Light control system based on voice recognition
By using a voice recognition-based lighting control system, combined with a microphone array and ambient light sensor, the system dynamically adjusts the brightness of the lights, solving the problem that existing systems cannot adjust the brightness under different lighting conditions, thus improving user experience and energy efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-31
- Publication Date
- 2026-04-07
AI Technical Summary
Existing voice-activated lighting control systems lack the ability to perceive environmental conditions, resulting in an inability to dynamically adjust the brightness of lights under different lighting environments, leading to energy waste and eye discomfort for users.
The lighting control system based on voice recognition includes a voice acquisition module, a sound source localization module, a user intent parsing module, a spatial model management module, an ambient light detection module, and a control strategy generation module. Through a microphone array, an ambient light sensor, and a wireless communication protocol, it automatically identifies the user's location and ambient light intensity, and dynamically adjusts the brightness of the lights.
It enables dynamic adjustment of light brightness based on real-time ambient light levels, avoiding eye strain from excessive light, reducing power consumption, and improving the user experience.
Smart Images
Figure CN121815519A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of lighting control technology, and in particular to a lighting control system based on voice recognition. Background Technology
[0002] With the rapid development of IoT and AI technologies, smart home systems have gradually entered ordinary households. Among them, voice interaction, due to its natural and convenient characteristics, has become one of the mainstream methods for controlling smart home devices. As a core component of the home environment, the voice control technology of lighting systems has attracted much attention.
[0003] Existing voice-activated lighting control systems lack the ability to perceive environmental conditions. Regardless of whether the ambient light is bright or dim, the system will only turn on the lights at a fixed brightness. During the day or when there is already sufficient ambient light indoors, turning on high-brightness lights not only wastes energy but also causes eye discomfort to users due to excessive light, affecting the user experience. Summary of the Invention
[0004] The purpose of this invention is to provide a voice recognition-based lighting control system that can dynamically adjust the brightness of the lights according to the real-time ambient light level, thereby avoiding eye discomfort caused by excessive light and improving the user experience.
[0005] To achieve the above objectives, the present invention provides a speech recognition-based lighting control system, including a speech acquisition module, a sound source localization module, a user intent parsing module, a spatial model management module, an ambient light detection module, a control strategy generation module, and a lighting control execution module;
[0006] The sound source localization module is connected to the voice acquisition module, the user intent parsing module is connected to the voice acquisition module, the control strategy generation module is connected to the sound source localization module, the user intent parsing module, the spatial model management module, and the ambient light detection module, respectively, and the lighting control execution module is connected to the control strategy generation module;
[0007] The voice acquisition module is used to acquire the user's voice control commands;
[0008] The sound source localization module is used to determine the first area where the sound source that issued the voice control command is located.
[0009] The user intent parsing module is used to recognize voice control commands and parse out the user intent;
[0010] The spatial model management module is used to store predefined indoor space area division information, as well as the adjacency relationship information between each area;
[0011] The ambient light detection module is used to detect the ambient light illuminance in each area in real time.
[0012] The control strategy generation module is used to generate brightness adjustment commands for the lamps based on user intent, first area information and ambient illuminance.
[0013] The lighting control execution module is used to send brightness adjustment commands to the corresponding lamps to control the lamps to adjust to the target brightness.
[0014] The user intent parsing module includes a speech recognition unit and a semantic understanding unit; the semantic understanding unit and the speech recognition unit are connected.
[0015] The speech recognition unit is used to recognize the text content of voice control commands;
[0016] The semantic understanding unit is used to parse the user's intent based on the text content of the recognized voice control command.
[0017] The user intent includes turning on the lights, turning off the lights, and the target area targeted by the voice control command.
[0018] The control strategy generation module includes a data receiving unit and a command generation unit; the command generation unit and the data receiving unit are connected.
[0019] The data receiving unit is used to receive user intent, first area information, and ambient light intensity.
[0020] The command generation unit is used to generate brightness adjustment commands for the lamps.
