer executes the following steps under the cooperation of the resource virtualization layer and the support of the hardware layer: collecting audio data; and carrying out noise reduction processing and / or abnormity identification processing on the collected audio data, and sending an audio processing result to the server-side equipment. The industrial microphone disclosed by the invention is a single device and can realize multiple functions of audio acquisition, noise reduction, abnormity identification, pushing and the like, and compared with an existing industrial microphone (a system consisting of a plurality of devices), the industrial microphone disclosed by the invention is relatively low in cost and relatively easy to maintain.

Description

Technical Field

[0001] This application relates to the technical field of audio data acquisition and processing, and particularly to an industrial microphone. Background Art

[0002] An industrial microphone is a device designed specifically for industrial environments, which can work stably under complex and harsh conditions. It is mainly used for collecting and analyzing audio data to support scenarios such as industrial automation, monitoring, control, and maintenance.

[0003] The current industrial microphone is a large system composed of an industrial sound pickup device, an audio processing device, an edge gateway device, and a background management system. Although this system can achieve the acquisition and analysis of audio data, due to its being a large system composed of multiple devices, the cost is huge and the maintenance is difficult. Summary of the Invention

[0004] In view of this, this application provides an industrial microphone to solve the problems of huge cost and difficult maintenance of the existing industrial microphone (a large system composed of multiple devices). The technical solution is as follows:

[0005] This application provides an industrial microphone, including: a hardware layer, a resource virtualization layer, and a software virtualization layer;

[0006] The hardware layer includes: a microphone array for collecting audio data, a processing unit for processing audio data, a storage unit for storing audio data, a communication unit for communicating with other devices, and a power management unit for providing power voltage for each unit;

[0007] The resource virtualization layer is used to interact with the hardware resources of the hardware layer;

[0008] The software virtualization layer is used to execute, with the cooperation of the resource virtualization layer and the support of the hardware layer:

[0009] Collect audio data;

[0010] Perform noise reduction processing and / or anomaly recognition processing on the collected audio data to obtain an audio processing result;

[0011] Send the audio processing result to the server device.

[0012] In a possible implementation, the industrial microphone further includes: a user interaction layer;

[0013] The user interaction layer is used to obtain the configuration information of the industrial microphone configured by the user in the microphone configuration interface. Among them, the configuration of the industrial microphone by the user includes some or all of the following configurations: pickup configuration, network configuration, transmission configuration, audio acquisition configuration, system configuration, device upgrade configuration;

[0014] The software virtualization layer is further used to process the configuration information.

[0015] In a possible implementation manner, the hardware layer further includes:

[0016] An audio processing unit for connecting an audio playback device.

[0017] In a possible implementation manner, when the software virtualization layer performs noise reduction processing and / or anomaly recognition processing on the collected audio data with the cooperation of the resource virtualization layer and the support of the hardware layer, it is specifically used for:

[0018] Perform noise reduction processing on the collected audio data to obtain noise-reduced audio data;

[0019] Obtain the acoustic features of the noise-reduced audio data;

[0020] Use the pre-trained anomaly recognition model to extract features from the acoustic features;

[0021] Use the anomaly recognition model to reconstruct the acoustic features based on the extracted features;

[0022] Based on the reconstructed acoustic features and the acoustic features of the noise-reduced audio data, determine whether the noise-reduced audio data is abnormal.

[0023] In a possible implementation manner, the audio processing result includes the noise-reduced audio data obtained by performing noise reduction processing on the collected audio data, and the anomaly recognition result obtained by performing anomaly recognition on the noise-reduced audio data;

[0024] When the software virtualization layer sends the audio processing result to the server device with the cooperation of the resource virtualization layer and the support of the hardware layer, it is specifically used for:

[0025] According to the State Grid standard protocol, encapsulate the noise-reduced audio data and the anomaly recognition result to obtain the first encapsulated data, and send the first encapsulated data to the server device;

[0026] and / or,

[0027] Encode the denoised audio data, encapsulate the encoded data and the abnormal recognition result according to the Open Network Video Interface Forum (ONVIF) protocol or the Video Surveillance Networking Protocol GB28181 to obtain the second encapsulated data, and send the second encapsulated data to the server device.

