A multifunctional intelligent window intercom system

Abstract

The present invention discloses a multifunctional intelligent window intercom system, including an external machine body and an internal machine body, an external machine microphone, an external machine speaker, a first voice digital codec module and a first audio power amplifier module are installed in the external machine body, an audio switch circuit with an MCU, a second voice digital codec module and a second audio power amplifier module is installed in the internal machine body, an internal machine microphone and an internal machine speaker are installed in the internal machine body, and also include a first AI voice processing unit, a second AI voice processing unit, a network AI intelligent translation server, a network cloud voice intercom server and a 4G wireless data transmission module, and the network AI intelligent translation server is connected to the MCU data bidirectional transmission through the 4G wireless data transmission module, and the control end of the MCU is electrically connected with a voice sending control button. The multifunctional intelligent window intercom system of the present invention is not only easy to use, but also can clearly transmit sound inside and outside the counter, and can also perform Chinese-English translation, reduce labor intensity, and improve work efficiency.

Description

Technical Field

[0001] The invention relates to the field of voice intercom equipment, and in particular to a multifunctional intelligent window intercom system. Background Art

[0002] At present, the intercoms used at counters such as bank windows, bus ticket windows, and train station ticket windows (i.e., ordinary intercoms used in various public places) generally have an external unit fixed on the window glass and transmitting sound to customers outside the window (i.e., the external unit has a speaker and a microphone) and an intercom with a straw that is located inside the window and fixed on the office seat. When the counter staff is talking, they need to press the intercom call button, and the external unit and the intercom are connected, and at the same time, they need to aim the straw at the mouth of the intercom staff with the straw, so that the external unit can clearly transmit sound to the customer outside the window. However, whether it is a bank window, a bus ticket window, a train station ticket window, or a window for urban administrative services, a large number of people are received every day. Relying on the existing window intercom for voice communication, accurate straw intercom and operation each time are time-consuming and laborious, which makes the staff exhausted after a day. When the staff's mouth is away from the straw, the sound becomes smaller, causing the voice microphone inside the straw to vibrate weakly and the converted electrical signal to be weak, resulting in the voice of customers outside the window being unclear or too soft to be heard clearly, which creates certain obstacles to normal voice communication.

[0003] In addition, the existing intercom machines only have the function of transmitting sound, but not the function of reducing noise. Whether it is a bank window, a bus ticket window, a train station ticket window or a window for urban administrative services, they are all public places with relatively noisy environments, which leads to the shortcomings of poor communication or poor experience of the existing intercom machines. Furthermore, if there are overseas customers at the counter, the existing intercom machines with straws do not have the function of automatic translation between Chinese and English, which leads to poor communication and affects business processing.

[0004] In view of this, the inventor of this case conducted in-depth research on the above-mentioned issues, which led to the emergence of this case. Summary of the invention

[0005] The purpose of the present invention is to provide a multifunctional intelligent window intercom system, which is not only simple in structure and easy to use, but also can clearly transmit sound inside and outside the counter, and can also perform Chinese-English translation, thereby reducing labor intensity, improving work efficiency and reducing costs.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] A multifunctional intelligent window intercom system comprises an external machine body arranged outside the counter window glass and an internal machine body arranged inside the counter window glass, the external machine body having an external machine shell, the internal machine body having an internal machine shell, the external machine shell having an external machine sound inlet hole and an external machine sound outlet hole, an external machine microphone and an external machine speaker are installed correspondingly to the positions of the external machine sound inlet hole and the external machine sound outlet hole in the external machine body, a first voice digital codec module and a first audio power amplifier module are installed in the external machine body, an audio switching circuit having an MCU, a second voice digital codec module and a second audio power amplifier module is installed in the internal machine body, the internal machine shell having an internal machine sound inlet hole and an internal machine sound outlet hole, an internal machine microphone and an internal machine speaker are installed correspondingly to the positions of the internal machine sound inlet hole and the internal machine sound outlet hole in the internal machine body, and also comprises a first AI voice processing unit and a second AI voice processing unit, a network AI intelligent translation server, and a 4G wireless data transmission module; the network AI intelligent translation server transmits data via the 4G wireless data transmission module The block is connected to the MCU for bidirectional data transmission, the external microphone is connected to the input end of the first AI voice processing unit for data transmission, the output end of the first AI voice processing unit is connected to the input end of the first voice digital codec module, the output end of the first voice digital codec module is connected to the corresponding input end of the MCU, the control end of the MCU is electrically connected with a voice sending control button, the input end of the first audio power amplifier module is connected to the corresponding output end of the MCU, and the output end of the first audio power amplifier module is connected to the external speaker; the internal microphone is connected to the input end of the second AI voice processing unit for data transmission, the output end of the second AI voice processing unit is connected to the input end of the second voice digital codec module, the output end of the second voice digital codec is connected to the corresponding input end of the MCU, the input end of the second audio power amplifier module is connected to the corresponding output end of the MCU, and the output end of the second audio power amplifier module is connected to the internal speaker.

[0008] The outer casing includes a first upper cover, a first bottom shell, a shock-absorbing pad and a support frame, the support frame having a vertical section and a horizontal section fixedly connected to the lower end of the vertical section, the first bottom shell is located in front of the vertical end, the shock-absorbing pad is located between the first bottom shell and the vertical section, a pit is recessed on the first bottom shell for accommodating the shock-absorbing pad, the rear side of the shock-absorbing pad is located outside the pit, and the first bottom shell is fixed to the vertical section by screws; the first upper cover is located in front of the first bottom shell, a groove is provided on the inner wall of the first upper cover corresponding to the edge portion of the first bottom shell, and a sealing ring is embedded in the groove.

[0009] The first upper cover is provided with a mounting hole, an alarm switch is mounted on the first upper cover through the mounting hole, the external sound inlet and the external sound outlet are both located on the first upper cover, and the external microphone, the external speaker and the alarm switch are arranged from top to bottom on the first upper cover.

[0010] It also includes a network cloud voice intercom server, a digital intercom module is installed in the inner machine body, the digital intercom module is connected with the MCU for two-way data transmission, and the digital intercom module is communicated with the network cloud voice intercom server through a 4G wireless data transmission module; the control end of the MCU is electrically connected with an intercom voice sending control button.

[0011] A power supply module is also provided in the inner machine body. The power supply module is located on the left side of the audio switch circuit. The power supply module is connected to the audio switch circuit for power supply via a wire. The incoming call detection circuit is located between the power supply module and the audio switch circuit.

[0012] A telephone module with display and dialing functions is also installed in the inner shell. The telephone module has a display screen and dialing buttons. The inner shell is provided with a hole structure for exposing the display screen and the dialing buttons outside the inner shell. The display screen is electrically connected to the corresponding output end of the MCU, and the dialing buttons are electrically connected to the corresponding input end of the MCU. A telephone interface is installed on the inner body, and the MCU is connected to the telephone interface.

