Passenger elevator capable of automatically calling elevator through voice control

Abstract

The invention discloses a voice-controlled automatic elevator calling passenger elevator, which comprises a voice acquisition module, a voice processing module, an elevator control module, an environment sensing module and a feedback module, wherein the voice acquisition module is used for acquiring an elevator calling voice instruction of a user; the voice processing module is connected with the voice acquisition module and is used for carrying out noise reduction, voiceprint verification and semantic analysis on the voice instruction; the elevator control module is connected with the voice processing module and used for controlling the elevator to run according to the analysis result; the environment sensing module is used for monitoring the environment noise level in real time and dynamically adjusting a voice recognition threshold value; the feedback module is used for providing voice and visual elevator calling feedback for the user, beam forming and voice separation technologies are adopted, the recognition rate is improved in a noisy environment, unauthorized use is prevented by combining voiceprint and face recognition, and voice instructions of multiple persons can be processed at the same time. The invention belongs to the field of elevators, and particularly relates to a voice-controlled passenger elevator capable of calling an elevator automatically.

Description

Technical Field

[0001] This invention belongs to the field of elevators, specifically referring to a passenger elevator with voice-controlled automatic call function. Background Technology

[0002] With the development of intelligent voice technology and the Internet of Things, elevator control systems are evolving towards contactless and intelligent operation. Traditional elevator calling methods mainly rely on physical buttons, which pose hygiene risks (especially during flu season) and are inconvenient to operate. Although some voice-activated elevator calling solutions exist, they still have the following shortcomings:

[0003] Insufficient recognition accuracy: The speech recognition rate is low in noisy environments and it is prone to false triggering;

[0004] Lack of authentication: Unable to distinguish between authorized and unauthorized users, posing a security risk;

[0005] Multi-person voice interference: When multiple people issue commands at the same time, the system has difficulty accurately identifying the target command;

[0006] Poor scene adaptability: The impact of different environmental noise on speech recognition thresholds is not considered.

[0007] For example, patent CN112850391A discloses an elevator voice call registration device, but it does not solve the problem of unregistered calls from temporary users such as delivery personnel. Patent CN202211543408.4 achieves elevator-home appliance linkage through a central control terminal, but it does not optimize the accuracy and response speed of voice calls. In addition, existing solutions mostly rely on cloud processing, which poses risks of delays and privacy leaks. Summary of the Invention

[0008] The present invention primarily addresses the aforementioned technical problems.

[0009] To solve the above problems, the technical solution adopted by the present invention is as follows: The passenger elevator with voice control and automatic call function proposed in this invention includes a voice acquisition module, a voice processing module, an elevator control module, an environmental perception module, and a feedback module:

[0010] The voice acquisition module is used to acquire the user's voice commands for calling the elevator;

[0011] The voice processing module is connected to the voice acquisition module and is used to perform noise reduction, voiceprint verification and semantic parsing on voice commands.

[0012] The elevator control module is connected to the voice processing module and is used to control the elevator operation based on the parsing results.

[0013] The environmental perception module is used to monitor the environmental noise level in real time and dynamically adjust the speech recognition threshold.

[0014] The feedback module is used to provide voice and visual elevator call feedback to the user.

[0015] Furthermore, the voice collection module includes a multi-microphone array and an infrared sensor:

[0016] The multi-microphone array supports beamforming technology and is used to directionally pick up the user's voice;

[0017] The infrared sensor is used to detect the user's approach and activate voice collection.

[0018] Furthermore, the voice processing module includes a voiceprint recognition unit, a voice separation unit, and an intention recognition unit:

[0019] The voiceprint recognition unit is used to verify the user's identity and permissions;

[0020] The voice separation unit uses a deep neural network to separate mixed voice streams;

[0021] The intention recognition unit is used to parse elevator control instructions such as "go to floor X" and "open the door".

[0022] Furthermore, the voice separation unit supports the blind source separation algorithm and can process at least 3 mixed voice instructions simultaneously.

[0023] Furthermore, the environmental perception module dynamically adjusts the recognition threshold by calculating the signal-to-noise ratio (SNR) in real time to:

[0024] Adopt the first loose threshold when SNR > 15dB;

[0025] Adopt the second standard threshold when 5dB < SNR ≤ 15dB;

[0026] Adopt the third strict threshold when SNR ≤ 5dB.

[0027] Furthermore, the elevator control module includes an instruction priority management unit and a safety isolation circuit:

[0028] The instruction priority management unit is used to handle conflicts when multiple people call the elevator simultaneously;

[0029] The safety isolation circuit prevents the voice system failure from affecting the safe operation of the elevator.