[0028] In a possible implementation manner, when the software virtualization layer sends the audio processing result to the server device with the cooperation of the resource virtualization layer and the support of the hardware layer, it is specifically used for:

[0029] When receiving the data sending start instruction sent by the server device, send the audio processing result to the server device;

[0030] The software virtualization layer is further used to execute, with the cooperation of the resource virtualization layer and the support of the hardware layer:

[0031] When receiving the data sending stop instruction sent by the server device, stop sending data to the server device.

[0032] In a possible implementation manner, the software virtualization layer sends the audio processing result to the server device through a Transmission Control Protocol (TCP) connection established with the server device;

[0033] The software virtualization layer receives the data sending start instruction and the data sending stop instruction sent by the server device through a Message Queuing Telemetry Transport (MQTT) connection established with the server device.

[0034] In a possible implementation manner, the microphone array is a circular microphone array.

[0035] In a possible implementation manner, the microphones in the microphone array are digital silicon microphones.

[0036] In a possible implementation manner, the communication unit includes a wired communication unit and a wireless communication unit;

[0037] The wired communication unit includes a Switch chip, which can be connected to a router device or a cluster gateway to implement data forwarding;

[0038] The wireless communication unit includes: a Bluetooth communication module and / or a WIFI communication module.

[0039] With the above technical solution, the industrial microphone provided by the present application includes a hardware layer, a resource virtualization layer, and a software virtualization layer. The hardware layer includes a microphone array, a processing unit, a storage unit, a communication unit, and a power management unit. The resource virtualization layer can interact with the hardware resources of the hardware layer, and the software virtualization layer can, with the cooperation of the resource virtualization layer and the support of the hardware layer, achieve audio data acquisition, noise reduction, anomaly recognition, and network transmission. The industrial microphone provided by the present application is a device integrating many functions such as audio acquisition, audio noise reduction, audio anomaly recognition, and audio push. Since the industrial microphone provided by the present application is a single device rather than a system composed of multiple devices, the cost and maintenance difficulty are greatly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings according to the provided drawings without creative efforts.

[0041] Figure 1 Schematic diagram of an existing industrial microphone;

[0042] Figure 2 Schematic structural diagram of the industrial microphone provided by an embodiment of the present application;

[0043] Figure 3 Specific structural diagram of the industrial microphone provided by an embodiment of the present application;

[0044] Figure 4 Schematic flow diagram of anomaly recognition processing for the denoised audio data provided by an embodiment of the present application;

[0045] Figure 5 Another specific structural diagram of the industrial microphone provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the embodiments part of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0047] The following describes the embodiments of the present application in conjunction with the drawings. Those of ordinary skill in the art know that with the development of technology and the emergence of new scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0048] In the description, claims and the above-mentioned drawings of this application, terms such as "first", "second", etc. are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that such terms can be interchanged under appropriate circumstances, which is only a way of distinguishing objects with the same attributes when describing embodiments of this application. In addition, the terms "comprising", "having" and any variations thereof are intended to cover non-exclusive inclusion, so that a process, method, system, product or device including a series of units does not have to be limited to those units, but may include other units not clearly listed or inherent to these processes, methods, products or devices.

[0049] Since existing industrial microphones are large systems composed of multiple devices (as shown in Figure 1 ), therefore, they have problems such as high cost and difficult maintenance. In view of the problems existing in existing industrial microphones, the inventors of this case conducted research. Through continuous research, they finally proposed an industrial microphone, which is a single device and can realize the acquisition, processing, pushing, etc. of audio data by itself.

[0050] Next, the industrial microphone provided by this application will be introduced through the following embodiments.

[0051] Please refer to Figure 2 , which shows a schematic structural diagram of an industrial microphone provided by an embodiment of this application. This industrial microphone is a multi-layer architecture device, which may include: a hardware layer 201, a resource virtualization layer 202, and a software virtualization layer 203.

[0052] The hardware layer 201 in this embodiment may include a microphone array, a processing unit, and a communication unit. In addition to the above three units, the hardware layer may also include a storage unit and a power management unit. Optionally, the hardware layer 201 may also include an audio processing unit.