[0013] The inner machine housing comprises a second upper cover and a second bottom shell, the audio switch circuit is fixed in the second bottom shell, the dial module is fixedly mounted on the second bottom shell, the second bottom shell is provided with a first through hole, the second upper cover is provided with a second through hole, the second through hole is located in front of the first through hole, and the first through hole and the second through hole are the above-mentioned hole structures; a transparent plate is mounted on the second upper cover, the inner machine sound outlet hole is located on the second upper cover and is located on the left side of the transparent plate and the second through hole, the inner machine sound inlet hole is located on the front edge of the second upper cover, the second upper cover is provided with a first knob switch and a second knob switch on the front side of the second through hole, the first knob switch and the second knob switch are arranged side by side on the second upper cover, the voice sending control button is located on the second upper cover and is located between the first knob switch and the second knob switch, the intercom voice sending control button is located on the second upper cover and is located between the first knob switch and the second knob switch, and the intercom voice sending control button is located on the left side of the voice sending control button.

[0014] The second bottom shell has an upper port, a waterproof rib is integrally formed on the upper port of the second bottom shell, a positioning step is provided on the upper edge of the second upper cover, a plurality of first reinforcing ribs are arranged on the inner wall of the second upper cover, second reinforcing ribs are arranged on the inner wall of the second bottom shell corresponding to the first reinforcing ribs, and a notch is provided on the first reinforcing rib.

[0015] A base is provided at the bottom of the inner casing, and two mortise grooves are provided on the base, and the two mortise grooves are located side by side on the top surface of the base and are respectively close to the left and right side edges of the base; two fixing rods are provided on the bottom of the inner casing, and the upper ends of the two fixing rods are respectively hinged at the two corners of the rear part of the inner casing, and two telescopic rods are provided on the bottom of the inner casing; the upper ends of the two telescopic rods are hinged at the corners of the front part of the inner casing, and the lower ends of the telescopic rods and the lower ends of the fixing rods both have a tenon structure matched with the mortise groove, and the length of the fixing rod is less than the length of the telescopic rod; a limit screw is fixedly installed on the groove wall of the mortise groove, and the telescopic rod has an inner section rod and an outer section rod, and a limit piece is provided on the outer section rod.

[0016] The beneficial effects of the present invention are:

[0017] (1) The indoor microphone and the indoor speaker are directly fixed in the indoor body. There is no need to use a straw or adjust the straw for conversation. The conversation is more convenient, which effectively reduces the labor intensity. The structure is simpler and more compact, which is also conducive to reducing space occupancy. The first AI voice processing unit and the second AI voice processing unit are installed on the audio switch circuit. The first AI voice processing unit and the second AI voice processing unit can actively identify the sound intensity, echo cancellation and noise suppression, so that there is no need to adjust the straw to ensure clear call transmission. When working, the first AI voice processing unit identifies the sound intensity of the external microphone to ensure that the voice is clearly transmitted to the first voice digital codec module, and is processed by the MCU on the audio switch circuit. After the MCU is processed, it is transmitted to the second audio power amplifier module to amplify the volume, and at the same time, it is emitted by the indoor speaker, so that the external body can clearly transmit the sound to the counter window. Similarly, the second AI voice processing unit identifies the sound intensity of the sound transmitted by the indoor microphone to ensure that the voice is clearly transmitted to the second voice digital codec module, and is processed by the MCU on the audio switch circuit. After processing by the MCU, it is transmitted to the first audio amplifier module to amplify the volume, and at the same time, it is emitted by the external speaker, so that the indoor unit body can clearly transmit the sound to the outside of the counter window. The setting is reasonable and the structure is simple, and it is easy to use. There is no need to use a straw or adjust the straw to make calls, so calls and communication are more convenient, more convenient, greatly reducing staff fatigue, and are cleaner and more hygienic. At the same time, it solves the problem that when using an intercom on the existing counter, the straw needs to be aimed at the mouth of the intercom staff with the straw to achieve clear sound transmission between the external unit and the customers outside the window.

[0018] (2) Since the system has a specific network AI intelligent translation server and a 4G wireless data transmission module; the network AI intelligent translation server is connected to the MCU through the 4G wireless data transmission module for two-way data transmission. The control end of the MCU is electrically connected to a voice sending control button. The voice sending control button controls the MCU to receive the data of the first voice digital codec module and send it out through the 4G wireless data transmission module, so that the translation can be controlled. When working, the staff in the counter presses the voice sending control button, and the 4G wireless data transmission module sends the data of the first voice digital codec module to the network AI intelligent translation server through the base station or satellite using the 4G wireless data transmission module. The network AI intelligent translation server translates the voice digital information and connects it to the 4G wireless data transmission module through the base station or satellite to transmit it back. The MCU receives the translated voice digital information and transmits it to the first audio power amplifier module or the second audio power amplifier module to amplify the volume. At the same time, it is emitted by the corresponding external speaker and internal speaker, realizing the Chinese-English automatic translation function, solving the problem that the counter has poor communication with overseas customers and affects business handling. Therefore, the multifunctional intelligent window intercom system of the present invention is not only simple in structure and easy to use, but also can clearly transmit sound inside and outside the counter, and can perform Chinese-English translation, thereby reducing labor intensity, improving work efficiency and reducing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] For ease of explanation, the present invention is described in detail with reference to the following specific embodiments and the accompanying drawings.

[0020] Figure 1 This is a principle block diagram of the multifunctional intelligent window intercom system of the present invention (the digital intercom module, etc. are omitted in the figure).

[0021] Figure 2 It is a schematic diagram of the structure of the external machine body in the present invention.

[0022] Figure 3 It is a cross-sectional view of the outer machine body in the present invention (i.e., a half cross-sectional view of the outer machine body).

[0023] Figure 4 It is a schematic diagram of the structure of the inner machine body in the present invention.

[0024] Figure 5 It is an exploded view of the inner machine body in the present invention.

[0025] Figure 6 It is a basic principle block diagram of the noise suppression algorithm in the present invention (i.e., a principle diagram of noise suppression of the first AI speech processing unit or the second AI speech processing unit).

[0026] Figure 7 The figure is a basic principle diagram of the intercom function of the multifunctional intelligent window intercom system of the present invention.

[0027] Figure 8 It is a schematic diagram of the structure of the inner machine body connected to the base in the present invention (a part of the inner machine body is shown in the figure). DETAILED DESCRIPTION

[0028] In order to further explain the technical solution of the present invention, the present invention is described in detail below through specific embodiments.

[0029] like Figure 1-8 As shown, the multifunctional intelligent window intercom system of the present invention comprises an external machine body 10 arranged outside the counter window glass and an internal machine body 20 arranged inside the counter window glass, the external machine body 10 has an external machine shell, the internal machine body 20 has an internal machine shell, the external machine shell has an external machine sound inlet hole 11 and an external machine sound outlet hole 12, an external machine microphone 13 and an external machine speaker 14 are installed in the external machine body 10 corresponding to the positions of the external machine sound inlet hole 11 and the external machine sound outlet hole 12, a first voice digital codec module 03 and a first audio power amplifier module 04 are installed in the external machine body 10, an audio switch circuit for controlling the wired call connection between the internal machine body 20 and the external machine body 10 is installed in the internal machine body 20, and the audio switch circuit has an MCU02, a second voice digital codec module 05 and a second audio power amplifier module 06. The inner machine shell is provided with an inner machine sound inlet hole 21 and an inner machine sound outlet hole 22, and an inner machine microphone 23 and an inner machine speaker 24 are installed in the inner machine body 20 at positions corresponding to the inner machine sound inlet hole 21 and the inner machine sound outlet hole 22; specifically, the audio switch circuit, MCU02, the first voice digital codec module 03, the first audio power amplifier module 04, the second voice digital codec module 05, the second audio power amplifier module 06, the external machine microphone 13, the external machine speaker 14, the inner machine microphone 23 and the inner machine speaker 24 are all existing ones, and are also connected in a conventional manner in the field, and are all controlled by MCU02 to realize the call connection between the inner machine body 20 and the external machine body 10.

