CN112672120B - Projector with speech analysis function and method for generating personal health data - Google Patents

Abstract

The invention relates to the field of projectors, in particular to a projector with a voice analysis function and a personal health data generation method, wherein the projector comprises: the circuit comprises a central control module, an optical projection module, an audio output module, a power supply module, a voice analysis module and an edge calculation voice processing module; the voice analysis module comprises a recording unit, an A/D conversion unit, a denoising unit and a voiceprint detection and extraction unit which are sequentially and electrically connected, the personal health data generation method comprises voice recording, signal conversion, denoising processing, voiceprint recognition and extraction and edge calculation, and the projector can collect information of daily life information, voiceprint characteristics and the like of the user and provide data support for health diagnosis of the user.

Description

Projector with speech analysis function and method for generating personal health data

technical field

The invention relates to the technical field of projectors, in particular to a projector with a speech analysis function and a method for generating personal health data.

Background technique

A projector is a device that can project images or videos onto a screen. Projectors are currently widely used in homes, multimedia conferences, schools, and entertainment venues. The current projectors have a single function and cannot collect information about users' daily life. Other physical and psychological information provides data support for the user's health diagnosis.

There are separate human health detectors in the prior art. For example, the patent with the patent number of CN109859843A can only detect the user's physiological data such as blood sugar, blood lipid and other data, and does not involve the collection and analysis of psychological and emotional changes, nor does it collect the user's daily diet. , drugs and other data to assist health diagnosis. In fact, if a doctor diagnoses a patient without considering the patient's long-term psychology, emotional state, daily diet, and drugs, misdiagnosis is likely to occur.

SUMMARY OF THE INVENTION

One object of the present invention is to provide a projector with a voice analysis function, which can realize the collection and processing of voice information and voiceprints, and provide data support for the user's physical health analysis; another object is to provide a voice analysis based on the above. A method of generating personal health data for a functional projector.

In order to achieve the above purpose, the present invention provides a projector with a voice analysis function, comprising: a casing and a circuit arranged inside the casing, the circuit includes a central control module, an optical projection module, an audio output module, a power supply module, and a voice analysis module. module, edge computing voice processing module; the central control module is respectively electrically connected with the optical projection module, audio output module, power supply module, voice analysis module, and edge computing voice processing module; the voice analysis module includes a recording unit, A/D The conversion unit, the denoising unit, the voiceprint detection and extraction unit, the recording unit is used to input the user's voice information and send the voice information to the A/D conversion unit to convert it into a digital voice signal, and the denoising unit denoises the digital voice signal. Then, it is transmitted to the voiceprint detection and extraction unit; the voiceprint detection and extraction unit receives the digital voice signal and extracts the voiceprint, and compares the voiceprint with the pre-stored user voiceprint. If the comparison is successful, the voiceprint and the digital voice signal are sent. To the edge computing voice processing module; the edge computing voice processing module is used to analyze and process the received voiceprint and digital voice signals to generate personal health data.

Further, the voice analysis module also includes a voice command control unit, and the voice command control unit is used to generate a command control signal according to the user's voice command and send it to the central control module, and the central control module controls the on-off of the projector and the projector according to the command control signal. The internal system operates; the voice command control unit is electrically connected with the denoising unit and the voiceprint detection and extraction unit.

Further, the circuit also includes a remote fingerprint switch module, the remote fingerprint switch module is electrically connected to the central control module or the remote fingerprint switch module is wirelessly connected to the central control module, the user controls the projector to turn on through the remote fingerprint switch module, and at the same time, the optical projection A white screen appears on the module; or, the user controls the projector to turn on by inputting a power-on voice command through the voice analysis module, and at the same time, a white screen appears on the optical projection module.

Further, the circuit further includes a heat dissipation module, and the heat dissipation module is electrically connected with the central control module, and is used for heat dissipation of the running projector.

Further, the power module includes a power control circuit, a lithium battery and an external power adapter interface, and the lithium battery and the external power adapter interface perform power switching and input through the power control circuit.

Further, the audio output module includes a built-in speaker and an audio output interface.