[0053] Among them, the microphone array is used to collect audio data, the processing unit is used to process audio data, the communication unit is used to communicate with other devices, for example, to send audio data to other devices, receive data sent by other devices, etc., the storage unit is used to store data, such as storing the data collected by the microphone array, storing the data processed by the processing unit, storing other data, etc., the power management unit is used to provide power voltage for each unit, and the audio processing unit is used to connect to an audio playback device, and the audio playback device can play the audio data collected by the microphone array or the audio data processed by the processing unit.

[0054] Such as Figure 3As shown, the microphone array of the hardware layer 201 can, but is not limited to, adopt a circular microphone array (such as a circular 8-microphone array. If a circular 8-microphone array is adopted, the maximum pickup distance can reach 10 meters). For the circular microphone array, directivity configuration can be performed according to the scene requirements, that is, configure an appropriate direction according to the scene requirements. Considering that the digital silicon microphone has a better and wider frequency response and stronger anti-interference ability than the traditional analog microphone, the microphones of the microphone array can adopt digital silicon microphones.

[0055] The processing unit of the hardware layer 201 can, but is not limited to, adopt a quad-core processor with 64-bit processing capabilities, such as the ARM Cortex-A53, which supports the ARM TrustZone security technology and can provide a secure operating environment for trusted programs.

[0056] As Figure 3 shown, the storage unit of the hardware layer 201 can, but is not limited to, include an embedded multimedia card eMMC and a double data rate synchronous dynamic random access memory DDR. Optionally, the storage unit of the hardware layer 201 can also include a TF memory card. Optionally, the storage unit of the hardware layer 201 can also include an SD card.

[0057] The communication unit of the hardware layer 201 can include a wired communication unit. As Figure 3 shown, the wired communication unit can include a Switch chip, and the Switch chip can be connected to a router device or a cluster gateway for realizing data forwarding. Optionally, in addition to the wired communication unit, the communication unit can also include a wireless communication unit. As Figure 3 shown, the wireless communication unit can include a Bluetooth communication unit and / or a WIFI communication unit (such as 2.4GHz Wi-Fi, supporting the 5GHz IEEE802.11 a / b / g / n / ac protocol).

[0058] The power management unit of the hardware layer 201 supports a wide voltage input of 9 - 18V and also supports a network port POE (IEEE802.3at) input of 42V~57V. It can support up to 5 device cascades within 100 meters.

[0059] Optionally, the hardware layer 201 can also include a display unit for displaying information, an input unit for inputting information, etc.

[0060] In this embodiment, the resource virtualization layer 202 is used to interact with the hardware resources of the hardware layer to realize the management of the hardware resources of the hardware layer. The resource virtualization layer 202 schedules and manages the virtual resource modules, and can communicate and interact with the hardware resources of the hardware layer by controlling the virtual resource modules. As Figure 3As shown in the figure, the virtual resource module may include: an audio driver module, a WIFI driver module, a Bluetooth driver module, a power management module, a flash driver module, an input driver module, a DDR driver module, and so on.

[0061] In this embodiment, the software virtualization layer 203 is used to execute, in cooperation with the resource virtualization layer 202 and with the support of the hardware layer 201: collect audio data; perform noise reduction processing and / or anomaly recognition processing on the collected audio data to obtain an audio processing result; and send the audio processing result to the server device.

[0062] Preferably, the collected audio data can be first subjected to noise reduction processing to obtain noise-reduced audio data (noise reduction processing is performed on multi-channel audio data to obtain single-channel noise-reduced audio data), and then anomaly recognition is performed on the noise-reduced audio data to obtain an anomaly recognition result. Furthermore, the noise-reduced audio data and the anomaly recognition result can be sent to the server device. Optionally, a noise reduction algorithm based on beamforming, a noise reduction algorithm based on a neural network, etc. can be used to perform noise reduction processing on the collected audio data.

[0063] The industrial microphone provided by the embodiment of the present application includes a hardware layer, a resource virtualization layer, and a software virtualization layer. The hardware layer includes a microphone array, a processing unit, a communication unit, etc. The resource virtualization layer can interact with the hardware resources of the hardware layer. The software virtualization layer can, in cooperation with the resource virtualization layer and with the support of the hardware layer, implement audio data collection, noise reduction, anomaly recognition, and network transmission. The industrial microphone provided by the embodiment of the present application is a device integrating many functions such as audio collection, audio noise reduction, audio anomaly recognition, and audio push. Since the industrial microphone provided by the embodiment of the present application is a single device rather than a system composed of multiple devices, the cost and maintenance difficulty are greatly reduced.