[0030] The multifunctional intelligent window intercom system of the present invention also includes a first AI voice processing unit 07 and a second AI voice processing unit 08 for sound intensity recognition, echo elimination and noise suppression, a network AI intelligent translation server for Chinese-English translation (the network AI intelligent translation server adopts the existing one, that is, the existing AI voice translation host), and a 4G wireless data transmission module 09 for wireless communication connection with the network AI intelligent translation server; the network AI intelligent translation server is connected to the MCU02 for data bidirectional transmission through the 4G wireless data transmission module 09, and the external microphone 13 is connected to the first AI voice processing unit 07. The input end of the first AI voice processing unit 07 is connected to the input end of the first voice digital codec module 03, the output end of the first voice digital codec is connected to the corresponding input end of the MCU02, the input end of the first audio power amplifier module 04 is connected to the corresponding output end of the MCU02, the output end of the first audio power amplifier module 04 is connected to the external speaker 14, and the control end of the MCU02 is electrically connected with a voice sending control button 25 for controlling the MCU02 to receive the data of the first voice digital codec module and send it out using the 4G wireless data transmission module 09. The voice sending control button 25 here uses an existing ordinary button switch; the internal microphone 23 is connected to the input end of the second AI voice processing unit 08 for data transmission, the output end of the second AI voice processing unit 08 is connected to the input end of the second voice digital codec module 05, the output end of the second voice digital codec module 05 is connected to the corresponding input end of the MCU02, the input end of the second audio power amplifier module 06 is connected to the corresponding output end of the MCU02, and the output end of the second audio power amplifier module 06 is connected to the internal speaker 24. The above-mentioned structural setting can solve the problem that traditional counter intercoms need to use straws and adjust the straws to align with the mouths of staff members in order to achieve clear and crisp conversations. It can realize conversations without using straws and adjusting the straws to align with the mouths of staff members. Therefore, no operation is required and no adjustment is required for different staff members. Conversations are more convenient and labor intensity is effectively reduced.

[0031] Compared with the traditional straw intercom on the counter, the present invention can make calls without using straw and adjusting straw, which makes calls more convenient, effectively reduces labor intensity, is simpler and more compact in structure, and is also conducive to reducing space occupancy. The first AI voice processing unit 07 and the second AI voice processing unit 08 for active language recognition (i.e., for sound intensity recognition), echo cancellation and noise suppression are installed on the audio switch circuit, so that there is no need to adjust the straw to ensure clear call transmission. During operation, the first AI voice processing unit 07 recognizes the sound transmission of the external microphone 13 to ensure that the voice is clearly transmitted to the first voice digital codec module 03, and is processed by the MCU02 on the audio switch circuit. After processing by MCU02, it is transmitted to the second audio power amplifier module 06 to amplify the volume, and at the same time, it is emitted by the internal speaker 24, so that the external machine body 10 can clearly transmit the sound to the counter window. Similarly, the second AI voice processing unit 08 identifies the sound transmission of the internal microphone 23 to ensure that the voice is clearly transmitted to the second voice digital codec module 05, and is processed by the MCU02 on the audio switch circuit. After processing by MCU02, it is transmitted to the first audio power amplifier module 04 to amplify the volume, and at the same time, it is emitted by the external speaker 14, so that the internal machine body 20 can clearly transmit the sound to the outside of the counter window. The setting is reasonable and the structure is simple, and it is easy to use. It is realized that there is no need to use straws and adjust straws for calls, so calls and communication are more convenient, more convenient, greatly reducing staff fatigue, and are also cleaner and more hygienic. At the same time, it solves the problem that when using an intercom on the existing counter, the straw needs to be aimed at the mouth of the intercom staff with the straw to achieve clear sound transmission from the external machine to the customer outside the window. It also solves the problem that clear and clear voice dialogue cannot be achieved when using an intercom on the existing counter. The first AI voice processing unit 07 and the second AI voice processing unit 08, as well as the arrangement of the internal sound inlet 21 and the internal sound outlet 22, the internal microphone 23 and the internal speaker 24, solve the problem that traditional counter intercoms need to use straws and adjust the straws to aim at the mouths of staff to achieve clear and crisp conversations. The solution is to create a straw-free environment where calls can be made directly, which is not only more hygienic, but also reduces the chance of virus infection and improves the epidemic prevention effect. It is both safe and practical, and is more beneficial to the health of counter staff.

[0032] Specifically, the 4G wireless data transmission module 09 is installed on the audio switch circuit, and the 4G wireless data transmission module 09 is wirelessly connected to the network AI intelligent translation server (specifically, the network AI intelligent translation server can use the existing Google cloud voice recognition service) through the base station and satellite. The 4G wireless data transmission module 09 here adopts the existing 4G module and the existing 5G network transmission module. The 4G wireless data transmission module 09 is bidirectionally connected to the MCU02. During normal operation, the sound received by the external microphone 13 is converted and compressed by the first voice digital codec module. When the staff in the counter triggers the voice sending control button 25, the data of the first voice digital codec module (referring to the data converted and compressed by the first voice digital codec module received by the external microphone 13) is sent out using the 4G wireless data transmission module 09, and is received and translated by the network AI intelligent translation server through the 4G wireless data transmission module 09 for bidirectional transmission (that is, the translated data is returned), and the data transmitted to the second voice digital codec module by the control MCU02 is decompressed and decoded, and finally broadcast by the internal speaker 24. On the contrary, the speaker translates Chinese into English, receives the sound through the internal microphone 23, converts and compresses it through the second voice digital codec module, and finally broadcasts the translated English through the external speaker 24. The automatic Chinese-English translation function is realized to solve the problem of poor communication between overseas customers when handling counter business.

[0033] The multifunctional intelligent window intercom system of the present invention also includes a network cloud voice intercom server for intercom with an intercom, and a digital intercom module (not shown in the figure) for wireless intercom with an intercom connected to the network cloud voice intercom server is installed in the inner machine body 20. The digital intercom module here adopts the existing one in the art, and the digital intercom module is connected to the MCU for data bidirectional transmission. The digital intercom module is connected to the network cloud voice intercom server through the 4G wireless data transmission module 09; the control end of the MCU is electrically connected to the intercom voice sending control button 26 for controlling the MCU02 to receive the data of the second voice digital codec module and send it out using the 4G wireless data transmission module 09. Here, the intercom voice sending control button 26 uses the existing ordinary push switch. Specifically, the 4G wireless data transmission module 09 is connected to the network cloud voice intercom server through a base station and a satellite. With the above-mentioned structure, when in use, when the staff in the counter touches the intercom voice sending control button 26, MCU02 receives the data of the second voice digital codec module and sends it out using the 4G wireless data transmission module 09, and transmits it to the network cloud voice intercom server. The information is received by other intercoms that are simultaneously connected to the network cloud voice intercom server by wireless communication. Other intercoms also send the intercom voice to the network cloud voice intercom server through wireless communication. At this time, the digital intercom module in the internal machine body 20 of the 4G wireless data transmission module 09 receives and transmits it to MCU02. The data transmitted to the second voice digital codec module by MCU02 is decoded and played through the internal machine speaker, thereby realizing intercom by wireless communication.

