Microstructured artificial larynx device

Abstract

The microstructure artificial laryngeal device provided by the application comprises a micro-porous structure audio sensor, a lower computer and an upper computer, wherein the micro-porous structure audio sensor is used for detecting a vibration signal of a throat and converting the vibration signal into a resistance signal; the lower computer is used for filtering and processing the resistance signal to remove noise interference; and the upper computer is used for accurately identifying audio information in the processed resistance signal. The micro-porous structure audio sensor is attached at a skin surface of a thyroid cartilage, the lower computer is worn at a neck, and the upper computer is worn at a small arm. The microstructure artificial laryngeal device provided by the application can be directly attached to a throat of a patient without being installed in the patient's body, thereby avoiding discomfort of the patient, in addition, the vibration signal of the throat can be better sensed through the micro-porous structure audio sensor, thereby completing sound detection, and in addition, the lower computer for removing noise interference is arranged, and the accuracy of speech recognition can be improved.

Description

Technical Field

[0001] This application relates to the field of voice-assisted recognition, specifically to a microstructured artificial larynx device. Background Technology

[0002] Currently, there are millions of people with language disorders worldwide. Some have language impairments due to congenital defects, while others have lost their language function due to acquired diseases. Language disorders bring great difficulties and inconvenience to their lives. The famous physicist Stephen Hawking suffered from Parkinson's disease and was unable to communicate normally through language. Engineers at Intel designed a typing system for him based on eye-tracking technology, allowing him to select the content he needed to express by moving his eyes. However, the technology was extremely expensive, making widespread adoption difficult.

[0003] In addition, an artificial larynx is a simple speech rehabilitation method. It is usually installed inside the mouth in the larynx. Airflow from the lungs is modulated by the tongue and lips, causing the artificial larynx membrane to vibrate and produce speech signals. As an assistive device for people with speech disorders, the artificial larynx needs to be installed inside the mouth, which can cause discomfort for patients.

[0004] Therefore, how to enable artificial larynx devices to assist people with speech impairments without causing them discomfort is a challenge that existing manufacturers need to overcome. Summary of the Invention

[0005] This application provides a microstructured artificial larynx device that can solve the technical problem that existing artificial larynx devices cause discomfort to patients when assisting those with speech impairments.

[0006] In a first aspect, this application provides a microstructured artificial larynx device, comprising: a microporous audio sensor, wherein the microporous audio sensor is used to detect vibration signals of the larynx and convert the vibration signals into resistance signals;

[0007] The lower-level machine is electrically connected to the microporous audio sensor. The lower-level machine is used to receive the resistance signal and filter the resistance signal to remove noise interference.

[0008] The host computer is electrically connected to the slave computer. The host computer is used to receive the processed resistance signal and classify the processed resistance signal to accurately identify the audio information in the processed resistance signal.

[0009] The microporous audio sensor provided in this application comprises a microporous array substrate, a pressure sensing layer, an electrode pattern layer, and an encapsulation layer stacked sequentially.

[0010] In the microporous audio sensor provided in this application, a plurality of first micropores are provided on one surface of the microporous array substrate, and the first micropores are arranged in an array.

[0011] In the micropore structure audio sensor provided in this application, the diameter of the first micropore is 0.09 mm to 0.11 mm, the depth of the first micropore is 0.48 mm to 0.52 mm, and the number of the first micropores is 380 to 420.

[0012] In the microporous audio sensor provided in this application, the electrode pattern layer includes multiple gold interdigitated electrode patterns, and the gold interdigitated electrode patterns are wavy.

[0013] In the microporous audio sensor provided in this application, the pressure sensing layer includes multiple carbon nanotube pressure sensing structures, and the carbon nanotube pressure sensing structures are arranged in a one-to-one correspondence with the gold interdigitated electrode pattern.

[0014] In the microporous audio sensor provided in this application, the lower-level machine includes a multiplexed signal circuit module, a microcontroller analog-to-digital converter module, and a wireless transmission module; wherein,

[0015] The multiplexed signal circuit module is electrically connected to the microporous audio sensor and the analog-to-digital converter module. The multiplexed signal circuit module is used to receive the resistance signal and output the resistance signal to the analog-to-digital converter module.

