Intelligent lifting table based on voice recognition

Abstract

The utility model relates to intelligent lifting table technical field especially based on speech recognition's intelligent lifting table, innovatively with speech recognition, environmental regulation and lifting function through customization hardware architecture and hierarchical control algorithm depth fusion. Its technical core includes: time multiplexing communication bus management multimode instruction, strain gauge type dynamic load monitoring and heat-mechanical-electric integrated structure design, significantly improved the function density and reliability of intelligent furniture, applicable to high -end office, medical treatment and education scene.

Description

Technical Field

[0001] This utility model relates to the field of intelligent height-adjustable desk technology, and in particular to an intelligent height-adjustable desk based on voice recognition. Background Technology

[0002] Traditional height-adjustable desks, while adjustable via motors, offer limited functionality and lack the ability to optimize the overall office environment. While existing smart home devices (such as voice-controlled fans and remotely dimmed lights) offer independent environmental control, their integration with height-adjustable desks still suffers from the following drawbacks:

[0003] 1. Functional fragmentation: Users need to operate the height-adjustable desk, lighting and ventilation equipment separately, which is cumbersome and the multiple devices take up space;

[0004] 2. Control delay: Insufficient coordination between voice recognition and motor drive, resulting in excessively long command response time (generally >1 second), affecting user experience;

[0005] 3. Structural conflict: External fans / lights are prone to interference with the lifting mechanism, and exposed cables pose safety hazards;

[0006] 4. Poor scalability: Existing solutions are difficult to integrate into the smart home ecosystem and cannot achieve scenario-based control such as temperature and humidity linkage.

[0007] Furthermore, the demand for health management in office settings is growing, and users urgently need smart furniture that can simultaneously optimize posture, lighting, and air circulation. Therefore, a highly integrated, responsive, and scalable height-adjustable desk system is urgently needed to address the aforementioned technical bottlenecks. Utility Model Content

[0008] To overcome the shortcomings of existing technologies, this invention provides an intelligent height-adjustable desk based on voice recognition.

[0009] The technical solution adopted by this utility model to solve its technical problem is as follows: a smart height-adjustable desk based on voice recognition, including a desktop panel, linear guide rail columns, a hidden cable management channel, and a base bracket; a multi-spectral LED lighting unit is fixed in the center of the desktop panel by a connecting rod, and mounting holes for embedding centrifugal silent fan units are symmetrically opened on the left and right edges; a hidden cable management channel and a base bracket are installed on the back, supported by symmetrical linear guide rail columns; a height-adjustable drive module is installed inside the linear guide rail columns; the centrifugal silent fan unit is fixed in the mounting holes of the desktop panel by a snap-on bracket; a control box is installed in the base bracket, and the control box has a built-in core control module. The core control module is based on an STM32F103ZET6 microcontroller and integrates a voice recognition unit and a wireless communication unit, and is connected to the multi-spectral LED lighting unit, the centrifugal silent fan unit, and the height-adjustable drive module through wires; the wires are installed in the hidden cable management channel.

[0010] According to another embodiment of the present invention, the lifting drive module is further comprising a dual-motor drive system and a load sensor. The dual-motor drive system is connected to the microcontroller via a CAN bus to pins PA11 (CAN_RX) and PA12 (CAN_TX). The load sensor is a strain gauge weighing module, and its output signal is conditioned by an AD623 instrumentation amplifier and then connected to the microcontroller's ADC1 channel 0 (pin PA0).

[0011] According to another embodiment of the present invention, the centrifugal silent fan unit further includes a small 12V DC cooling fan, which is connected to the TIM1 channel 1 (pin PA8) of the microcontroller via a PWM speed control circuit.

[0012] According to another embodiment of the present invention, the LED lighting unit further includes a digital dimming chip TLC59116, which achieves 16-bit precision brightness control by connecting to the microcontroller's I²C interface (pin PB6 / SCL, pin PB7 / SDA).

[0013] According to another embodiment of the present invention, the wireless communication unit is a WiFi module based on the ESP8266 chip, which is connected to the microprocessor through the USART2 serial communication interface (pin PA2 / TX, pin PA3 / RX), and realizes full-duplex communication with the cloud platform through the MQTT protocol.

[0014] 6. According to another embodiment of the present invention, the voice recognition unit further includes an LD3320 chip, which is connected to the microprocessor via a USART3 interface (pins PB10 / RX, PB11 / TX).