Further, the circuit further includes a wireless module, and the wireless module is electrically connected with the central control module.

Further, the circuit further includes a storage module, and the storage module is electrically connected with the central control module and the edge computing voice analysis module.

Further, the edge computing voice analysis module includes a high-efficiency cache unit, a neural network acceleration unit, a data visualization unit, and an abnormal network information feedback unit, wherein: the neural network acceleration unit is used to analyze the digital voice signal and voiceprint transmitted by the voice analysis module. Process and generate personal health data and send it to the high-efficiency cache unit for caching; the data visualization unit is used to visualize the personal health data generated by the neural network acceleration unit and the digital voice signals and voiceprints that cannot be processed by the neural network acceleration unit. The network information feedback unit performs feedback.

The present invention also provides a method for generating health data based on the above-mentioned projector with a speech analysis function, the method comprising:

S101: The projector is turned on by voice or switch;

S102: The voice analysis module starts to input user voice information;

S103: Convert the user voice information entered in step S103 into a digital voice signal and extract voiceprints;

S104: determine whether the extracted voiceprint matches the pre-stored voiceprint, if so, perform step S105, otherwise, end;

S105: Send the digital voice signal and voiceprint to the edge computing voice processing module;

S106: The edge computing voice processing module performs semantic recognition on the received digital voice signal and divides the received digital voice signal into diet information, medication information, emotional information and life language information according to the semantic recognition result;

S107: The edge computing voice processing module analyzes and processes the dietary information, medication information, emotional information, life speech information and voiceprint information respectively to generate personal health data.

The present invention has the following beneficial effects:

1. The projector with the speech analysis function provided by the present invention adds a speech analysis module and an edge computing speech processing module, which can collect the user's diet information, medication information, emotional information and life language information through the speech analysis module, and extract the voiceprint. Features, and perform edge computing on these data to analyze the user's emotional and psychological changes, as well as daily diet and medication conditions, provide data support for subsequent user health diagnosis, and reduce the rate of misdiagnosis by doctors.

2. The voice analysis module in the present invention has a voice control function, which can recognize user commands, thereby realizing voice control of the projector.

Description of drawings

1 is a schematic diagram of the internal circuit structure of a projector according to an embodiment of the present invention;

2 is a schematic diagram of a circuit structure of a speech analysis module according to an embodiment of the present invention;

3 is a schematic diagram of an internal circuit structure of a projector according to another embodiment of the present invention;

FIG. 4 is a schematic structural diagram of a power module circuit according to an embodiment of the present invention;

5 is a schematic structural diagram of an audio output module circuit according to an embodiment of the present invention;

6 is a schematic diagram of a circuit structure of an edge computing voice processing module according to an embodiment of the present invention;

7 is a flowchart of a method for generating personal health data according to an embodiment of the present invention;

1-central control module, 2-optical projection module, 3-audio output module, 4-power supply module, 5-edge computing voice processing module, 6-voice analysis module, 7-remote fingerprint switch module, 8-cooling module, 9 -Wireless module, 10-storage module, 301-audio output interface, 302-built-in speaker, 401-power control circuit, 402-lithium battery, 403-external power adapter interface, 501-high-efficiency buffer unit, 502-neural network acceleration Units, 503 - data visualization unit, 504 - abnormal network information feedback unit, 601 - recording unit, 602 - A/D conversion unit, 603 - denoising unit, 604 - voiceprint detection and extraction unit, 605 - voice command control unit.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

The present invention provides a projector with a speech analysis function, which includes a casing and a circuit arranged inside the casing. The schematic diagram of the internal circuit structure of an embodiment provided by the present invention is shown in FIG. 1 . The circuit structure inside the casing is shown in FIG. 1 . It includes a central control module 1 , an optical projection module 2 , an audio output module 3 , a power supply module 4 , an edge computing voice processing module 5 and a voice analysis module 6 . The central control module 1 is electrically connected to the optical projection module 2, the audio output module 3, the power supply module 4, the edge computing voice processing module 5, and the voice analysis module 6, respectively, and controls the normal operation of each module, such as controlling the on-off and on-off operations of the projector. System, mainly responsible for system control, data transfer, etc. The analysis and processing of the voice information is mainly completed by the edge computing voice analysis module 5 . The power supply module 4 is electrically connected to the central control module 1, the optical projection module 2, the audio output module 3, the speech analysis module 6 and the edge computing speech processing module 5, respectively, to supply power to each module and ensure the normal operation of each module.