[0016] The microcontroller analog-to-digital converter module is electrically connected to the wireless transmission module, filters the resistance signal to remove noise interference, and transmits the processed resistance signal to the wireless transmission module.

[0017] The wireless transmission module is used to transmit the processed resistance signal to the host computer.

[0018] In the micro-hole structure audio sensor provided in this application, the lower-level machine further includes a microcontroller module, which is electrically connected to the multiplexed signal circuit module and is used to control the operation of the multiplexed signal circuit module.

[0019] In the microporous audio sensor provided in this application, the host computer includes a wireless receiving and communication module, a signal storage module, and a one-dimensional convolutional neural network classifier module; wherein,

[0020] The wireless receiving and communication module is electrically connected to the lower-level machine and the signal storage module. The wireless receiving and communication module is used to receive the processed resistance signal and transmit the processed resistance signal to the signal storage module.

[0021] The signal storage module is connected to the one-dimensional convolutional neural network classifier module. The signal storage module is used to store the processed resistance signal and output the processed resistance signal to the one-dimensional convolutional neural network classifier module after multiple sets of the processed resistance signal are received.

[0022] The one-dimensional convolutional neural network classifier module is used to classify and identify the processed resistance signal in order to obtain the audio information in the processed resistance signal.

[0023] In the microporous audio sensor provided in this application, the recognition result display module is electrically connected to the one-dimensional convolutional neural network classifier module, and the recognition result display module is used to display the audio information in the processed resistance signal.

[0024] The microstructured artificial larynx device provided in this application includes a microporous audio sensor, a lower-level computer, and a higher-level computer. The microporous audio sensor detects vibration signals from the larynx and converts them into resistance signals. The lower-level computer receives the resistance signals and filters them to remove noise interference. The higher-level computer receives the processed resistance signals and classifies them for accurate identification of audio information. The microporous audio sensor is attached to the skin surface of the thyroid cartilage, the lower-level computer is worn in the neck area, and the higher-level computer is worn in the forearm area. The microstructured artificial larynx device provided in this application can be directly attached to the patient's larynx without being implanted inside the patient's body, thus avoiding discomfort. Furthermore, the microporous audio sensor allows for better detection of larynx vibration signals, enabling sound detection. The inclusion of a noise-removing lower-level computer further improves the accuracy of speech recognition. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the microstructured artificial larynx device provided in the embodiments of this application.

[0026] Figure 2 This is a schematic diagram of the first structure of the microporous audio sensor provided in an embodiment of this application.

[0027] Figure 3 This is a schematic diagram of the second structure of the microporous audio sensor provided in an embodiment of this application.

[0028] Figure 4 A schematic diagram of the structure of the electrode pattern layer provided in the embodiments of this application.

[0029] Figure 5 This is a first structural schematic diagram of the lower-level machine provided in an embodiment of this application.

[0030] Figure 6 This is a schematic diagram of the second structure of the lower-level machine provided in an embodiment of this application.

[0031] Figure 7 The circuit diagram of the lower-level machine provided in the embodiments of this application.

[0032] Figure 8 This is a first structural schematic diagram of the host computer provided in an embodiment of this application.

[0033] Figure 9 This is a schematic diagram of the second structure of the host computer provided in an embodiment of this application. Detailed Implementation

[0034] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0035] Furthermore, the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish different objects, not to describe a specific order. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. Since the source and drain of the transistor used in this application are symmetrical, their sources and drains are interchangeable. According to the configuration shown in the drawings, the middle terminal of the transistor is designated as the gate, the signal input terminal as the source, and the output terminal as the drain.

[0036] Please see Figure 1 , Figure 1 This is a first structural schematic diagram of the microstructured artificial larynx device provided in an embodiment of this application. Figure 1 As shown, the microstructured artificial larynx device 10 provided in this application embodiment includes a microporous audio sensor 101, a lower-level computer 102, and a higher-level computer 103.

[0037] The microporous audio sensor 101 is used to detect vibration signals from the throat and convert them into resistance signals. The lower-level computer 102 is electrically connected to the microporous audio sensor 101. The lower-level computer 102 receives the resistance signals and filters them to remove noise interference. The upper-level computer 103 is electrically connected to the lower-level computer 102. The upper-level computer 103 receives the processed resistance signals and classifies them for accurate identification of audio information within the processed resistance signals.