[0034] It is further explained that the first AI voice processing unit 07 and the second AI voice processing unit 08 both use existing DSP chips. It is known that the DSP chip is a fast and powerful microprocessor, which is unique in that it can process data in real time. The DSP chip uses a Harvard structure with separate programs and data, and has a dedicated hardware multiplier, which can be used to quickly implement various digital signal processing algorithms. More specifically, the first AI voice processing unit 07 and the second AI voice processing unit 08 both use self-developed AI chips (i.e., custom-developed AI chips with model NF008), or use the AI ​​voice chips launched by the existing American startup Syntiant, or use the existing Soul Core II A (i.e., the domestic Soul Core II A chip, which can complete hundreds of billions of floating-point operations in one second, and the single-core performance exceeds 4 times that of similar processing cores in the current international market, and it can be directly interconnected with high-speed ADC and DAC). It is known that ADC is the abbreviation of analog-to-digital converter, which refers to analog / digital converter. ADC is a circuit or device that converts analog input signals into digital signals. DAC is the abbreviation of digital-to-analog converter, which refers to digital / analog converter. Digital quantity is represented by codes combined in digits. For weighted codes, each bit of code has a certain bit weight. In order to convert digital quantity into analog quantity, each bit of code must be converted into a corresponding analog quantity according to its bit weight, and then these analog quantities are added together to obtain a total analog quantity proportional to the digital quantity, thereby realizing digital / analog conversion, having relevant timing interfaces, and realizing P-band RF direct sampling software radio processing. The computing performance of Soul Core No. 2A is 3 times higher than that of Texas Instruments TMS320C6678). The network AI intelligent translation server (i.e., network AI automatic translation host) is also selected from existing ones.

[0035] In this embodiment, the outer housing includes a first upper cover 101, a first bottom shell 102, a shock-absorbing pad 103 and a support frame 104. The support frame 104 has a vertical section and a horizontal section fixedly connected to the lower end of the vertical section. Specifically, the clamping angle between the vertical section and the horizontal section of the support frame 104 is less than 90 degrees, so that the outer body 10 is fixed on the support frame 104 and faces the people outside the counter, which is more conducive to receiving voice. At the same time, it is also convenient to fix and place the outer body 10. The first bottom shell 102 is located at the front side of the vertical end, the shock absorbing pad 103 is located between the first bottom shell 102 and the vertical section, the first bottom shell 102 is provided with a recess for accommodating the shock absorbing pad 103, the rear side of the shock absorbing pad 103 is located outside the recess, and the first bottom shell 102 is fixed to the vertical section by screws; the first upper cover 101 is located at the front side of the first bottom shell 102, the inner wall of the first upper cover 101 is provided with a groove corresponding to the edge of the first bottom shell 102, and the groove is embedded with a sealing ring 105 for sealing when the first upper cover 101 and the first bottom shell 102 cooperate, so that the external machine body 10 outside the counter can be prevented from being damaged by water vapor such as rain. With the above-mentioned structural setting, the shock absorbing pad 103 is provided, and when the external machine body 10 is working (that is, the vibration generated by the external machine speaker 14 when the external machine body 10 is talking), it can be filtered out by the shock absorbing pad 103 and will not be transmitted to the support frame 104, which can effectively avoid noise. In addition, when the external unit body 10 is installed on the support frame 104, it can be tilted toward the people outside the counter, so that the sound waves and sound intensity of the voices of the staff inside the counter can be received in a wider range, which can ensure better voice reception and maximize the hearing of the voices of the staff inside the counter, which is more conducive to clear sound transmission and the second AI voice processing unit 08 to recognize the sound.

[0036] In this embodiment, the first upper cover 101 is provided with a mounting hole, and an alarm switch 15 is installed on the first upper cover 101 through the mounting hole to cause the internal speaker 24 to emit a whistle sound in a loop-connected manner. The external unit sound inlet 11 and the external unit sound outlet 12 are both located on the first upper cover 101, and the external unit microphone 13, the external unit speaker 14 and the alarm switch 15 are arranged from top to bottom on the first upper cover 101 (i.e., the external unit sound inlet 11 is located below the external unit sound outlet 12, the alarm switch 15 is located below the external unit sound inlet 11, and the alarm switch 15 is located below the external unit sound inlet 11). With the above-mentioned structural arrangement, it is possible to realize that the alarm switch 15 can be used to generate a reminder for the personnel inside the counter to receive the response request of the personnel outside the counter.

[0037] In this embodiment, a telephone module is also installed in the inner shell body. Here, the telephone module is selected from the telephone module in the existing fixed telephone (that is, the same as the telephone module in the traditional fixed telephone, and the telephone module has the functions of display, dialing, dialing pronunciation and caller ID). The telephone module has a display screen 014 and a dial button 011. The inner shell body is provided with a hole structure for exposing the display screen and the dial button outside the inner shell body. The display screen is electrically connected to the corresponding output end of the MCU02, and the dial button is electrically connected to the corresponding input end of the MCU. A telephone interface for connecting to an indoor telephone line is installed on the inner body 20, and the MCU02 is connected to the telephone interface. The telephone module also has a dial tone generator 012 and an incoming call detection circuit (not shown in the figure), the dial tone generator 012 is installed on the dial button 011, the dial button 011 and the dial tone generator 012 are electrically connected to the corresponding input end and the corresponding output end of the MCU02, respectively, and the display screen 014 is fixed on the dial button 011, and the display screen 014 here uses an existing liquid crystal display screen 014, which is mainly used to display the dial number and the incoming call display number. The display screen 014 is electrically connected to the corresponding output end of the MCU02. Specifically, the display screen 014, the incoming call detection circuit, the dial button 011 and the dial tone generator 012 are all electrically connected to the MCU02 in a conventional manner in the art (that is, the display screen 014, the dial button 011, the incoming call detection circuit and the dial tone generator 012 are the same as the display screen, dial button, dial tone generator 012 and the incoming call detection circuit in the existing fixed-line telephone, and are connected in the same manner as in the existing fixed-line telephone). The dial tone sounder 012 uses an existing small speaker. With the above structure, when working, the incoming call detection circuit can receive the incoming call information and transmit it to MCU02. The received incoming call information is processed by MCU02 and displayed on the display screen 014 (i.e., the number is displayed). When dialing is performed through the dial button 011, the staff in the counter triggers the corresponding button on the dial button 011, and the corresponding circuit on the dial button 011 is connected, and then the dial tone sounder 012 sounds, generating a sound feedback reminder for the staff in the counter when dialing.