Specifically, as shown in FIG. 2 , the speech analysis module 6 includes a recording unit 601 , an A/D conversion unit 602 , a denoising unit 603 , and a voiceprint detection and extraction unit 604 .

The recording unit 601 is used to collect the user's voice, and the recording unit 601 is electrically connected to the A/D conversion unit 602. Optionally, the recording unit 601 can be a microphone and a commercially available recording chip, such as a voice chip with models of WTR050 and WT2000, etc. , is used to collect or record the user's voice information. The user records oral personal daily diet, taking medicine, etc., and speech information in daily life. The A/D conversion unit 602 is an analog-to-digital converter. Optionally, a dual eight-bit high-speed ADC converter MXT2002 can be used to convert the analog voice signal into a digital voice signal that the projector can process later. The A/D conversion unit 602 is connected to The denoising unit 603 is electrically connected, and outputs the converted digital voice signal to the denoising unit 603 for denoising processing. After the denoising is completed, the denoised digital voice signal is respectively transmitted to the voiceprint detection and extraction unit 604 . After denoising by the denoising unit 603, the noise in the digital voice signal can be effectively eliminated, and a more accurate digital voice signal can be obtained, which is convenient for the voiceprint detection and extraction by the voiceprint detection and extraction unit 604 and the edge computing voice processing module 5. Analysis and calculation, reduce the difficulty of voiceprint detection, extraction and analysis and calculation, improve the accuracy of voiceprint detection, extraction, analysis and calculation, and ensure the sound effect of the recording. Optionally, the denoising unit 603 may be a commercially available filter.

The voiceprint detection and extraction unit 604 is electrically connected to the edge computing voice processing module 5, and performs voiceprint and sound wave detection on the input digital signal, extracts the voiceprint features, and is pre-stored in the voiceprint detection and extraction unit 604. The voice of the projector user. pattern for comparison. Optionally, the voiceprint detection and extraction unit 604 may also use a commercially available voiceprint recognition chip, such as a voiceprint recognition chip of SpeakIn. The voiceprint detection and extraction unit 604 has pre-stored the user's voiceprint. When the voiceprint detection and extraction unit 604 detects a voiceprint, it compares the current voiceprint with the pre-stored voiceprint to determine whether the current user is a projection. If the voiceprint of the current user matches the preset user's voiceprint successfully, the voiceprint detection and extraction unit 604 sends the extracted voiceprint feature and the received digital voice signal to the edge computing voice processing module 5 Perform processing calculations.

Optionally, the voiceprint detection and extraction unit 604 in the projector provided by the present invention can pre-store multiple user information, including voiceprint information, and the edge computing voice processing module 5 can also establish multiple user-independent personal information databases. For temporary caching of personal data. When the pre-stored user information of the projector is multiple, when the voiceprint detection and extraction unit 604 performs detection and extraction, it is necessary to detect which user the voiceprint feature belongs to, and combine the user identity information with the voiceprint and digital voice. The signal is sent to the edge computing voice processing module 5 together, and the edge computing voice processing module 5 can know which user's voice information is the voice information such as the voiceprint and digital voice signal currently analyzed and calculated, so as to combine the voice information and analysis calculation. The results are stored in the user's personal information database.