[0064] In another embodiment of the present application, the software virtualization layer in the above embodiment is introduced.

[0065] In a possible implementation manner, as Figure 3 shown, the software virtualization layer may include an audio collection module, an audio noise reduction module, an audio anomaly recognition module, and an audio encapsulation and push module.

[0066] Among them, the audio acquisition module realizes audio acquisition with the cooperation of the resource virtualization layer and the support of the microphone array in the hardware layer. The audio noise reduction module realizes noise reduction processing on the acquired audio with the cooperation of the resource virtualization layer and the support of the processing unit in the hardware layer. The audio anomaly recognition module realizes anomaly recognition processing on the noise-reduced audio with the cooperation of the resource virtualization layer and the support of the processor in the hardware layer. The audio encapsulation and push module realizes the encapsulation and push of the noise-reduced audio data and the anomaly recognition result with the cooperation of the resource virtualization layer and the support of the processing unit and the communication unit in the hardware layer.

[0067] In a possible implementation, as Figure 4 shown, the process of the audio anomaly recognition module in the software virtualization layer performing anomaly recognition processing on the noise-reduced audio may include:

[0068] Step S401: Obtain the acoustic features of the noise-reduced audio data.

[0069] The acoustic features of the noise-reduced audio data are features that can characterize information such as the physical properties, frequency distribution, and energy change of sound.

[0070] Step S402: Use the pre-trained anomaly recognition model to extract features from the acoustic features.

[0071] In a possible implementation, the anomaly recognition model can be an unsupervised machine learning model, which can learn effective feature representations of audio data from a large amount of unlabeled training audio data. When training the model with the training audio data, the training audio data can be first enhanced, then the acoustic features of the enhanced data are obtained, and then the obtained acoustic features are input into the model for training.

[0072] The anomaly recognition model in this embodiment may include an encoder and a decoder. The encoder obtains the effective feature representation of the input data, and the decoder reconstructs the input data according to the effective feature representation obtained by the encoder.

[0073] The acoustic features of the noise-reduced audio data can be input into the encoder of the anomaly recognition model, and the encoder extracts features from the input acoustic features.

[0074] Step S403: Use the anomaly recognition model to reconstruct the acoustic features according to the extracted features.

[0075] The extracted features can be input into the decoder of the anomaly recognition model, and the decoder reconstructs the acoustic features according to the input features.

[0076] Step S404: Determine whether there is an anomaly in the noise-reduced audio data according to the acoustic features of the noise-reduced audio data and the reconstructed acoustic features.

[0077] Specifically, the reconstruction residual value can be calculated based on the acoustic features of the denoised audio data and the reconstructed acoustic features. The reconstruction residual can characterize the difference between the acoustic features of the denoised audio data and the reconstructed acoustic features. Furthermore, based on the calculated reconstruction residual value and a set threshold, it can be determined whether there is an abnormality in the denoised audio data. Specifically, if the reconstruction residual value is greater than the set threshold, it is determined that there is an abnormality in the denoised audio; if the reconstruction residual value is less than or equal to the set threshold, it is determined that there is no abnormality in the denoised audio.

[0078] Optionally, after obtaining the anomaly recognition result, if the anomaly recognition result indicates that there is an abnormality in the denoised audio data, further anomaly category recognition can be performed, that is, the specific anomaly category existing in the denoised audio data can be identified.

[0079] After obtaining the denoised audio data and the anomaly recognition result, the audio encapsulation and push module of the software virtualization layer can encapsulate the denoised audio and the anomaly recognition result, and then send the encapsulated data to the server device.

[0080] In a possible implementation manner, according to the State Grid specification protocol, the denoised audio data and the anomaly recognition result can be encapsulated to obtain the first encapsulated data, and then the first encapsulated data is sent to the server device.

[0081] In another possible implementation manner, the denoised audio data can be encoded, and according to the Open Network Video Interface Forum (ONVIF) protocol or the Video Surveillance Networking Protocol GB28181, the encoded data and the anomaly recognition result are encapsulated to obtain the second encapsulated data, and then the second encapsulated data is sent to the server device.