[0038] It should be noted that the microporous audio sensor 101 can better sense vibration signals from the throat and convert them into resistance signals. Resistance signals are easier to detect than vibration signals and facilitate subsequent recognition, thus improving sound detection. The lower-level computer 102 can also perform filtering to remove noise interference, facilitating subsequent speech recognition and improving accuracy. The upper-level computer 103 performs speech recognition by first classifying the speech before recognition, which not only improves accuracy but also increases efficiency.

[0039] The microporous audio sensor 101 is attached to the skin surface of the thyroid cartilage, the lower unit 102 is worn at the neck, and the upper unit 103 is worn at the forearm. The microstructured artificial larynx device 10 provided in this application can be directly attached to the patient's larynx without being installed inside the patient's body, thus avoiding discomfort to the patient.

[0040] Please visit Figure 2 as well as Figure 3 , Figure 2 This is a schematic diagram of the first structure of the microporous audio sensor provided in an embodiment of this application. Figure 3 This is a schematic diagram of the second structure of the microporous audio sensor provided in an embodiment of this application. Figure 2 as well as Figure 3 As shown, the microporous audio sensor 101 comprises a microporous array substrate 1011, a pressure sensing layer 1012, an electrode pattern layer 1013, and an encapsulation and attachment layer 1014, which are stacked sequentially.

[0041] Among them, a plurality of first micropores 1011a are provided on one surface of the micropore array substrate 1011, and the first micropores 1011a are arranged in an array.

[0042] It should be noted that by setting multiple first micropores 1 0 1 1a arranged in an array, the vibration of the patient's larynx can be better sensed, thereby obtaining accurate vibration signals. Specifically, since the microporous audio sensor 1 01 is attached to the surface of the human body, and due to airtightness and body surface temperature, there is a slightly higher air pressure inside each first micropore 1 0 1 1a than outside, which allows the vibration of the vocal cord muscles to be transmitted more clearly to its upper surface, thus obtaining accurate vibration signals.

[0043] The diameter of the first micropore 1011a is between 0.09 mm and 0.11 mm. Specifically, the diameter of the first micropore 1011a is 0.090 mm, 0.092 mm, 0.094 mm, 0.096 mm, 0.100 mm, 0.104 mm, or 0.110 mm. The specific diameter of the first micropore 1011a is determined by the specific requirements of the microstructured artificial larynx device 10 provided in the embodiments of this application.

[0044] The depth of the first micropore 1011a is 0.48 mm to 0.52 mm. Specifically, the depth of the first micropore 1011a is 0.480 mm, 0.482 mm, 0.484 mm, 0.492 mm, 0.500 mm, 0.510 mm, or 0.520 mm. The specific depth of the first micropore 1011a is determined by the specific requirements of the microstructured artificial larynx device 10 provided in this application embodiment.

[0045] The number of first micropores 1011a is between 380 and 420. Specifically, the number of first micropores 1011a is 380, 385, 390, 395, 400, 410, or 420. The specific number of first micropores 1011a is determined by the specific requirements of the microstructured artificial larynx device 10 provided in the embodiments of this application.

[0046] The microporous array substrate 1011 is made of silicone. Specifically, it is manufactured by pouring a silicone solution into an acrylic mold, followed by pressing and drying in a drying oven. It should be noted that silicone is chosen as the material for the microporous array substrate 1011 because it possesses good elasticity, which not only reduces vibrations caused by swallowing and breathing but also exhibits good biocompatibility. Of course, other elastic materials can also be used to make the microporous array substrate 1011; no specific limitations are made here.

[0047] Please visit Figure 4 , Figure 4This is a schematic diagram of the structure of the electrode pattern layer provided in an embodiment of this application. Figure 4 As shown, the electrode pattern layer 1013 includes multiple gold interdigitated electrode patterns 1013a, and the gold interdigitated electrode patterns 1013a are wavy.

[0048] It should be noted that by setting the gold infeed electrode pattern 1013a to a wavy shape, firstly, it facilitates the setting and arrangement of other components; secondly, it can improve the conductivity of the gold infeed electrode pattern 1013a, enabling the acquisition of more accurate resistance signals; and thirdly, since the gold infeed electrode pattern 1013a is a delicate component, the wavy shape of the gold infeed electrode pattern 1013a can withstand the application concentration caused by attaching to curved surfaces, thereby preventing damage to the gold infeed electrode pattern 1013a.