The edge computing speech processing module 5 can extract speech rate and speech frequency information, at the same time recognize semantics, analyze the user's wording, and analyze the change of voiceprint characteristics. Because human emotions will have a certain impact on speechprints, speech rate, speech frequency, etc. , there are differences in the voiceprint, speech rate, and speech frequency of the voice made when the user is angry or happy. The edge computing voice processing module 5 can determine the current user's mood and psychological state through the above analysis, and carry out the process within a certain period of time. Comprehensive analysis can also analyze the user's current emotional and psychological state fluctuations. And the edge computing voice processing module 5 can recognize semantics, so it can analyze the user's diet information and medication information and convert these information into text for analysis and calculation to obtain the user's diet status and medication status. Information such as emotion, psychology, diet, medication information and other information obtained by the edge computing voice processing module 5 can be used as personal health data to provide health data support for doctors during health diagnosis and reduce the misdiagnosis rate of doctors.

In this embodiment, an edge computing voice processing module 5 is added to the projector with the voice analysis function. Optionally, the edge computing voice processing module 5 can be the XForce edge AI computing platform of Horizon Company, TPU of Google, Cambrian DianNao series, Eyeriss NPU, Intel Xeon D-2100 processor and other AI chips. The edge computing voice processing module 5 can also be a GPU with edge computing function. The GPU is a graphics processor. The current GPU is no longer limited to 3D graphics processing. The development of GPU general computing technology has attracted a lot of attention in the industry. It also proves that in some calculations such as floating-point operations and parallel computing, GPUs can provide dozens or even hundreds of times the performance of CPUs. Adding chips with edge computing functions to current GPUs can quickly process big data.

The edge computing voice processing module 5 can realize short-term storage of the above-mentioned voiceprint, diet, medication and other voice information, so it can realize the short-term calculation and analysis of the voice information on the local projector with the voice analysis function, without the need to store the voice information. Transfer to the cloud for analysis in the cloud. The addition of the edge computing voice processing module 5 can produce the following beneficial effects: 1. A large number of computing tasks can be processed at the data collection end without being transmitted to the cloud, which greatly relieves the transmission pressure of the network and also reduces the cost of data center Pressure; 2. The characteristics of distributed edge computing and its proximity to the device end doom it to its advantages in real-time processing, so it can better support real-time processing and execution of local services. The task is processed at the edge. Therefore, the time required for data transmission in the network Sharply reduced, the user will get a faster response time; 3. Since the user's voiceprint features extracted by the projector are very private data, if these personal privacy data are uploaded to the cloud center for calculation and processing, there will be certain security risks , adding an edge computing voice processing module 5 to the projector can better protect private data.

In another embodiment of the present invention, optionally, as shown in FIG. 2 , the voice analysis module further includes a voice command control unit 605, and the voice command control unit 605 is configured to generate a command control signal according to the user's voice command and send it to the central Control module 1, the central control module 1 controls the power on and off of the projector and the operation of the internal system of the projector according to the command control signal; the voice command control unit 605 is electrically connected with the denoising unit 603 and the voiceprint detection and extraction unit 604. The extraction unit 604 detects that the currently extracted voiceprint is consistent with the pre-stored user voiceprint, which proves that the current user is a user of the projector and can perform subsequent operations. The switch or system operation command identified in the received digital voice signal generates the corresponding command control signal, and sends the command control signal to the central control module 1, and the central control module 1 controls the projector to switch on and off or the system according to the command control signal. run.

In another embodiment of the present invention, as shown in FIG. 3 , the circuit further includes a remote fingerprint switch module 7, which is electrically connected to the central control module 1 or wirelessly connected to the central control module, The user controls the projector to turn on through the remote fingerprint switch module 7, and at the same time, a white screen appears on the optical projection module. Specifically, the remote fingerprint switch module 7 is mainly used to control the startup of the projector with the voice analysis function, and the power is turned on by pressing the switch, and the user's fingerprint information is input by pressing at the same time. After the fingerprint information is successfully matched, the projector is turned on; Yes, for the convenience of use, the user can input the power-on voice command through the voice analysis module to control the projector to turn on, and at the same time, a white screen appears on the optical projection module. The specific principle of voice control has already been mentioned above, and will not be repeated here.