[0021] The command generation unit generates a brightness adjustment command for the lamps, which includes: when the parsed user intent is an on command without a specific area, the spatial model management module obtains at least one second area adjacent to the first area, and combines the ambient light intensity of the first and second areas to generate an on and brightness adjustment command for the lamps in the first and second areas.
[0022] The command generation unit generates a brightness adjustment command for the lamps, which further includes: when the parsed user intent contains a specific area reference, ignoring the first area, generating a control command for controlling the lamps within the specific area.
[0023] The command generation unit generates a brightness adjustment command for the lighting fixtures, which further includes generating a control command for controlling the lighting fixtures in all areas of the room when the parsed user intent is a global control instruction.
[0024] This invention discloses a voice recognition-based lighting control system. The voice acquisition module includes multiple microphone arrays deployed in different indoor areas to collect user voice control commands. In this embodiment, the microphone arrays are preferably four-element circular microphone arrays, embedded in the ceiling at the center of each functional area such as the living room, dining room, and bedroom, to ensure effective coverage of the entire area and high-quality voice signal acquisition. The sound source localization module processes the voice signals collected by the microphone arrays. By calculating the time difference of sound arrival at different microphones, it determines the first area where the sound source issuing the voice control command is located. Specifically, this embodiment uses the generalized cross-correlation function method to calculate the time difference, and then uses a triangulation algorithm to accurately calculate the horizontal coordinates of the sound source, matching them with a pre-stored spatial map to determine the user's current first area, such as the "living room." The user intent parsing module is used to recognize the text content of the voice control commands and parse the user intent. The module first performs noise reduction and enhancement processing on the voice signal, then converts the voice into text using an automatic speech recognition engine, such as converting the user's spoken "turn on the lights" into a text command. Subsequently, it uses natural language processing technology for semantic understanding to parse the core user intent (such as "turn on the lights" or "turn off the lights") and whether the command contains a specific target area (such as "living room" or "bedroom"). The space model management module stores predefined indoor space area division information and the adjacency relationship information between each area. During system initialization, the user needs to draw a floor plan using the accompanying mobile application and manually divide logical areas such as "living room," "dining room," and "corridor." The system automatically or the user manually defines the adjacency relationships between each area; for example, defining "living room" as adjacent to "dining room" and "corridor." The ambient light detection module includes ambient light sensors deployed in each area to detect the ambient light intensity in each area in real time. The ambient light sensors are preferably digital light intensity sensors, connected to the main controller via the I2C communication protocol, and report the current illuminance value of the area in real time.The control strategy generation module receives user intent, first area information, and ambient illuminance. When the parsed user intent is an on / off command without a specific area, it obtains at least one second area adjacent to the first area based on the spatial model management module, and generates on / off and brightness adjustment commands for the lamps in the first and second areas by combining the ambient illuminance of the first and second areas. When the parsed user intent contains a specific area, it ignores the first area and generates a control command for controlling the lamps in the specific area. When the parsed user intent is a global control command, it generates a control command for controlling the lamps in all areas of the room. The lighting control execution module sends the brightness adjustment command to the corresponding smart lamps. In this embodiment, the lighting control execution module sends control commands containing the target lamp device address, on / off status, and brightness percentage to the smart lamps in each area via the Zigbee wireless communication protocol to control them to turn on and adjust to the target brightness. This invention automatically determines the user's location using sound source localization technology and only turns on lights in the user's area and its adjacent areas. This completely avoids unnecessary activation of all lights in the house due to a single voice command, reducing power consumption at the source and aligning with the concept of green environmental protection. By introducing an ambient light detection mechanism, the system can dynamically adjust light brightness based on real-time ambient light levels. It provides sufficient illumination in dim environments and a softer brightness in environments with some existing light, avoiding sudden excessive brightness that could irritate the user's eyes and creating a more comfortable and healthy home lighting environment. This invention can dynamically adjust light brightness based on real-time ambient light levels, preventing eye discomfort caused by excessive light and improving the user experience. Attached Figure Description
[0025] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.