[0015] According to another embodiment of the present invention, the hidden cable management channel is provided on the back of the desktop panel, with a depth of 10mm, to accommodate motor power cables, LED power cables and sensor signal cables.

[0016] According to another embodiment of the present invention, the base bracket is further provided with an EMI filter at its bottom, and the input power supply AC 220V is filtered and distributed to each module in the control box.

[0017] The beneficial effects of this utility model;

[0018] 1. Enhanced Functional Integration: Deeply integrates lifting adjustment, environmental control (lighting / ventilation), and voice / remote operation, reducing equipment redundancy and saving more than 30% of desktop space; multi-spectral LED lighting supports color temperature adjustment from 2700K to 6500K, meeting the ISO 9241-307 visual ergonomics standard.

[0019] 2. Optimized response speed: The end-to-end latency from voice command recognition to execution is ≤0.5 seconds (compared to ≥1.2 seconds in traditional solutions), with a false trigger rate of <3%; the motor closed-loop control accuracy is ±0.5mm, significantly reducing lifting and lowering jitter.

[0020] 3. Enhanced stability and security: The time-division multiplexing communication bus improves the system's anti-interference capability, and the EMC test meets the 10V / m radio frequency field immunity requirement; the dual-cavity heat dissipation design extends the motor life to >100,000 lifting cycles.

[0021] 4. Intelligent expansion capabilities: Supports integration with third-party smart home platform protocols, expanding scene rules to more than 20 types; based on user behavior data, the adaptive algorithm can reduce energy consumption by 15% (through intelligent power on / off and automatic adjustment of brightness / fan speed).

[0022] 5. Industrial applicability verification: No deformation after bearing 150kg for 1 hour; Noise index is superior to GB / T 3768-2017 in all aspects, and the sound power level under full load is ≤45dB(A). Attached Figure Description

[0023] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0024] Figure 1 This is a schematic diagram of the structure of this utility model;

[0025] Figure 2 This is the electrical connection diagram of this utility model.

[0026] In the diagram: 1. Desktop panel; 2. Linear guide rail column; 3. Concealed cable management channel; 4. Base bracket; 5. Multispectral LED lighting unit; 6. Centrifugal silent fan unit; 7. Lifting drive module; 8. Control box; 9. Core control module; 10. Voice recognition unit; 11. Wireless communication unit; 12. Dual motor drive system; 13. Load sensor. Detailed Implementation

[0027] like Figure 1 This is a structural schematic diagram of a utility model, a smart height-adjustable desk based on voice recognition, characterized by comprising a desktop panel 1, linear guide rail columns 2, a concealed cable management channel 3, and a base bracket 4; a multi-spectral LED lighting unit 5 is fixed in the center of the desktop panel 1 via a connecting rod, and mounting holes for embedding centrifugal silent fan units 6 are symmetrically opened on the left and right edges; the concealed cable management channel 3 and the base bracket 4 are installed on the back, supported by symmetrical linear guide rail columns 2; a height-adjustable drive module 7 is installed inside the linear guide rail columns 2; the centrifugal silent fan unit 6 is fixed in the mounting holes of the desktop panel 1 by a snap-on bracket; a control box 8 is provided inside the base bracket 4, and the control box 8 has a built-in core control module 9, which is based on an STM32F103ZET6 microcontroller and integrates a voice recognition unit 10 and a wireless communication unit 11, and is connected to the multi-spectral LED lighting unit 5, the centrifugal silent fan unit 6, and the height-adjustable drive module 7 via wires; the wires are installed in the concealed cable management channel 3.

[0028] Specifically, the STM32F103 series microcontrollers serve as the main control chips. Their ARM Cortex-M3 processor cores can meet the requirements for fast and efficient performance, and have advantages such as low power consumption, large storage capacity, and rich peripheral interfaces. They are widely used in the field of industrial automation.

[0029] According to another embodiment of the present invention, the lifting drive module 7 is further comprising a dual-motor drive system 12 and a load sensor 13. The dual-motor drive system 12 is connected to the microcontroller via a CAN bus to pins PA11CAN_RX and PA12CAN_TX. The load sensor 13 is a strain gauge weighing module, and its output signal is conditioned by an AD623 instrumentation amplifier and then connected to the microcontroller's ADC1 channel 0 pin PA0.