In yet another embodiment of the present invention, as shown in FIG. 3 , the circuit further includes a heat dissipation module 8 , which is electrically connected to the central control module 1 and used to dissipate heat from the projector in operation. Specifically, during the operation of the projector, especially when a large amount of data is transmitted and calculated, the projector will inevitably cause the temperature of the machine to rise rapidly due to high-speed operation. In order to better protect the various parts of the projector from burning out and ensure For the high-speed operation of the projector, a heat dissipation module 8 needs to be installed to dissipate heat from the projector or main components in operation. The heat dissipation module 8 may use a relatively mature heat dissipation plate or the like, which is not specifically limited in the present invention.

In yet another embodiment, as shown in FIG. 4 , the power module 4 includes a power control circuit 401 , a lithium battery 402 and an external power adapter interface 403 , and the lithium battery 402 and the external power adapter interface 403 are powered by the power control circuit 401 switching and input. After the machine is started, each voltage regulator chip in the power control circuit 401 provides the required power supply voltage for each module inside the projector, and makes each functional module and each drive circuit start to work normally. At the same time, the power control circuit 401 determines whether to perform a charging operation according to the power of the lithium battery 402. When the external power supply is not used, the lithium battery 402 is discharged to supply power to each module inside the projector.

In yet another embodiment, as shown in FIG. 5 , the audio output module 3 includes an audio output interface 301 and a built-in speaker 302 . The built-in speaker 302 plays sound through the speaker located on the lower surface of the casing, and the audio output interface 301 is electrically connected to external speakers, earphones and other devices.

In yet another embodiment, as shown in FIG. 3 , the circuit in the housing further includes a wireless module 9 , and the wireless module 9 is electrically connected to the central control module 1 . Optionally, the wireless module 9 includes a wireless router and an infrared remote control signal receiver. With the Bluetooth device, the projector can interact with other smart devices through the wireless module 9, for example, it can interact with other health detection equipment, collect the human physiological data information measured by these health detection equipment, and pass the central control. Module 1 is sent to edge computing voice processing module 5 for processing and calculation. The wireless module 9 can also output the result of calculation and analysis by the edge computing voice processing module to other smart devices of the user, so as to facilitate the user to view the analysis result.

In another embodiment of the present invention, as shown in FIG. 3 , the circuit further includes a storage module 10 , and the storage module 10 is electrically connected to the central control module 1 and the edge computing speech analysis module 5 . The storage module 10 has a large-capacity memory, which can store the personal health data generated by the edge computing voice processing module 5 for a long time. Although the edge computing voice processing module 5 can also temporarily cache the personal health data, due to the amount of personal health data. The edge computing voice processing module 5 cannot store these data for a long time, and can only store data within a certain period of time, so that only short-term analysis can be performed during analysis and processing. The long-term monitoring and tracking of personal health data provides support for the long-term comprehensive calculation and analysis of data by the edge computing voice processing module 5. The result of the analysis and calculation performed by the edge computing voice processing module 5 is stored in the storage module 10 for storage, which relieves the storage pressure of the edge computing voice processing module 5 . With the increase of the number of times and the time of use of the user, the amount of personal health data stored in the storage module 10 is increasing, which in turn provides sufficient data support for subsequent edge computing and the like. In addition, the storage module 10 of the projector of the present invention can be classified and stored according to the type of personal health data actually stored, such as mood change database, psychological state change database, medication situation change database, diet situation change database and so on.

Further, on the basis of the above embodiments of the present invention, as shown in FIG. 6 , the edge computing voice processing module 5 includes a high-efficiency cache unit 501, a neural network acceleration unit 502, a data visualization unit 503, and an abnormal network information feedback unit 504, wherein :

The neural network acceleration unit 502 is configured to analyze and process the speech information delivered by the speech analysis module 6 in real time to generate personal health data and send it to the high-efficiency caching unit 501 and the storage module 10 for caching. Specifically, the neural network acceleration unit 502 first analyzes and calculates the voice information received within a certain period of time: voiceprint, diet information, medication information, life language information, etc., and obtains the user's mood, psychology, diet, medication and other data for transmission. Perform visual processing on the data visualization unit 503 to generate textual or graphical conclusions that can be viewed by the user, and store the generated textual or graphical conclusions in the storage module 10; , diet, medication and other data are sent to the high-efficiency cache unit 501 and the storage module 10 for storage.