[0026] Figure 1 This is a schematic diagram of the overall structure of the present invention.
[0027] 1-Voice acquisition module, 2-Sound source localization module, 3-User intent parsing module, 4-Spatial model management module, 5-Ambient light detection module, 6-Control strategy generation module, 7-Lighting control execution module, 8-Voice recognition unit, 9-Semantic understanding unit, 10-Data receiving unit, 11-Command generation unit. Detailed Implementation
[0028] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, but should not be construed as limiting the present invention.
[0029] Please see Figure 1 ,in, Figure 1 This is a schematic diagram of the overall structure of the present invention.
[0030] This invention provides a voice recognition-based lighting control system, comprising a voice acquisition module 1, a sound source localization module 2, a user intent parsing module 3, a spatial model management module 4, an ambient light detection module 5, a control strategy generation module 6, and a lighting control execution module 7. The user intent parsing module 3 includes a voice recognition unit 8 and a semantic understanding unit 9. The control strategy generation module 6 includes a data receiving unit 10 and a command generation unit 11. Through the aforementioned scheme, the brightness of the lights can be dynamically adjusted according to the real-time ambient light intensity, avoiding eye discomfort caused by excessive light and improving the user experience.
[0031] In this specific embodiment, the sound source localization module 2 is connected to the voice acquisition module 1, the user intent parsing module 3 is connected to the voice acquisition module 1, the control strategy generation module 6 is connected to the sound source localization module 2, the user intent parsing module 3, the spatial model management module 4, and the ambient light detection module 5 respectively, and the lighting control execution module 7 is connected to the control strategy generation module 6;
[0032] The voice acquisition module 1 is used to acquire the user's voice control commands;
[0033] The sound source localization module 2 is used to determine the first area where the sound source that issued the voice control command is located.
[0034] The user intent parsing module 3 is used to recognize voice control commands and parse out the user intent;
[0035] The space model management module 4 is used to store predefined indoor space area division information and the adjacency relationship information between each area;
[0036] The ambient light detection module 5 is used to detect the ambient light illuminance of each area in real time.
[0037] The control strategy generation module 6 is used to generate brightness adjustment commands for the lamps based on user intent, first area information and ambient illuminance.
[0038] The lighting control execution module 7 is used to send brightness adjustment commands to the corresponding lamps to control the lamps to adjust to the target brightness.
[0039] In this embodiment, the voice acquisition module 1 includes multiple microphone arrays deployed in different areas of the room for acquiring the user's voice control commands. Preferably, the microphone arrays are four-element circular microphone arrays, embedded in the ceiling at the center of each functional area such as the living room, dining room, and bedroom, to ensure effective coverage of the entire area and acquisition of high-quality voice signals. The sound source localization module 2 processes the voice signals acquired by the microphone arrays. By calculating the time difference between the arrival times of the sound at different microphones, it determines the first area where the sound source issuing the voice control command is located. Specifically, this embodiment uses the generalized cross-correlation function method to calculate the time difference, and then uses a triangulation algorithm to accurately calculate the horizontal coordinates of the sound source, matching them with a pre-stored spatial map to determine the user's current first area, such as the "living room." The user intent parsing module 3 is used to recognize the text content of the voice control command and parse the user intent. The user intent parsing module 3 first performs noise reduction and enhancement processing on the voice signal, and then converts the voice into text through an automatic speech recognition engine, such as converting the user's spoken "turn on the lights" into a text command. Subsequently, semantic understanding is performed through natural language processing technology to parse out the core user intent (such as "turn on the lights" or "turn off the lights") and whether the command contains a specific target area (such as "living room" or "bedroom"). The space model management module 4 stores predefined indoor space area division information and the adjacency relationship information between each area. During the system initialization phase, the user needs to draw a floor plan through the accompanying mobile application and manually divide logical areas such as "living room", "dining room", and "corridor". The system automatically or the user manually defines the adjacency relationship between each area. For example, "living room" is defined as an adjacent area with "dining room" and "corridor". The ambient light detection module 5 includes ambient light sensors deployed in various areas for real-time detection