[0049] It should be noted that the end of the gold interdigitated electrode pattern 1013a has a rectangular electrode interface for connecting to an external circuit. Of course, other specific interface shapes can also be used; this application embodiment does not impose specific limitations here, but is determined by the specific requirements of the microstructure artificial larynx device 10 provided in this application embodiment.

[0050] It should be noted that the distance between each parallel gold interdigitated electrode pattern 1013a is 49 to 51 micrometers. Specifically, the distance between each parallel gold interdigitated electrode pattern 1013a is 49 micrometers, 49.2 micrometers, 49.4 micrometers, 49.6 micrometers, 50 micrometers, 50.4 micrometers, or 51 micrometers. The specific distance between each parallel gold interdigitated electrode pattern 1013a is determined by the specific requirements of the microstructured artificial larynx device 10 provided in the embodiments of this application.

[0051] The total length of the parallel gold interdigitated electrode pattern 1013a is 2.4 to 2.6 mm.

[0052] The pressure-sensing layer 1012 includes multiple carbon nanotube pressure-sensing structures 1012a, which are arranged in a one-to-one correspondence with the gold interdigitated electrode pattern 1013a. Furthermore, the upper and lower edges of the carbon nanotube pressure-sensing structures 1012a and the gold interdigitated electrode pattern 1013a are aligned.

[0053] It should be noted that the carbon nanotube pressure sensing structure 1012a is mainly used to sense the vibration of the patient's larynx to obtain the vibration signal, while the gold finger electrode pattern 1013a is used to convert the vibration signal into a more easily identifiable resistance signal. By setting the carbon nanotube pressure sensing structure 1012a and the gold finger electrode pattern 1013a in a one-to-one correspondence, a one-to-one detection can be guaranteed, which can not only avoid detection errors and improve detection accuracy, but also improve the efficiency of voice detection.

[0054] It should be noted that the pressure sensing layer 1012 is fabricated by first depositing multiple carbon nanotube pressure sensing films on a copper foil, and then transferring them to a microporous array substrate 1011.

[0055] The height of the carbon nanotube pressure-sensing structure 1012a is 24 mm to 26 mm. Specifically, the height of the carbon nanotube pressure-sensing structure 1012a is 24 mm, 24.2 mm, 24.4 mm, 24.6 mm, 25 mm, 25.4 mm, or 26 mm. The specific height of the carbon nanotube pressure-sensing structure 1012a is determined by the specific requirements of the microstructured artificial larynx device 10 provided in the embodiments of this application.

[0056] The width of the carbon nanotube pressure-sensing structure 1012a is 4 mm to 6 mm. Specifically, the height of the carbon nanotube pressure-sensing structure 1012a is 4 mm, 4.2 mm, 4.4 mm, 4.6 mm, 5 mm, 5.4 mm, or 6 mm. The specific height of the carbon nanotube pressure-sensing structure 1012a is determined by the specific requirements of the microstructured artificial larynx device 10 provided in the embodiments of this application.

[0057] The encapsulation layer 1014 is made of polyimide. It should be noted that polyimide is a type of polymer containing imide rings in its main chain. It is one of the best-performing organic polymer materials, possessing high temperature resistance and high insulation properties, thus providing excellent encapsulation.

[0058] The encapsulation attachment layer 1014 has a length of 49 to 51 mm. Specifically, the length of the encapsulation attachment layer 1014 is 49 mm, 49.2 mm, 49.4 mm, 49.6 mm, 50 mm, 50.4 mm, or 51 mm. The specific length of the encapsulation attachment layer 1014 is determined by the specific requirements of the microstructured artificial larynx device 10 provided in the embodiments of this application.

[0059] The width of the encapsulation layer 1014 is 49 to 51 mm. Specifically, the width of the encapsulation layer 1014 is 49 mm, 49.2 mm, 49.4 mm, 49.6 mm, 50 mm, 50.4 mm, or 51 mm. The specific width of the encapsulation layer 1014 is determined by the specific requirements of the microstructured artificial larynx device 10 provided in the embodiments of this application.