[0038] In this embodiment, the inner machine body 20 is also provided with a power supply module connected to the indoor power supply and providing power to the inner machine body 20 and the outer machine body 10. Here, the power supply module adopts the existing ones in the art. The power supply module is located on the left side of the audio switch circuit. The power supply module is connected to the audio switch circuit through a wire, and the incoming call detection circuit is located between the power supply module and the audio switch circuit. In this way, the overall structure of the inner machine body 20 is more compact, the internal space layout is more reasonable, and there will be no mutual influence or interference during operation. Specifically, the power supply module is connected to the power input end of the audio switch circuit through an ordinary wire to realize a voltage-stabilized power supply for the audio switch circuit.

[0039] In this embodiment, the inner housing includes a second upper cover 201 and a second bottom shell 202, the dial button 011 is fixedly mounted on the second bottom shell 202, the dial button 011 has a trigger end for people to dial, the trigger end is distributed at the transparent plate 203 and the second upper cover 201, the transparent plate 203 is provided with a first through hole for exposing the display screen 014, the second upper cover 201 is provided with a second through hole for exposing the trigger end, the second through hole is located in front of the first through hole, the first through hole and the second through hole are the above-mentioned hole structure The second through hole is located on the front side of the transparent plate 203 in the second upper cover 201; the second upper cover 201 is provided with a transparent plate 203 for protecting the display screen 014 in a covering manner and for projecting the information of the display screen 014 out of the second upper cover 201; the inner machine sound outlet 22 is located on the second upper cover 201 and is located on the left side of the transparent plate 203 and the second through hole; the inner machine sound inlet 21 is located on the front side edge of the second upper cover 201; the second upper cover 201 is provided with a device for adjusting the outer machine body 10 on the front side portion of the second through hole The first knob switch 204 for volume and the second knob switch for adjusting the volume of the indoor unit body 20, the first knob switch 204 and the second knob switch 205 are arranged side by side on the second upper cover 201, the first knob switch 204 is on the left side of the second knob switch 205, the first knob switch 204 and the second knob switch 205 are both electrically connected to the MCU02, here the first knob switch 204 and the second knob switch 205 are conventionally connected to the audio switch circuit (that is, the first knob switch 204 and the second knob switch 205 are connected to the corresponding input end of the audio switch circuit by wire or welding, and the corresponding loop of the audio switch circuit is connected), the voice sending control button 25 is located on the second upper cover 201, and is between the first knob switch 204 and the second knob switch 205. With the above structural setting, the layout is more reasonable, the structure is more compact, and the space occupancy is smaller. The first knob switch 204 and the second knob switch 205 are mainly used for the staff inside the counter to adjust the playback volume of the internal unit 20 and the playback volume of the external unit 10 respectively, to ensure that people with poor hearing outside the counter (i.e. the elderly or people with weak hearing, etc.) can hear clearly, and at the same time, to avoid howling caused by the playback volume of the internal unit 20 and the playback volume of the external unit 10, to ensure that the volume is controllable and facilitate normal communication.

[0040] In this embodiment, the second bottom shell 202 has an upper port that is inclined downward from the back to the front and is matched with the second upper cover 201. This upper port is mainly formed because the rear side wall of the second bottom shell 202 is higher than the front side wall. The upper port of the second bottom shell 202 is integrally formed with waterproof ribs that protrude upward along the edge of the port. The waterproof ribs here mainly play a waterproof and shockproof effect. The upper edge of the second upper cover 201 is provided with a positioning step that fits closely with the outer edge and upper edge of the waterproof rib. The inner wall of the second upper cover 201 is provided with a plurality of vertical first reinforcing ribs used to reinforce its side wall. The inner wall of the second bottom shell 202 is provided with second reinforcing ribs corresponding to the first reinforcing ribs, and the first reinforcing ribs are provided with notches that match the second reinforcing ribs. With the above-mentioned structural setting, when the second upper cover 201 is matched with the second bottom shell 202, it can still remain inclined toward the staff in the counter, which is convenient for the staff in the counter to operate. At the same time, through this inclination (that is, when the second upper cover 201 is matched on the second bottom shell 202, it remains inclined toward the staff in the counter), it is possible to receive the sound waves of the voice of the staff in the counter in a wider range, which can ensure a better voice reception effect, so that the voice of the staff in the counter can be heard to the maximum extent, which is more conducive to clear sound transmission and the second AI voice processing unit 08 to recognize the sound.

[0041] In this embodiment, the audio switch circuit is fixed in the second bottom shell 202, and the second bottom shell 202 is provided with a USB interface 1, a SIM1 card slot 2, a SIM2 card slot 3, a sound input interface 4, a sound output interface 5, a network interface 6, a power interface 8 and a headphone jack which are electrically connected to the MCU02 respectively, and the telephone interface 7 is in the second bottom shell, and the USB interface 1, the SIM1 card slot 2, the SIM2 card slot 3, the sound input interface 4, the sound output interface 5, the network interface 6, the telephone interface 7 and the power interface 8 are arranged from left to right in the second bottom shell 202, and a third through hole for exposing the headphone jack is provided on the left side wall of the second bottom shell 202, and a third through hole for exposing the headphone jack is provided on the rear side wall of the second bottom shell. SIM2 card slot 3, voice input interface 4, voice output interface 5, network interface 6, telephone interface 7, power interface 8 are exposed in the fourth through hole, and the second bottom shell 202 is also equipped with a main switch located directly above the SIM1 card slot 2 and connected to the wire between the power supply module and the audio switch circuit. Here, the USB interface 1, SIM1 card slot 2, SIM2 card slot 3, voice input interface 4, voice output interface 5, network interface 6, telephone interface 7 and power interface 8 are all electrically connected to the audio switch circuit in a conventional manner in the art. With the above-mentioned structure, the headphone jack is set, and the voice of the person outside the counter can be listened to by plugging in headphones. The SIM1 card slot 2 and SIM2 card slot 3 are set mainly to meet the different network signal coverage in different regions and different operators can be selected. The USB interface 1 uses a MICRO USB interface (i.e., MICRO refers to a small USB interface commonly used in the industry). Data can be imported and exported from the internal machine body 20 through the USB interface 1, and the Android phone is powered by the data cable. The network interface 6 and telephone interface 7 are set to facilitate the access of the network cable and the access of the telephone line respectively. The power jack 8 is used in conjunction with an existing power adapter to facilitate the connection of the output end of the existing power adapter after the indoor power is transformed. The sound input interface 4 is connected to a hand-held meter or a microphone to input sound to the indoor unit body 20, or to transmit commands through the indoor unit body 20 through voice input via an external playback device. The SIM1 card slot 2 and the SIM2 card slot 3 are provided, and by inserting a SIM card (i.e., inserting a SIM card of China Mobile, China Unicom, or China Telecom), the 4G wireless data transmission module 09 can also realize intercom with an existing intercom or mobile phone through a base station.