of ambient light illuminance in each area. The ambient light sensors are preferably digital light intensity sensors, which are connected to the main controller via the I2C communication protocol and report the illuminance value of the current area in real time.The control strategy generation module 6 receives user intent, first area information, and ambient illuminance. When the parsed user intent is an on / off command without a specific area, it obtains at least one second area adjacent to the first area based on the spatial model management module 4, and generates on / off and brightness adjustment commands for the lamps in the first and second areas by combining the ambient illuminance of the first and second areas. When the parsed user intent contains a specific area, it ignores the first area and generates a control command for controlling the lamps in the specific area. When the parsed user intent is a global control command, it generates a control command for controlling the lamps in all areas of the room. The lighting control execution module 7 sends the brightness adjustment command to the corresponding smart lamps. In this embodiment, the lighting control execution module 7 sends control commands containing the target lamp device address, on / off status, and brightness percentage to the smart lamps in each area via the Zigbee wireless communication protocol to control them to turn on and adjust to the target brightness. This invention automatically determines the user's location using sound source localization technology and only turns on lights in the user's area and its adjacent areas. This completely avoids unnecessary activation of all lights in the house due to a single voice command, reducing power consumption at the source and aligning with the concept of green environmental protection. By introducing an ambient light detection mechanism, the system can dynamically adjust light brightness based on real-time ambient light levels. It provides sufficient illumination in dim environments and a softer brightness in environments with some existing light, avoiding sudden excessive brightness that could irritate the user's eyes and creating a more comfortable and healthy home lighting environment. This invention can dynamically adjust light brightness based on real-time ambient light levels, preventing eye discomfort caused by excessive light and improving the user experience.
[0040] Furthermore, the voice recognition unit 8 is used to recognize the text content of voice control commands;
[0041] The semantic understanding unit 9 is used to parse the user's intent based on the text content of the recognized voice control command. The user's intent includes turning on the lights, turning off the lights, and the target area targeted by the voice control command.
[0042] In this embodiment, the speech recognition unit 8 first performs noise reduction and enhancement processing on the speech signal, and then converts the speech into text through an automatic speech recognition engine, such as converting the user's spoken "turn on the lights" into a text command. Subsequently, the semantic understanding unit 9 performs semantic understanding through natural language processing technology, and parses out the core user intent (such as "turn on the lights" or "turn off the lights") and whether the command contains a clear target area (such as "living room" or "bedroom").
[0043] Secondly, the data receiving unit 10 is used to receive user intent, first area information, and ambient light intensity;
[0044] The command generation unit 11 is used to generate brightness adjustment commands for the lamps.
[0045] The command generation unit 11 generates brightness adjustment commands for lighting fixtures, including: when the parsed user intent is an on / off command without a specific area target, based on the spatial model management module 4, acquiring at least one second area adjacent to the first area, and combining the ambient illuminance of the first and second areas to generate on / off and brightness adjustment commands for lighting fixtures within the first and second areas. The command generation unit 11 also generates brightness adjustment commands for lighting fixtures, including: when the parsed user intent includes a specific area target, ignoring the first area, generating a control command for controlling lighting fixtures within that specific area. Furthermore, the command generation unit 11 generates brightness adjustment commands for lighting fixtures, including: when the parsed user intent is a global control command, generating a control command for controlling lighting fixtures in all areas of the room.
[0046] In this embodiment, the data receiving unit 10 receives user intent, first area information, and ambient illuminance. When the parsed user intent is an activation command without a specific area, the command generation unit 11 obtains at least one second area adjacent to the first area based on the spatial model management module 4, and generates activation and brightness adjustment commands for the lamps in the first and second areas by combining the ambient illuminance of the first and second areas. When the parsed user intent contains a specific area, the first area is ignored, and a control command for controlling the lamps in the specific area is generated. When the parsed user intent is a global control command, a control command for controlling the lamps in all areas of the room is generated.