Every once in a while, the neural network acceleration unit 502 will extract the emotions, psychology, diet, medication and other data stored in the high-efficiency cache unit 501, and then conduct a comprehensive analysis to generate the user's mood, psychology, diet, medication and other fluctuations during this period of time, After the extraction, the data stored in the high-efficiency cache unit 501 will be cleared, so that the high-efficiency cache unit 501 can continue to store the newly generated data. The generated fluctuations in mood, psychology, diet, medication, etc. are sent again to the high-efficiency cache unit 501 and the storage module 10 for storage. Here, the time can be set to perform comprehensive processing and analysis every 5 days, 10 days, half a month, etc. The specific time setting can be determined according to the actual situation, which is not limited in the present invention.

The data visualization unit 503 is used to visualize the personal health data generated by the neural network acceleration unit 502 (including data on emotions, psychology, diet, medication, etc., as well as fluctuations in emotions, psychology, diet, medication, etc.), and transmit it to the data visualization unit 503. The abnormal network information feedback unit 504 performs feedback and transmits it to the storage module 10 for storage. At the same time, if the neural network acceleration unit 502 has voice information that cannot be analyzed and processed, it can also be sent to the data visualization unit 503 for visualization processing and transmitted to the abnormal network information feedback unit 504 for feedback.

The abnormal network information feedback unit 504 transmits the received visualized data to the cloud platform through the wireless module or communication protocol to a third party, or displays it to the user himself through the connected optical projection module 2 . In the projector of the present invention, any data to be transmitted for viewing by a user or a third party must be transmitted to the abnormal network information feedback unit 504 for feedback.

In an embodiment of the present invention, as shown in FIG. 7 , a method for generating health data based on the above-mentioned projector with a speech analysis function is also provided, the method comprising:

S101: The projector is turned on by voice or switch;

S102: The voice analysis module starts to input user voice information;

S103: Convert the entered user voice information into digital voice signals and extract voiceprints;

S104: Voiceprint matching;

Specifically, it is judged whether the extracted voiceprint matches the pre-stored voiceprint, if so, go to step S105, otherwise, end;

S105: Send the digital voice signal and voiceprint to the edge computing voice processing module;

S106: Perform semantic recognition to classify the received digital speech signal;

Specifically, the edge computing voice processing module performs semantic recognition on the received digital voice signal and divides the received digital voice signal into diet information, medication information and life language information according to the semantic recognition result.

S107: Analyze and process the classified digital voice signal and voiceprint information to generate personal health data.

Specifically, the edge computing voice processing module analyzes and processes dietary information, medication information, daily speech information and voiceprint information to generate personal health data. Using the health data generating method provided by the present invention to generate personal health data has the advantages of simple operation and strong practicability, and the generated personal health data can provide strong data support for the user's physical health analysis.

The projector with speech analysis function provided by the present invention, in addition to having the functions of projection, playing video, pictures, music, etc., which are possessed by ordinary projectors, by adding a speech analysis module and an edge computing speech processing module on the projector, also It can collect and extract the user's physical and physiological information such as voiceprint information and record the user's daily life information such as daily diet, medication, life language information, and provide data support for the user's health diagnosis. The user's physiological information includes the user's voiceprint information. , through the feature extraction and calculation of the voiceprint information, the user's mood and psychological state can also be obtained. Since a person's voiceprint will change slightly with the person's mood, it is possible to understand the person's psychological and emotional state through the voiceprint. , the user's daily life information includes the user's diet, the daily medicines the user takes, life language information, etc. By recording and calculating the user's diet, the user's diet rule is obtained. When the doctor makes a health diagnosis for the user, you can Provide the recent user's diet and daily medicines, so that doctors can make a correct health diagnosis for the user based on the above detailed personal data information, and reduce the misdiagnosis rate.

The present invention has been further described above with the help of specific embodiments, but it should be understood that the specific description herein should not be construed as a limitation on the spirit and scope of the present invention. Various modifications made in the embodiments all belong to the protection scope of the present invention.