[0042] In this embodiment, the 4G wireless data transmission module 09 has an antenna 28, and the right side wall of the second bottom shell 202 is rotatably mounted with an antenna 28 for enhancing the signal generation and reception. The antenna uses an existing 4G antenna. The above-mentioned structure is set up, and the antenna 28 is set up mainly to enhance the signal, reduce the blocking of the second bottom shell 202 and the second upper cover 201 to the 4G wireless data transmission module 09 signal, enhance the transceiver capability between the 4G wireless data transmission module 09 and the network, and is more conducive to signal stability, and more conducive to the stable transceiver of the network AI intelligent translation server translation information.

[0043] In this embodiment, the inner housing is also equipped with an intercom voice sending control button 26 (i.e., the PPT key for intercom calls on existing intercoms) that is electrically connected to the audio switch circuit and controls the network cloud voice intercom server to communicate in a circuit connection loop mode. The inner housing is also equipped with a key switch 27 for answering wired calls. The key switch 27 is connected in a conventional manner. The key switch 27 uses an existing ordinary one. The intercom voice sending control button 26, the key switch 27, and the voice sending control button 25 are arranged side by side on the second upper cover 201 and are located between the first knob switch 204 and the second knob switch 205. The key switch 27 is located between the intercom voice sending control button 26 and the voice sending control button 25. The intercom voice sending control button 26 is located on the left side of the key switch 27. With the above-mentioned structural arrangement, the intercom voice sending control button 26 is mainly used for switching between wired telephone calls and wireless telephone calls. It is easy to use and has two functions in one. When the wired circuit fails, the key switch 27 can be used to switch to a wireless phone call. This solves the problem that the traditional phone can only rely on the wired circuit for phone calls, reduces the dependence on the telephone line, avoids the impact of the telephone line disconnection, and still ensures normal use, so it is more convenient to use.

[0044] As is known to all, the staff in the counter have to deal with a lot of documents and reconciliation bills. The counter desktop space is very limited, so the desktop is messy and seems insufficient. When not in use, the internal machine body 20 used for internal and external calls takes up a certain amount of desktop space. The internal machine body 20 is placed horizontally on the counter surface, so the dialing buttons on the internal machine body 20 are easily squeezed by a whole pile of messy documents or reconciliation bills, resulting in misdialing, etc. In addition, the position of the internal machine body 20 on the counter is also fixed, so it is impossible to quickly approach the staff, fail to receive sufficient sound intensity, and cannot ensure clear calls. How to effectively utilize the counter desktop is further studied. Therefore, the present invention is more preferably provided with a base 1' fixed to the countertop in a conventional manner at the bottom of the inner housing, and two tongue grooves 2' extending along the front-rear direction of the inner housing are provided on the base 1', and the two tongue grooves 2' are arranged side by side on the top surface of the base 1' and are respectively close to the left and right side edges of the base 1'; the bottom of the inner housing is provided with two fixing rods (not shown in the figure) whose lower ends are respectively against the groove walls of the two tongue grooves 2', and the two fixing rods are provided on the bottom of the inner housing. The upper ends of the rods are respectively hinged at the two corners of the rear part of the inner housing, and the bottom of the inner housing is provided with two telescopic rods 3', whose lower ends are respectively against the groove walls of the two tongue grooves 2', and cause the front part of the inner housing to tilt downward to the rear; the upper ends of the two telescopic rods 3' are hinged at the corners of the front part of the inner housing, and specifically, the inner housing is integrally formed with an articulated seat, and the upper ends of the telescopic rods 3' are provided with an articulated ring, and the articulated seat and the articulated ring are connected by a pin. The lower end of the telescopic rod 3' and the lower end of the fixed rod both have a tenon structure that matches the tongue groove 2', and the tenon structure refers to the lower end of the telescopic rod 3' and the lower end of the fixed rod. There is a square block below the lower end of the fixed rod, and the height of the fixed rod just extends out of the tongue groove 2' and satisfies the inner housing to be re-stacked on the bottom bracket 1'. The length of the fixed rod is much smaller than the length of the telescopic rod 3'; the retracted length of the telescopic rod 3' is 5 times the length of the fixed rod, and a limit screw (not shown in the figure) is screwed on the groove wall of the mortise 2', which is located between the fixed rod and the telescopic rod 3' and is used to prevent the lower end of the fixed rod from sliding out of the front part of the mortise 2'. The limit screw is an existing self-tapping screw. The telescopic rod 3' has an inner section rod and an outer section rod sleeved outside the inner section rod, and the outer section rod is provided with a limit member that controls the inner section rod to extend the longest out of the outer section rod and not to separate from the outer section rod. Specifically, a hinge ring is integrally formed at the upper end of the inner section of the telescopic rod 3', and a tenon structure is integrally formed at the lower end of the outer section of the telescopic rod 3'. The telescopic rod 3' is similar to the telescopic rod 3' of an existing ordinary umbrella (that is, the inner section is extended through a limiter to be fixed to the outer section in the same way as the telescopic rod 3' on an existing ordinary umbrella). In other words, the limiter is the inner ball and spring or spring sheet of the telescopic rod 3' of the existing umbrella.With the above structure, when the counter staff needs to talk to people outside the counter, they can directly move the inner housing closer to the counter staff by relying on the tongue and groove 2' on the bottom bracket 1', so as to ensure that the inner microphone in the inner body 20 can receive sufficient sound intensity and ensure clear conversation. When the counter staff is sorting out the documents and reconciliation bills, they can lift the front side of the inner housing on the bottom bracket 1' upwards. Due to the support of the fixed rod and the telescopic rod 3', the inner housing is kept lifted up at the front of the bottom bracket 1' and tilted backwards. At this time, a space for temporarily accommodating the documents and reconciliation bills is formed between the inner housing and the bottom bracket 1', so as to avoid the problem that the documents and reconciliation bills fall on the floor or are lost, which is more convenient. At the same time, the use space occupied by the inner housing on the desktop is expanded, and the utilization rate of the desktop is effectively increased. When the counter staff finishes sorting out the documents and reconciliation bills, after taking out the documents and reconciliation bills, the telescopic rod 3' is retracted and slides along the tongue and groove 2', and directly and automatically lies in the tongue and groove 2', so that the inner housing is stacked on the bottom bracket 1' again.

[0045] Window intercom and echo cancellation and noise suppression part: The external unit body 10 and the internal unit body 20 collect the speaker (i.e. the person speaking to the external unit microphone 13 outside the counter and the person speaking to the internal unit microphone 23 inside the counter) and the environmental noise through the corresponding external unit microphone 13 and the corresponding internal unit microphone 23 respectively, and design a Deep Neural Network that recognizes the characteristics of human voice through the AI ​​voice processing unit (i.e. the first AI voice processing unit 07 and the second AI voice processing unit 08, the same below) and echo cancellation and noise suppression. The DNN network has a preset standard answer (i.e., clean and clear human voice), and combines the standard answer with various noises. The input network allows the DNN network to find the appropriate model and parameters to obtain the result that matches the standard answer. It has built-in up to 400,000,000 trainings to ensure that the model and parameters have a large enough adaptability, such as typical noises: restaurants / gathering places, buses, subways, train stations, wind noise, music noise, etc. After the training is completed, the working method is to identify the voice features. No matter what kind of noise, the same parameters / methods are used to obtain clean human voices, so that the human voice can be recognized and selected like a human, achieving environmental noise suppression. Built-in dual DSP for basic operations on voice data, exclusive offline AI voice processing unit, 3-layer network with 1028 nodes, computing power: 1012 times / second.