[0030] According to another embodiment of the present invention, the centrifugal silent fan unit 6 further includes a small 12V DC cooling fan, which is connected to the TIM1 channel 1 pin PA8 of the microcontroller via a PWM speed control circuit.

[0031] Specifically, the fan is a small 12V DC cooling fan, square in shape, measuring 40mm × 40mm, with an airflow of 3.1 CFM, noise level below 25dB, and a speed of 6000 rpm. This fan employs a dual ball bearing design, ensuring long lifespan and stable, reliable operation. During system operation, the fan is controlled via an STM32 microcontroller, adjusting the PWM wave duty cycle based on voice or mobile app input to automatically regulate the fan speed.

[0032] According to another embodiment of the present invention, the LED lighting unit 5 further includes a digital dimming chip TLC59116, which achieves 16-bit precision brightness control by connecting to the I²C interface pins PB6 / SCL and PB7 / SDA of the microcontroller.

[0033] According to another embodiment of the present invention, the wireless communication unit 11 is a WiFi module based on the ESP8266 chip, which is connected to the microcontroller through the USART2 serial communication interface pins PA2 / TX and PA3 / RX, and achieves full-duplex communication with the cloud platform through the MQTT protocol.

[0034] Specifically, the ESP8266 is a low-cost WiFi module that can be used as an add-on module for microcontrollers to provide wireless connectivity to devices. The ESP8266 module integrates a WiFi interface and processor, can communicate with the main controller via serial port, and can be configured and controlled via AT commands. It supports multiple WiFi security protocols and network modes, and can connect to existing WiFi networks or provide network services as a standalone WiFi access point.

[0035] According to another embodiment of the present invention, the voice recognition unit 10 is equipped with an LD3320 chip and is connected to the microcontroller via USART3 interface pins PB10 / RX and PB11 / TX.

[0036] Specifically, the LD3320 chip is a dedicated voice recognition chip. It directly integrates a voice recognition processing module and external circuitry, enabling it to perform voice recognition, voice control, and other functions. This system uses the LD3320 chip as its voice recognition module. The LD3320 captures ambient sound signals through its built-in microphone and converts them into digital signals. After processing the digital signals using algorithms such as spectrum analysis, feature extraction, and voice recognition, the system matches the voice recognition results with preset commands and outputs corresponding control signals. When a user issues voice commands such as "raise," "lower," "turn on light," "turn off light," "turn on fan," "turn off fan," "speed 1," "speed 2," and "speed 3," the LD3320 recognizes the command in the voice, converts it into a corresponding recognition code, and outputs it to the STM32F103ZET6 main control chip for parsing and execution of the corresponding command. The entire process is a loop: the microphone continuously collects voice signals, the LD3320 continuously processes the voice signals and outputs the recognition results, and the main control chip continuously parses the commands and executes the corresponding operations.

[0037] According to another embodiment of the present invention, the hidden cable management channel 3 is provided on the back of the desktop panel 1, with a depth of 10mm, to accommodate the motor power cable, LED power cable and sensor signal cable.

[0038] According to another embodiment of the present invention, the base bracket 4 is provided with an EMI filter at its bottom, and the input power supply AC 220V is filtered and distributed to each module in the control box 8.

[0039] The specific operation process is as follows: 1. After powering on the STM32F103ZET6, initialize each peripheral: configure USART2 (115200bps) to connect to the ESP8266 module and subscribe to OneNET cloud platform topics; load the LD3320 voice command library through USART3 (9600bps); initialize TIM1 channel 1 (PWM frequency 25kHz) to control the fan speed, and configure the TLC59116 dimming chip through the I²C interface; self-test the lifting drive module: read the motor status (temperature, position) through the CAN bus, and trigger an LED alarm if there is an abnormality.