[0082] In actual application, according to the requirements of the actual scenario, one of the above two implementation manners can be selected for data encapsulation and pushing.

[0083] In addition, it should be noted that the server device can send a data sending start instruction and a data sending stop instruction to the industrial microphone. When the software virtualization layer of the industrial microphone receives the data sending start instruction sent by the server device, it encapsulates and pushes the denoised audio and the anomaly recognition result. When it receives the data sending stop instruction sent by the server device, it stops sending data to the server device.

[0084] In a possible implementation manner, the industrial microphone can establish a Transmission Control Protocol (TCP) connection with the server device, and then the software virtualization layer transmits the encapsulated data obtained by encapsulating the denoised audio data and the anomaly recognition result according to the State Grid specification protocol to the server device through the TCP connection.

[0085] In a possible implementation, the industrial microphone can establish a Message Queuing Telemetry Transport (MQTT) connection with the server device. Furthermore, the server device can send a data transmission start instruction and a data transmission stop instruction to the industrial microphone through the MQTT connection. The data transmission start instruction and the data transmission stop instruction are MQTT messages, and the MQTT messages in JSON format are shown in the following table.

[0086] Table 1 MQTT Messages in JSON Format

[0087]

[0088] In another embodiment of the present application, as Figure 5 shown, in addition to including a hardware layer 201, a resource virtualization layer 202, and a software virtualization layer 203, the industrial microphone may further include a user interaction layer 204.

[0089] The user interaction layer 204 is used to obtain the configuration information for the user to configure the industrial microphone. Furthermore, the software virtualization layer 203 can process the configuration information. Specifically, the software virtualization layer 203 may further include a configuration information processing module, and the configuration information processing module is used to process the configuration information for the user to configure the industrial microphone.

[0090] The user can connect the terminal to the industrial microphone in a wired or wireless manner. Furthermore, the user can log in to the fixed IP address through the terminal to enter the configuration interface of the industrial microphone. The user can perform some or all of the following configurations in the configuration interface of the industrial microphone: pickup configuration, network configuration, transmission configuration, audio acquisition configuration, system configuration, device upgrade, etc.

[0091] Among them, the configuration item examples of the pickup configuration are as follows:

[0092] (1) Direction configuration: Pickup angle.

[0093] (2) Sampling rate configuration: Configure the sampling rate to 48 kHz or 192 kHz.

[0094] (3) Sampling bit depth configuration: Configure the sampling bit depth to 16 bit or 32 bit.

[0095] (4) Gain: Add or subtract the decibel value, and only integer settings are supported.

[0096] (5) Enable local recording: After checking, the recording will be stored in the TF card, and it is not checked by default.

[0097] (6) Slice length: The time length of a single file stored locally, in seconds, with a default of 60 s. Only positive integers are supported for configuration, and this item can only be configured after local recording is enabled.

[0098] (7) Save: After modifying the above configurations, click the Save button.

[0099] (8) Reset: Click the "Reset" button, and the parameters will be restored to the factory settings and take effect after saving.

[0100] (9) SD Card Status: When the SD card is not inserted, the status, capacity, and format button are not displayed. When inserted, they are displayed.

[0101] (10) Format: Perform a formatting operation on the SD card.

[0102] Among them, the configuration item examples of network configuration (Ethernet) are as follows:

[0103] (1) Local IP: The default is 192.168.1.66, and custom modification is supported.

[0104] (2) Subnet Mask: The default is 255.255.255.0, and custom modification is supported.

[0105] (3) Default Gateway: The default is 192.168.1.1, and custom modification is supported.

[0106] (4) MAC Address: The MAC address of the wired network card, and modification is not supported.

[0107] (5) Save: After modifying the above configurations, click the Save button, and it will take effect after restarting the device according to the extraction language.

[0108] (6) Reset: After clicking the "Reset" button, the parameters will be restored to the factory settings and take effect after clicking Save and restarting.

[0109] Among them, the configuration item examples of network configuration (WIFI) are as follows:

[0110] (1) Turn on WIFI: The default is on, and turning off is supported.

[0111] (2) Type: The default is AP mode, and switching between AP and STA modes is supported. When switching to STA mode, connect to the network by filling in the WIFI account and password.