[0046] The basic principle block diagram of the noise suppression algorithm: Figure 6As shown, after the internal and external microphones 13 collect voice and environmental noise, the analog voice is digitized by the AI ​​voice processing unit and the built-in DSP (i.e., DSP digital signal processing), and then restored to human voice by CODEC (i.e., built-in precise 24-bit stereo ADC and DAC, both ADC and DAC are selected from existing ones, and both are well known in the art), the sound of the internal and external units is switched to the internal and external unit audio power amplifier circuit for amplification through the audio switch circuit, and then output to the internal and external unit speakers 14 respectively, to restore the human voice of the internal and external units. At the same time, because the internal and external unit speakers 14 and the microphone are in the same structural cavity, in order to prevent howling and echo, the output of the internal and external unit speakers 14 simultaneously sends a signal to the AI ​​voice processing unit for detection, and the built-in high-computing-power DSP is used to eliminate the echo algorithmically, so that when the internal unit body 20 and the external unit body 10 are used in conjunction, no echo or howling is generated.

[0047] Basic principle of intercom function: Based on mobile Internet technology, this system adopts the data transmission and processing system of 4G wireless data transmission module 09 (that is, the existing 4G LTE module, and similar ones), and connects this intercom system to the network cloud voice intercom server through TCP / IP link. When the local machine initiates a call, the voice is first compressed and converted into a digital signal through the 4G LTE data transmission and processing system, and uploaded to the voice intercom server through the TCP / IP network. After the server analyzes the data, it forwards the data to all intercom terminals in the same group as the initiating terminal. After receiving the data, each intercom terminal decompresses the data and restores the voice, and an intercom call ends. The multifunctional smart window intercom system of the present invention has a dedicated SIM card interface and an antenna 28 interface connected to the 4G LTE data transmission and processing system.

[0048] The basic block diagram of the network intercom server: Figure 7 As shown in the figure, the server cluster is used to support all core capabilities and services of the entire cloud intercom platform, such as calls, conversations, messages, etc.; DNS, as a front-end service, provides LVS load balancing for DM. DM allocates SPs to users based on its own balancing strategy. Multiple DMs are distributed in a peer-to-peer manner, and data access operations are performed in a shared library manner. Multiple SPs are distributed in a peer-to-peer manner and communicate based on a routing mechanism. Each SP manages one or more MDSs. SP is the core processing service that carries the status of online users and maintains voice conversations and instant messages for all heartbeat users. Multiple SPs are load balanced through the DM running monitoring module. Each SP is completely peer-to-peer and addresses and communicates with each other through a routing mechanism. Specific functions: responsible for the maintenance of user online status, presentation of Presence user status, processing voice calls and message services between users, providing message push channels, and the intelligent routing module is responsible for addressing and communication between SPs, responsible for media communication and balanced distribution of MDS services.

[0049] Server cluster - DM (Data Manager) data management service: DM service is responsible for all data management tasks related to DB interaction in the entire system, providing TCP access interface to the client, and also providing interaction interface corresponding to DB operation for other services. DB can be deployed as a service on different devices separately, using MySQL database, Master / Slaver mode, and MemCache mechanism well known in the art. Specific functions: responsible for monitoring the operation status of each functional network element, dynamically allocating functional server IP and Port to users, supporting multi-point deployment of SP and MDS, load balancing control, responsible for interactive processing with the core interface on the terminal side, carrying core data related logic processing capabilities, and using a streamlined and encrypted transmission protocol with the terminal side interface. Server cluster - MDS (Media Server) media service: Media service is mainly responsible for the voice call and group voice right control of users, and is divided into two service forms: 1) temporary group call, 2) fixed group or channel call. Specific functions: Provide temporary group calls, i.e. 1-to-1 and 1-to-many real-time voice call services, responsible for group-level communication capabilities such as channel voice control, media transmission, and in-channel messages, responsible for supporting channel management modes, responsible for audio codec conversion, and audio filtering processing, etc. The terminal is used to rely on the server cluster and realize intercom calls with the dispatch control center, management center, and user center through the network intercom platform (i.e. network AI intelligent translation server).

[0050] The network intercom server achieves the following effects: first, it realizes account and group management; second, it realizes selective calling and group calling functions; third, it realizes audio and video transmission, location and motion trajectory playback; fourth, the voice is linked to the network cloud voice intercom server through TCP / IP, so that counter staff can communicate or dispatch across departments and regions through real-time intercom. Through mobile Internet technology, it can also solve the signal blind spots or communication limitations caused by traditional intercoms due to floor shielding or long-distance communication needs across regions due to distance limits.

[0051] The working principle of the Chinese-English automatic translation is: based on the technical foundation of the above-mentioned intercom server platform, the voice is compressed and converted into a digital signal through the data transmission and processing system of 4GLTE, and uploaded to the AI ​​automatic translation server platform (i.e., network AI intelligent translation server) through the TCP / IP network, and combined with the neural network speech recognition technology (DNN), the Chinese or English voice is converted into the corresponding English or Chinese data respectively, and then the data is forwarded to the multifunctional intelligent window intercom system of the present invention through the intercom server platform, and switched to the audio power amplifier of the internal and external machines through the audio voice switch circuit, and then output to the speaker to realize the Chinese-English automatic translation function. Solve the problem that the overseas customers outside the counter have poor communication and affect the business handling. The MCU02 in the multifunctional intelligent window intercom system of the present invention mainly controls the time-sharing opening and closing of the functions of each subsystem. The present invention provides a multifunctional intelligent window intercom system, which integrates multiple functions and uses an AI voice processing unit and network technology (i.e., a first AI voice processing unit 07, a second AI voice processing unit 08, a 4G wireless data transmission module 09, a digital intercom module, and a telephone module) to solve the rigid needs of work at all levels of the industry without the need for frequent switching between multiple devices, thereby improving work efficiency and increasing the utilization rate of the work surface.

[0052] The above embodiments and drawings do not limit the product form and style of the present invention. Any appropriate changes or modifications made by ordinary technicians in the relevant technical field should be deemed to be within the patent scope of the present invention.