[0060] The surface of the encapsulation layer 1014 is further provided with multiple uniformly distributed second micropores. The second micropores are formed using a laser processing technology, which enables the encapsulation layer 1014 to have good air permeability and biocompatibility.

[0061] The lower surface of the encapsulation and attachment layer 1014 is adhesive, which facilitates the encapsulation of the microstructure artificial larynx device 10 provided in this application embodiment and the attachment of the microstructure artificial larynx device 10 provided in this application embodiment to the human body surface.

[0062] Please visit Figure 5 , Figure 5 This is a first structural schematic diagram of the lower-level machine provided in an embodiment of this application. (See attached diagram.) Figure 5 As shown, the lower-level machine 102 includes a multiplexed signal circuit module 1021, a microcontroller analog-to-digital converter module 1022, and a wireless transmission module 1023.

[0063] The multiplexing circuit module 1021 is electrically connected to the micro-hole structure audio sensor 101 and the microcontroller analog-to-digital converter module 1022. The multiplexing circuit module 1021 is used to receive the resistance signal and output the resistance signal to the analog-to-digital converter module 1022.

[0064] It should be noted that the multiplexing circuit module 1021 can connect different paths on the microporous audio sensor 101 to receive multiple signals at once, thereby improving the efficiency of speech recognition.

[0065] The microcontroller analog-to-digital converter module 1022 is electrically connected to the wireless transmitter module 1023. It filters the resistance signal to remove noise interference and transmits the processed resistance signal to the wireless transmitter module 1023.

[0066] It should be noted that the microcontroller analog-to-digital converter module 1022 can process the resistor signal using the Kalman filter algorithm, thereby reducing noise interference.

[0067] The wireless transmission module 1023 is used to transmit the processed resistance signal to the host computer 103.

[0068] Please visit Figure 6 as well as Figure 7 , Figure 6 This is a schematic diagram of the second structure of the lower-level machine provided in an embodiment of this application. Figure 7 The circuit diagram of the lower-level machine provided in the embodiments of this application. Figure 6 The lower-level machine 102 shown is... Figure 5 The difference between the lower-level machine 102 shown is that the lower-level machine 102 also includes a microcontroller controller module 1024. The microcontroller controller module 1024 is electrically connected to the multiplexer circuit module 1021, and the microcontroller controller module 1024 is used to control the operation of the multiplexer circuit module 1021.

[0069] It should be noted that the lower-level machine 102 can also use other control modules, such as inductive control modules and pressure-type control modules. This application embodiment does not impose specific limitations here; the specific control method adopted by the lower-level machine 102 is determined by the specific requirements of the microstructure artificial larynx device 10 provided in this application embodiment.

[0070] Please visit Figure 8 , Figure 8 This is a first structural diagram of the host computer provided in an embodiment of this application. Figure 8 As shown, the host computer 103 includes a wireless receiving and communication module 1031, a signal storage module 1032, and a one-dimensional convolutional neural network classifier module 1033. Among them,

[0071] The wireless receiving and communication module 1031 is electrically connected to the lower-level machine 102 and the signal storage module 1032. The wireless receiving and communication module 1031 is used to receive the processed resistance signal and transmit the processed resistance signal to the signal storage module 1032.

[0072] The signal storage module 1032 is connected to the one-dimensional convolutional neural network classifier module 1033. The signal storage module 1032 is used to store the processed resistance signals and outputs the processed resistance signals to the one-dimensional convolutional neural network classifier module 1033 after multiple sets of processed resistance signals are received.

[0073] It should be noted that multiple sets of processed resistance signals are first stored in the signal storage module 1032, and then sent to the one-dimensional convolutional neural network classifier module 1033 after all channels have been acquired. This can reduce the operating pressure caused by the system reading signals in real time.

[0074] Among them, the one-dimensional convolutional neural network classifier module 1 0 3 3 is used to classify and identify the processed resistance signal in order to obtain the audio information in the processed resistance signal.

[0075] It should be noted that the one-dimensional convolutional neural network classifier module 1033 has a built-in deep learning model that can classify and analyze different collected signals, thereby improving the accuracy and efficiency of speech recognition.

[0076] Please visit Figure 9 , Figure 9 This is a schematic diagram of the second structure of the host computer provided in an embodiment of this application. Figure 9 The host computer 103 shown is... Figure 8 The difference between the host computer 103 shown is that the host computer 103 also includes a recognition result display module 1034, which is electrically connected to the one-dimensional convolutional neural network classifier module 1033. The recognition result display module 1034 is used to display the audio information in the processed resistance signal.