[0112] (3) Save: After modifying the above configurations, click the Save button, and it will take effect after restarting the device according to the extraction language. The network information obtained in STA mode can be viewed on this page, and the obtained network information does not support custom modification.

[0113] (4) Reset: After clicking the "Reset" button, the parameters will be restored to the factory settings and take effect after clicking Save and restarting.

[0114] Among them, the configuration item examples of transmission configuration are as follows:

[0115] (1) Enable upload: Default is off. Data can only be uploaded to the server after enabling.

[0116] (2) Server port: Fill in the server port. The default is 18090.

[0117] (3) Save: Click the save button after modifying the above configuration.

[0118] (4) Upload data type: The default is "one-way data". This one-way data is the audio data of one channel after noise reduction from multiple (such as 8) microphone channels; if the original data is selected, the uploaded data is the data without noise reduction.

[0119] (5) Enable control: Default is off. A connection to the MQTT server can only be established after enabling.

[0120] (6) MQTT server address: Fill in the IP address of the MQTT server. The default is 192.168.1.67.

[0121] (7) MQTT server port: Fill in the port of the MQTT server. The default is 1883.

[0122] (8) Clientid: Fill in the client ID. The default is empty.

[0123] (9) Listening Topic: Fill in the subscription topic. The default is / control.

[0124] (10) Qos: Fill in the service quality level. The default is 2, and 0, 1, 2 are supported.

[0125] (11) Username: Fill in the username of the MQTT server. The default is empty.

[0126] (12) Password: Fill in the password of the MQTT server. The default is empty.

[0127] (13) Save: Click the save button after modifying the above settings.

[0128] Among them, the configuration item examples of the audio acquisition configuration are as follows:

[0129] (1) Play: Click the "Play" button to play the real-time audio. A time-domain graph is displayed on the page. Click "Pause" to stop playing.

[0130] (2) Start recording: Click to start caching the audio to the browser.

[0131] (3) End recording: Click to download the audio to the PC browser.

[0132] Among them, the configuration item examples of the system configuration are as follows:

[0133] (1) System account: Customize the web login account.

[0134] (2) System password: Customize the web login password.

[0135] (3) Save: After modifying the above configuration, you need to click the Save button, and it will take effect after logging out.

[0136] (4) Reset: After clicking the “Reset” button, the username and password will be restored to factory settings. The settings will take effect after you save and log out.

[0137] The following are examples of configuration items for device upgrade:

[0138] (1) Place the system OTA package / software rpm package on the local PC.

[0139] (2) On the web page, click "Select File" and select the corresponding package.

[0140] (3) Click the Upgrade button.

[0141] (4) The device automatically restarts after the upgrade is completed.

[0142] In addition to the above configurations of the industrial microphone, users can also perform the following configurations: configure the industrial microphone to upload audio data to the customer's private server; configure the industrial microphone to send the collected original audio data or noise-reduced audio data to the edge gateway, so that the edge gateway can perform abnormality identification processing on the original audio data or noise-reduced audio data.

[0143] It should be noted that after collecting audio data, the industrial microphone provided in the embodiment of the present application can directly perform noise reduction, abnormality identification and other processing on the collected audio data. After collecting audio data, the industrial microphone provided in the embodiment of the present application can also perform noise reduction processing on the collected audio data, and then send the noise-reduced audio data to the edge gateway, and the edge gateway performs abnormality identification on the noise-reduced audio data (an abnormality identification model is deployed on the edge gateway). After collecting audio data, the industrial microphone provided in the embodiment of the present application can also send the collected audio data directly to the edge gateway, and the edge gateway performs noise reduction, abnormality identification and other processing on the collected audio data (the edge gateway may also only perform abnormality identification processing on the collected audio data).

[0144] The industrial microphone provided by the embodiment of the present application is a single device, which can realize many functions such as audio acquisition, audio noise reduction, audio anomaly recognition, audio encapsulation, and audio push by itself. Compared with the existing industrial microphones (a large system composed of multiple devices), the industrial microphone provided by the embodiment of the present application has a lower cost and is easier to maintain. In addition, the industrial microphone provided by the embodiment of the present application can use a ring silicon microphone array to collect audio data. Since the ring silicon microphone array has characteristics such as a higher collection frequency, configurable directivity, and good sound pickup effect, the industrial microphone provided by the embodiment of the present application has a better audio collection effect and applicability. In addition, the industrial microphone provided by the embodiment of the present application supports the national grid standard protocol and the MQTT transmission protocol.