[0030] Furthermore, the feedback module includes an LED display screen and a voice synthesis unit:

[0031] The LED display screen is used to display the target floor and the elevator position;

[0032] The voice synthesis unit is used to broadcast call confirmation information and operation status.

[0033] Furthermore, it also includes an emergency response module for:

[0034] Automatically switches to traditional button mode when the voice system malfunctions;

[0035] In the event of a power outage, the backup power supply will be activated to maintain basic voice functions.

[0036] The beneficial effects achieved by the present invention using the above structure are as follows:

[0037] 1. The passenger elevator with voice-controlled automatic call proposed in this solution adopts non-contact operation: reducing contact with elevator buttons and reducing the risk of pathogen transmission.

[0038] 2. The passenger elevator with voice-controlled automatic call proposed in this solution adopts beamforming and voice separation technology, which improves the recognition rate by more than 40% in noisy environments. Combined with voiceprint and facial recognition, it prevents unauthorized use and can process voice commands from 3-5 people at the same time to avoid interference.

[0039] 3. The passenger elevator with voice-controlled automatic call proposed in this solution automatically adjusts the recognition threshold according to the ambient noise, thereby improving robustness. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the overall system architecture of the present invention;

[0041] Figure 2 This is a schematic diagram of the workflow of the present invention.

[0042] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0043] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0044] like Figure 1-2 As shown, the present invention proposes a voice-controlled automatic elevator for passengers, comprising:

[0045] Voice acquisition module: Employs a multi-microphone array, supports beamforming technology, and can directionally acquire user voice while suppressing ambient noise;

[0046] Voice processing module: includes: voiceprint recognition unit: verifies user identity and ensures that only authorized users can call the elevator;

[0047] Speech separation unit: Employs a deep neural network (DNN) to separate mixed speech, supporting simultaneous command recognition by multiple users;

[0048] Intent recognition unit: parses voice commands and extracts target floor information;

[0049] Environmental perception module: Real-time monitoring of environmental parameters such as noise level, temperature and humidity inside and outside the car, and dynamic adjustment of voice recognition threshold;

[0050] Elevator control module: Based on the output of the voice processing module, it generates elevator operation commands and controls the motor response;

[0051] Feedback module: Provides users with feedback on the elevator call status through an LED display and voice synthesis technology.

[0052] Workflow:

[0053] S1. Voice wake-up: When a user approaches the elevator, the infrared sensor or facial recognition module activates the voice acquisition system;

[0054] S2. Speech Acquisition and Preprocessing: Speech is acquired using a multi-microphone array, and noise reduction, framing, and feature extraction are performed.

[0055] S3. Voiceprint Verification: Compare the user's voiceprint with the pre-stored voiceprint database to verify identity and permissions;

[0056] S4. Speech Separation and Recognition: If multiple people are detected speaking at the same time, blind source separation technology is used to extract independent speech streams;

[0057] S5. Intent Analysis: Identify key commands such as "go to floor X" and "open the door" to extract the target floor;

[0058] S6. Dynamic threshold adjustment: Automatically adjusts speech recognition sensitivity based on ambient noise levels;

[0059] S7. Elevator Control: Sends the target floor command to the elevator main controller to schedule elevator operation;

[0060] S8. Feedback and Confirmation: Inform the user via voice and display screen that the elevator call was successful and the estimated waiting time.

[0061] The specific usage is as follows:

[0062] 1. Voice acquisition module: It adopts a ring-shaped 6-microphone array, supports 360° sound pickup, sampling rate of 16kHz, signal-to-noise ratio >65dB; it integrates an infrared temperature sensor (accuracy ±0.3℃), which can simultaneously detect the user's body temperature during the epidemic.

[0063] 2. Voice Processing Module: The main control chip uses Rockchip RK3588, which has an built-in NPU acceleration unit and supports 8-channel parallel voice processing; Voiceprint recognition uses an i-vector + x-vector hybrid model, with an equal error rate (EER) < 3%; Voice separation uses the Conv-TasNet network, and the separation accuracy is > 85% at a signal-to-noise ratio of -5dB.

[0064] 3. Elevator Control Interface: It uses RS485 communication, supports the Modbus protocol, and is compatible with mainstream elevator controllers; It has a safety isolation design to prevent the voice system failure from affecting the safe operation of the elevator.

[0065] Software Algorithms:

[0066] 1. Noise Suppression: It uses spectral subtraction combined with Wiener filtering, which has a good suppression effect on both steady-state and non-steady-state noises; The training dataset contains 20 common elevator environment noises (such as fan noise, alarm sound, etc.).

[0067] 2. Dynamic Threshold Adjustment: It defines three environmental modes: quiet, normal, and noisy; The recognition threshold is dynamically switched according to the real-time signal-to-noise ratio (SNR): SNR > 15dB: loose threshold, supporting far-field wake-up; 5dB < SNR ≤ 15dB: standard threshold; SNR ≤ 5dB: strict threshold, and the user needs to speak close to the microphone.