Claims (7)

1. A projector with a voice analysis function, wherein the projector comprises: a casing and a circuit arranged inside the casing, the circuit comprising a central control module, an optical projection module, and an audio output module , power module, storage module, wireless module, edge computing voice processing module and voice analysis module; the central control module is respectively connected with the optical projection module, the audio output module, the power module, the wireless module, the the edge computing voice processing module and the voice analysis module are electrically connected; the storage module is respectively electrically connected with the central control module and the edge computing voice processing module; The voice analysis module includes a recording unit, an A/D conversion unit, a denoising unit, and a voiceprint detection and extraction unit that are electrically connected in sequence; the recording unit is used to record the user's voice information and send the voice information to the The A/D conversion unit converts the digital voice signal into a digital voice signal, and the denoising unit denoises the digital voice signal and transmits it to the voiceprint detection and extraction unit; The voiceprint detection and extraction unit receives the digital voice signal and extracts the voiceprint, and compares the voiceprint with the pre-stored user voiceprint. The voiceprint, the user identity information of the user to which the voiceprint belongs, and the digital voice signal are sent to the edge computing voice processing module; The edge computing voice processing module includes a high-efficiency cache unit, a neural network acceleration unit, a data visualization unit, and an abnormal network information feedback unit, and the neural network acceleration unit is used to analyze and process the voice information transmitted by the voice analysis module to generate personal health data. And send it to the high-efficiency cache unit and the storage module for caching; the voice information received by the neural network acceleration unit at least includes: voiceprint, diet information, medication information and life voice information; The data visualization unit is used to perform data visualization processing on the personal health data generated by the neural network acceleration unit, and transmit it to the abnormal network information feedback unit for feedback and to the storage module for storage; at the same time, the neural network acceleration unit Send the voice information that cannot be analyzed and processed to the data visualization unit for visualization processing and transmit it to the abnormal network information feedback unit for feedback, and the abnormal network information feedback unit sends the received visualized data to the cloud platform through the wireless module. third parties. 2. The projector with a voice analysis function according to claim 1, wherein the voice analysis module further comprises a voice command control unit, and the voice command control unit is used to generate a command control signal according to a user's voice command Send it to the central control module, and the central control module controls the on-off of the projector and the operation of the internal system of the projector according to the instruction control signal; The voice command control unit is electrically connected with the denoising unit and the voiceprint detection and extraction unit. 3. The projector with speech analysis function according to claim 1, wherein the circuit further comprises a remote fingerprint switch module, the remote fingerprint switch module is electrically connected with the central control module or the remote fingerprint switch The switch module is wirelessly connected with the central control module, the user controls the projector to start up through the remote fingerprint switch module, and at the same time, a white screen appears on the optical projection module. 4 . The projector with speech analysis function according to claim 1 , wherein the circuit further comprises a heat dissipation module, the heat dissipation module is electrically connected to the central control module, and is used for the operation of the projector. 5 . instrument to dissipate heat. 5. The projector with speech analysis function according to claim 1, wherein the power supply module comprises a power supply control circuit, a lithium battery and an external power supply adapter interface, and the lithium battery and the external power supply adapter interface pass through. The power control circuit performs switching and input of power. 6. The projector with speech analysis function according to claim 1, wherein the audio output module comprises a built-in speaker and an audio output interface. 7. A method for generating personal health data based on the projector with speech analysis function according to any one of claims 1-6, wherein the method comprises: S101: The projector is turned on by voice or switch; S102: The voice analysis module starts to input user voice information; S103: Convert the user voice information entered in step S102 into a digital voice signal and extract voiceprints; S104: determine whether the extracted voiceprint matches the pre-stored voiceprint, if so, perform step S105, otherwise, end; S105: Send the digital voice signal and voiceprint to the edge computing voice processing module; S106: The edge computing voice processing module performs semantic recognition on the received digital voice signal and divides the received digital voice signal into diet information, medication information and life language information according to the semantic recognition result; S107: The edge computing voice processing module analyzes and processes the diet information, medication information, life language information and voiceprint information respectively to generate personal health data.

Images