[0077] The microstructured artificial larynx device provided in this application includes a microporous audio sensor, a lower-level computer, and a higher-level computer. The microporous audio sensor detects vibration signals from the larynx and converts them into resistance signals. The lower-level computer receives the resistance signals and filters them to remove noise interference. The higher-level computer receives the processed resistance signals and classifies them for accurate identification of audio information. The microporous audio sensor is attached to the skin surface of the thyroid cartilage, the lower-level computer is worn around the neck, and the higher-level computer is worn in the forearm. The microstructured artificial larynx device provided in this application can be directly attached to the patient's larynx without being implanted inside the patient's body, thus avoiding discomfort. Furthermore, the microporous audio sensor allows for better detection of larynx vibration signals, enabling sound detection. The inclusion of a noise-removing lower-level computer further improves the accuracy of speech recognition.

[0078] The above provides a detailed description of a microstructured artificial larynx device provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A microstructured artificial larynx device, characterized by The utility model relates to a throat audio recognition system, comprising: a micro-porous structure audio sensor for detecting a vibration signal of a throat and converting the vibration signal into an electric resistance signal; a lower computer electrically connected with the micro-porous structure audio sensor, configured to receive the electric resistance signal and filter the electric resistance signal to remove noise interference; an upper computer electrically connected with the lower computer, configured to receive the processed electric resistance signal and classify the processed electric resistance signal to accurately identify audio information in the processed electric resistance signal; the micro-porous structure audio sensor comprises a micro-porous array substrate, a pressure sensing layer, an electrode pattern layer and a packaging and attaching layer arranged in sequence; a surface of the micro-porous array substrate is provided with a plurality of first micro-holes arranged in an array; the electrode pattern layer comprises a plurality of gold interdigital electrode patterns in a wave shape; the pressure sensing layer comprises a plurality of carbon nanometer pressure sensing structures corresponding to the gold interdigital electrode patterns.

2. The microstructured artificial larynx device of claim 1, wherein, the diameter of the first micro-hole is 0.09-0.11 mm, the depth of the first micro-hole is 0.48-0.52 mm, and the number of the first micro-holes is 380-420.

3. The microstructured artificial larynx device of claim 1, wherein, the lower computer comprises a multi-channel signal multiplexing circuit module, a single-chip microcomputer analog-digital conversion module and a wireless transmitting module; wherein the multi-channel signal multiplexing circuit module is electrically connected with the micro-porous structure audio sensor and the single-chip microcomputer analog-digital conversion module, configured to receive the electric resistance signal and output the electric resistance signal to the analog-digital conversion module; the single-chip microcomputer analog-digital conversion module is electrically connected with the wireless transmitting module, configured to filter the electric resistance signal to remove noise interference and transmit the processed electric resistance signal to the wireless transmitting module; the wireless transmitting module is configured to transmit the processed electric resistance signal to the upper computer.

4. The microstructured artificial larynx device of claim 3, wherein, the lower computer further comprises a single-chip microcomputer controller module electrically connected with the multi-channel signal multiplexing circuit module, configured to control the multi-channel signal multiplexing circuit module to work.

5. The microstructured artificial larynx device of claim 1, wherein, the upper computer comprises a wireless receiving communication module, a signal storage module and a one-dimensional convolutional neural network classifier module; wherein the wireless receiving communication module is electrically connected with the lower computer and the signal storage module, configured to receive the processed electric resistance signal and transmit the processed electric resistance signal to the signal storage module; the signal storage module is connected with the one-dimensional convolutional neural network classifier module, configured to store the processed electric resistance signal and output a plurality of groups of the processed electric resistance signal to the one-dimensional convolutional neural network classifier module when the plurality of groups of the processed electric resistance signal are received. The one-dimensional convolutional neural network classifier module is used for classifying and identifying the processed resistance signal to obtain audio information in the processed resistance signal.

6. The microstructured artificial larynx device of claim 5, wherein, The upper computer further comprises an identification result display module, which is electrically connected with the one-dimensional convolutional neural network classifier module, and is used for displaying the audio information in the processed resistance signal.

Images