[0068] 3. Intent Understanding: It supports 10 dialect variants, covering Mandarin, Cantonese, etc.; Floor recognition uses a CRF + BiLSTM model, with an accuracy rate > 98%.

[0069] Typical Application Scenarios:

[0070] 1). Home Mode:

[0071] The resident passes the face recognition at the unit access control;

[0072] The system automatically wakes up the elevator voice module: "Hello, please say the target floor";

[0073] The user says "Go to the 25th floor", and after the system verifies the voiceprint:

[0074] Dispatch the elevator to the 1st floor;

[0075] Turn on the corridor lighting on the 25th floor in advance (if the smart home system is linked);

[0076] When the elevator arrives, there is a voice prompt: "The elevator has arrived. Please go to the 25th floor".

[0077] 2). Visitor Mode:

[0078] The visitor says "Call the elevator" in the elevator hall;

[0079] The system responded: "Please state the name of the resident you wish to visit."

[0080] After the visitor states their name, the system:

[0081] Check the list of visitors pre-stored by the resident;

[0082] Authorize temporary elevator access;

[0083] Automatically register the target floor.

[0084] 3) Multi-person elevator scenarios:

[0085] A says "Go to the 8th floor," and B simultaneously says "Go to the 15th floor."

[0086] The system separates the two audio streams and recognizes them separately.

[0087] Elevators are dispatched according to time sequence or priority (such as VIP users).

[0088] Security mechanisms:

[0089] 1) Prevent accidental touch: Facial or voiceprint verification is required before calling the elevator; the elevator will be locked for 1 minute after 3 consecutive failed recognition attempts.

[0090] 2) Anti-malicious calls: Repeated calls from the same user within 5 minutes require secondary confirmation; abnormal calls trigger background monitoring alarms.

[0091] Emergency measures: Automatically switch to traditional button mode in case of voice system failure; maintain basic voice function for 30 minutes by backup power in case of power failure.

[0092] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A passenger elevator with voice-controlled automatic call function, characterized in that, It includes a voice collection module, a voice processing module, an elevator control module, an environment perception module, and a feedback module: The voice collection module is used to collect the user's elevator call voice command; The voice processing module, connected to the voice collection module, is used to perform noise reduction, voiceprint verification, and semantic parsing on the voice command; The elevator control module, connected to the voice processing module, is used to control the elevator operation according to the parsing result; The environment perception module is used to monitor the environmental noise level in real time and dynamically adjust the voice recognition threshold; The feedback module is used to provide voice and visual elevator call feedback to the user.

2. A voice-controlled automatic elevator for passengers according to claim 1, characterized in that: The voice collection module includes a multi-microphone array and an infrared sensor: The multi-microphone array, supporting beamforming technology, is used to directionally pick up the user's voice; The infrared sensor is used to detect the user's approach and activate voice collection.

3. A voice-controlled automatic elevator for passengers according to claim 1, characterized in that: The voice processing module includes a voiceprint recognition unit, a voice separation unit, and an intention recognition unit: The voiceprint recognition unit is used to verify the user's identity and permission; The voice separation unit uses a deep neural network to separate the mixed voice stream; The intention recognition unit is used to parse elevator control commands such as "go to the Xth floor" and "open the door".

4. A voice-controlled automatic call passenger elevator according to claim 3, characterized in that: The voice separation unit supports the blind source separation algorithm and can process at least 3 mixed voice commands simultaneously.

5. A voice-controlled automatic elevator for passengers according to claim 1, characterized in that: The environment perception module dynamically adjusts the recognition threshold to: Adopt the first loose threshold when SNR > 15dB; Adopt the second standard threshold when 5dB < SNR ≤ 15dB; Adopt the third strict threshold when SNR ≤ 5dB.

6. A voice-controlled automatic call passenger elevator according to claim 1, characterized in that: The elevator control module includes an instruction priority management unit and a safety isolation circuit: The instruction priority management unit is used to handle the conflict of multiple people calling the elevator at the same time; The safety isolation circuit prevents the voice system failure from affecting the safe operation of the elevator.

7. A voice-controlled automatic call passenger elevator according to claim 1, characterized in that: The feedback module includes an LED display screen and a voice synthesis unit: The LED display screen is used to display the target floor and the elevator position; The voice synthesis unit is used to broadcast the elevator call confirmation information and the running status.

8. A voice-controlled automatic call passenger elevator according to any one of claims 1-7, characterized in that: It also includes an emergency processing module, which is used for: Automatically switching to the traditional button mode when the voice system fails; Enabling the backup power supply to maintain the basic voice function when power is off.

Images