Claims

1. A multifunctional intelligent window intercom system, comprising an external machine body arranged outside the counter window glass and an internal machine body arranged inside the counter window glass, the external machine body having an external machine shell, the internal machine body having an internal machine shell, the external machine shell having an external machine sound inlet hole and an external machine sound outlet hole, an external machine microphone and an external machine speaker are installed in the external machine body at positions corresponding to the external machine sound inlet hole and the external machine sound outlet hole, a first voice digital codec module and a first audio power amplifier module are installed in the external machine body, an audio switch circuit having an MCU, a second voice digital codec module and a second audio power amplifier module is installed in the internal machine body, the internal machine shell having an internal machine sound inlet hole and an internal machine sound outlet hole, an internal machine microphone and an internal machine speaker are installed in positions corresponding to the internal machine sound inlet hole and the internal machine sound outlet hole in the internal machine body, characterized in that: It also includes a first AI voice processing unit and a second AI voice processing unit for sound intensity recognition, echo elimination and noise suppression, a network AI intelligent translation server for Chinese-English translation, and a 4G wireless data transmission module for wireless communication connection with the network AI intelligent translation server; the network AI intelligent translation server is connected to the MCU for data bidirectional transmission via the 4G wireless data transmission module, the external microphone is connected to the input end of the first AI voice processing unit for data transmission, the output end of the first AI voice processing unit is connected to the input end of the first voice digital codec module, the output end of the first voice digital codec module is connected to the corresponding input end of the MCU, and the control end of the MCU is electrically connected to a useful The voice sending control button is used to control the MCU to receive the data of the first voice digital codec module and send it out using the 4G wireless data transmission module. The input end of the first audio power amplifier module is connected to the corresponding output end of the MCU, and the output end of the first audio power amplifier module is connected to the external speaker; the internal microphone is connected to the input end of the second AI voice processing unit for data transmission, the output end of the second AI voice processing unit is connected to the input end of the second voice digital codec module for data transmission, the output end of the second voice digital codec is connected to the corresponding input end of the MCU, the input end of the second audio power amplifier module is connected to the corresponding output end of the MCU, and the output end of the second audio power amplifier module is connected to the internal speaker; The outer casing comprises a first upper cover, a first bottom shell, a shock-absorbing pad and a support frame, wherein the support frame has a vertical section and a transverse section fixedly connected to the lower end of the vertical section, the first bottom shell is located at the front side of the vertical section, the shock-absorbing pad is located between the first bottom shell and the vertical section, a pit for accommodating the shock-absorbing pad is concavely provided on the first bottom shell, the rear side surface of the shock-absorbing pad is located outside the pit, and the first bottom shell is fixed to the vertical section by screws; the first upper cover is located at the front side of the first bottom shell, a groove is provided on the inner wall of the first upper cover corresponding to the edge portion of the first bottom shell, and a sealing ring is embedded in the groove for sealing when the first upper cover and the first bottom shell cooperate; The bottom of the inner casing is provided with a base for fixing on the counter top, and the base is provided with two tongues and grooves extending in the front-rear direction of the inner casing, and the two tongues and grooves are arranged side by side on the top surface of the base and close to the left and right side edges of the base respectively; the bottom of the inner casing is provided with two fixing rods whose lower ends respectively lean against the groove walls of the two tongues and grooves, and the upper ends of the two fixing rods are respectively hinged at the two corners of the rear part of the inner casing, and the bottom of the inner casing is provided with two telescopic rods whose lower ends respectively lean against the groove walls of the two tongues and cause the front part of the inner casing to tilt downward to the rear; the upper ends of the two telescopic rods are hinged at the corners of the front part of the inner casing, and the lower ends of the telescopic rods and the lower ends of the fixing rods both have tenon structures matched with the tongues and grooves, and the length of the fixing rods is less than the length of the telescopic rods; A limiting screw is installed on the groove wall of the mortise and tenon, which is located between the fixed rod and the telescopic rod and is used to prevent the lower end of the fixed rod from sliding out of the front part of the mortise and tenon. The telescopic rod has an inner section rod and an outer section rod sleeved outside the inner section rod. The outer section rod is provided with a limiting piece that controls the inner section rod to extend to the longest extent of the outer section rod without separating from the outer section rod.

2. A multifunctional intelligent window intercom system according to claim 1, characterized in that: The first upper cover is provided with a mounting hole, and an alarm switch is installed on the first upper cover through the mounting hole to connect a circuit and cause the internal unit speaker to emit a whistle sound. The external unit sound inlet and the external unit sound outlet are both located on the first upper cover, and the external unit microphone, the external unit speaker and the alarm switch are arranged from top to bottom on the first upper cover.

3. A multifunctional intelligent window intercom system according to claim 1 or 2, characterized in that: It also includes a network cloud voice intercom server for intercom with the intercom, a digital intercom module for wireless intercom with the intercom connected to the network cloud voice intercom server is installed in the inner machine body, the digital intercom module is connected to the MCU for two-way data transmission, and the digital intercom module is connected to the network cloud voice intercom server via a 4G wireless data transmission module; the control end of the MCU is electrically connected to an intercom voice sending control button for controlling the MCU to receive data from the second voice digital codec module and send it out using the 4G wireless data transmission module.

4. A multifunctional intelligent window intercom system according to claim 3, characterized in that: The inner unit body is also provided with a power module which is connected to the indoor power supply and provides power to the inner unit body and the outer unit body. The power module is located on the left side of the audio switch circuit. The power module is connected to the audio switch circuit for power supply via a wire, and the incoming call detection circuit is located between the power module and the audio switch circuit.

5. A multifunctional intelligent window intercom system according to claim 4, characterized in that: A telephone module with display and dialing functions is also installed in the inner shell. The telephone module has a display screen and dialing buttons. The inner shell is provided with a hole structure for exposing the display screen and the dialing buttons outside the inner shell. The display screen is electrically connected to the corresponding output end of the MCU, and the dialing buttons are electrically connected to the corresponding input end of the MCU. A telephone interface for connecting to an indoor telephone line is installed on the inner body, and the MCU is connected to the telephone interface.

6. A multifunctional intelligent window intercom system according to claim 5, characterized in that: The inner housing comprises a second upper cover and a second bottom shell, the audio switch circuit is fixed in the second bottom shell, the dial module is fixedly mounted on the second bottom shell, the second bottom shell is provided with a first through hole for the display screen to be exposed outside the second bottom shell, the second upper cover is provided with a second through hole for the dial module to be partially exposed outside the upper cover, the second through hole is located in front of the first through hole, the first through hole and the second through hole are the above-mentioned hole structures; the second upper cover is provided with a transparent plate for covering the display screen and for projecting the display screen information outside the second upper cover, the inner sound outlet hole is located on the second upper cover and is located on the left side of the transparent plate and the second through hole, the inner sound inlet hole is located on the front edge of the second upper cover, the second upper cover is provided with a first knob switch for adjusting the volume of the outer speaker and a second knob switch for adjusting the volume of the inner speaker, the first knob switch and the second knob switch are arranged side by side on the second upper cover, and the voice sending control button is located on the second upper cover and is located between the first knob switch and the second knob switch.

7. A multifunctional intelligent window intercom system according to claim 6, characterized in that: The second bottom shell has an upper port which is inclined downward from the back to the front and is matched with the second upper cover. The upper port of the second bottom shell is integrally formed with a waterproof rib which protrudes upward along the edge of the port. The upper edge of the second upper cover is provided with a positioning step which closely matches the outer edge and the upper edge of the waterproof rib. The inner wall of the second upper cover is provided with a plurality of vertical first reinforcing ribs which are used to reinforce the side wall thereof. Second reinforcing ribs corresponding to the first reinforcing ribs are provided on the inner wall of the second bottom shell, and the first reinforcing ribs are provided with notches which match the second reinforcing ribs.

8. A multifunctional intelligent window intercom system according to claim 7, characterized in that: The intercom voice sending control button is located on the second upper cover and between the first knob switch and the second knob switch, and the intercom voice sending control button is located on the left side of the voice sending control button.

Images