[0040] 2. User Command Processing Flow: Voice Control: When the user issues the command "Raise to standing position," the LD3320 acquires the voice signal through a MEMS microphone array. The signal is filtered by FIR and matched with keywords using the DTW algorithm to generate an identification code (0x01), which is sent to the STM32 via USART3. After parsing the command, the STM32 sends the target position (110cm) to the dual motors via the CAN bus and simultaneously activates PWM output (30% duty cycle) to control the fan at speed 1 and adjusts the I²C dimming to 5000K / 300lux. The motors provide real-time position data feedback, and the STM32 dynamically adjusts the PID parameters to ensure smooth lifting (speed 30mm / s, acceleration 0.5m / s²). APP Remote Control: The user sends the JSON command {"height":75, "light":2700K, "fan":0} via the APP. After the ESP8266 receives the data, the STM32 parses and executes the command: the desktop is lowered to 75cm, the lights are switched to warm light, and the fan is turned off. The device status is transmitted back to the APP interface via the MQTT protocol, updating real-time parameters. Adaptive scenario: The system monitors historical data (such as daily standing time). When continuous use exceeds 1 hour, it automatically triggers the "health reminder" mode: the desktop vibrates slowly to alert the user, the light flashes 3 times, and the fan switches to the second level to enhance ventilation; if the user does not respond, it automatically rises to the preset standing height after 10 seconds.

[0041] 3. Data storage and expansion: User preference data is stored in SPI Flash (W25Q128) and supports power-off data retention.

[0042] The above description is illustrative only and not restrictive of this utility model. Those skilled in the art will understand that many modifications, variations or equivalents can be made without departing from the spirit and scope defined by the appended claims, and all such modifications, variations or equivalents will fall within the protection scope of this utility model.

Claims

1. A smart height-adjustable desk based on voice recognition, characterized in that, It includes a desktop panel (1), linear guide columns (2), a concealed cable management channel (3), and a base bracket (4); the desktop panel (1) has a multi-spectral LED lighting unit (5) fixed in the center by a connecting rod, and symmetrical mounting holes for embedding centrifugal silent fan units (6) are provided on the left and right edges. The back is equipped with a concealed cable management channel (3) and a base bracket (4), which are supported by symmetrical linear guide columns (2); the linear guide columns (2) are equipped with a lifting drive module (7); the centrifugal silent fan unit ( 6) It is fixed in the mounting hole of the desktop panel (1) by a snap-on bracket; the base bracket (4) is provided with a control box (8), the control box (8) has a built-in core control module (9), the core control module (9) is based on the STM32F103ZET6 microcontroller, integrates a voice recognition unit (10) and a wireless communication unit (11), and is connected to the multispectral LED lighting unit (5), the centrifugal silent fan unit (6), and the lifting drive module (7) by wires; the wires are installed in the hidden cable management channel (3).

2. The intelligent height-adjustable desk based on voice recognition according to claim 1, characterized in that, The lifting drive module (7) consists of a dual-motor drive system (12) and a load sensor (13). The dual-motor drive system (12) is connected to the microcontroller's pin PA11 (CAN_RX) and pin PA12 (CAN_TX) via a CAN bus. The load sensor (13) is a strain gauge weighing module. Its output signal is conditioned by an AD623 instrumentation amplifier and then connected to the microcontroller's ADC1 channel 0 (pin PA0).

3. The intelligent height-adjustable desk based on voice recognition according to claim 1, characterized in that, The centrifugal silent fan unit (6) includes a small 12V DC cooling fan, which is connected to the TIM1 channel 1 (pin PA8) of the microcontroller through a PWM speed control circuit.

4. The intelligent height-adjustable desk based on voice recognition according to claim 1, characterized in that, The LED lighting unit (5) includes a digital dimming chip TLC59116, which achieves 16-bit precision brightness control by connecting to the microcontroller's I²C interface (pin PB6 / SCL, pin PB7 / SDA).

5. The intelligent height-adjustable desk based on voice recognition according to claim 1, characterized in that, The wireless communication unit (11) is a WiFi module based on the ESP8266 chip. It is connected to the microcontroller through the USART2 serial communication interface (pin PA2 / TX, pin PA3 / RX) and achieves full-duplex communication with the cloud platform through the MQTT protocol.

6. The intelligent height-adjustable desk based on voice recognition according to claim 1, characterized in that, The voice recognition unit (10) uses an LD3320 chip and is connected to the microcontroller via the USART3 interface (pins PB10 / RX and PB11 / TX).

7. The intelligent height-adjustable desk based on voice recognition according to claim 1, characterized in that, The hidden cable management channel (3) is located on the back of the desktop panel (1), with a depth of 10mm, and accommodates the motor power cable, LED power cable and sensor signal cable.

8. The intelligent height-adjustable desk based on voice recognition according to claim 1, characterized in that, The base bracket (4) is equipped with an EMI filter at the bottom. The input power AC 220V is filtered and distributed to each module in the control box (8).

Images