[0145] Finally, it should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover a non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0146] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0147] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An industrial microphone, characterized in that: include: Hardware layer, resource virtualization layer, and software virtualization layer; The hardware layer includes: a microphone array for collecting audio data, a processing unit for processing audio data, a storage unit for storing audio data, a communication unit for communicating with other devices, and a power management unit for providing power supply voltage to each unit; The resource virtualization layer is used to interact with the hardware resources of the hardware layer; The software virtualization layer is used to perform, with the cooperation of the resource virtualization layer and the support of the hardware layer: Collect audio data; Performing noise reduction processing and / or abnormality recognition processing on the collected audio data to obtain an audio processing result; The audio processing result is sent to the server device.

2. The industrial microphone according to claim 1, characterized in that The industrial microphone further comprises: a user interaction layer; The user interaction layer is used to obtain configuration information of the industrial microphone configured by the user in the microphone configuration interface, wherein the configuration of the industrial microphone by the user includes part or all of the following configurations: sound pickup configuration, network configuration, transmission configuration, audio acquisition configuration, system configuration, and equipment upgrade configuration; The software virtualization layer is also used to process the configuration information.

3. The industrial microphone according to claim 1, characterized in that The hardware layer also includes: An audio processing unit for connecting to an audio playback device.

4. The industrial microphone according to claim 1, characterized in that: When the software virtualization layer performs noise reduction processing and / or abnormality identification processing on the collected audio data with the cooperation of the resource virtualization layer and the support of the hardware layer, it is specifically used to: Performing noise reduction processing on the collected audio data to obtain noise-reduced audio data; Acquiring acoustic features of the denoised audio data; Extracting features from the acoustic features using a pre-trained anomaly recognition model; Reconstructing acoustic features based on the extracted features using the anomaly recognition model; According to the reconstructed acoustic features and the acoustic features of the denoised audio data, it is determined whether the denoised audio data has an abnormality.

5. The industrial microphone according to claim 1, characterized in that The audio processing result includes denoised audio data obtained by performing denoising processing on the collected audio data, and an abnormality recognition result obtained by performing abnormality recognition on the denoised audio data; When the software virtualization layer sends the audio processing result to the server device with the cooperation of the resource virtualization layer and the support of the hardware layer, it is specifically used to: According to the State Grid standard protocol, the denoised audio data and the abnormal recognition result are packaged to obtain first packaged data, and the first packaged data is sent to the server device; and / or, The noise-reduced audio data is encoded, and the encoded data and the abnormal identification result are packaged according to the Open Network Video Interface Forum ONVIF protocol or the video surveillance networking protocol GB28181 to obtain second packaged data, and the second packaged data is sent to the server device.

6. The industrial microphone according to claim 1, characterized in that When the software virtualization layer sends the audio processing result to the server device with the cooperation of the resource virtualization layer and the support of the hardware layer, it is specifically used to: When receiving a data transmission start instruction sent by the server device, sending the audio processing result to the server device; The software virtualization layer is further used to execute, with the cooperation of the resource virtualization layer and the support of the hardware layer: When receiving a data transmission stop instruction sent by the server device, stop sending data to the server device.

7. The industrial microphone according to claim 6, characterized in that The software virtualization layer sends the audio processing result to the server device through a transmission control protocol TCP connection established with the server device; The software virtualization layer receives a data transmission start instruction and a data transmission stop instruction sent by the server device through a message queue telemetry transmission MQTT connection established with the server device.

8. The industrial microphone according to claim 1, characterized in that The microphone array is a ring-shaped microphone array.

9. The industrial microphone according to claim 1, characterized in that: The microphones in the microphone array are digital silicon microphones.

10. The industrial microphone according to claim 1, characterized in that The communication unit includes a wired communication unit and a wireless communication unit; The wired communication unit includes a Switch chip, which can be connected to a router device or a cluster gateway to achieve data forwarding; The wireless communication unit includes: a Bluetooth communication module and / or a WIFI communication module.

Images