[0047] This invention discloses a voice recognition-based lighting control system. The voice acquisition module 1 includes multiple microphone arrays deployed in different indoor areas to collect user voice control commands. In this embodiment, the microphone arrays are preferably four-element circular microphone arrays, embedded in the ceiling at the center of each functional area such as the living room, dining room, and bedroom, to ensure effective coverage of the entire area and high-quality voice signal acquisition. The sound source localization module 2 processes the voice signals collected by the microphone arrays. By calculating the time difference between the arrival times of the sound at different microphones, it determines the first area where the sound source issuing the voice control command is located. Specifically, this embodiment uses the generalized cross-correlation function method to calculate the time difference, and then uses a triangulation algorithm to accurately calculate the horizontal coordinates of the sound source, matching them with a pre-stored spatial map to determine the user's current first area, such as the "living room." The user intent parsing module 3 is used to recognize the text content of the voice control command and parse the user intent. The user intent parsing module 3 first performs noise reduction and enhancement processing on the voice signal, and then converts the voice into text through an automatic speech recognition engine, such as converting the user's spoken "turn on the lights" into a text command. Subsequently, semantic understanding is performed through natural language processing technology to parse out the core user intent (such as "turn on the lights" or "turn off the lights") and whether the command contains a specific target area (such as "living room" or "bedroom"). The space model management module 4 stores predefined indoor space area division information and the adjacency relationship information between each area. During the system initialization phase, the user needs to draw a floor plan through the accompanying mobile application and manually divide logical areas such as "living room", "dining room", and "corridor". The system automatically or the user manually defines the adjacency relationship between each area. For example, "living room" is defined as an adjacent area with "dining room" and "corridor". The ambient light detection module 5 includes ambient light sensors deployed in various areas for real-time detection of ambient light illuminance in each area. The ambient light sensors are preferably digital light intensity sensors, which are connected to the main controller via the I2C communication protocol and report the illuminance value of the current area in real time.The control strategy generation module 6 receives user intent, first area information, and ambient illuminance. When the parsed user intent is an on / off command without a specific area, it obtains at least one second area adjacent to the first area based on the spatial model management module 4, and generates on / off and brightness adjustment commands for the lamps in the first and second areas by combining the ambient illuminance of the first and second areas. When the parsed user intent contains a specific area, it ignores the first area and generates a control command for controlling the lamps in the specific area. When the parsed user intent is a global control command, it generates a control command for controlling the lamps in all areas of the room. The lighting control execution module 7 sends the brightness adjustment command to the corresponding smart lamps. In this embodiment, the lighting control execution module 7 sends control commands containing the target lamp device address, on / off status, and brightness percentage to the smart lamps in each area via the Zigbee wireless communication protocol to control them to turn on and adjust to the target brightness. This invention automatically determines the user's location using sound source localization technology and only turns on lights in the user's area and its adjacent areas. This completely avoids unnecessary activation of all lights in the house due to a single voice command, reducing power consumption at the source and aligning with the concept of green environmental protection. By introducing an ambient light detection mechanism, the system can dynamically adjust light brightness based on real-time ambient light levels. It provides sufficient illumination in dim environments and a softer brightness in environments with some existing light, avoiding sudden excessive brightness that could irritate the user's eyes and creating a more comfortable and healthy home lighting environment. This invention can dynamically adjust light brightness based on real-time ambient light levels, preventing eye discomfort caused by excessive light and improving the user experience.
[0048] The above-disclosed embodiments are merely one or more preferred embodiments of this application and should not be construed as limiting the scope of this application. Those skilled in the art can understand that all or part of the processes for implementing the above embodiments and equivalent changes made in accordance with the claims of this application still fall within the scope of this application.
Claims
1. A voice recognition-based lighting control system, characterized in that, It includes a voice acquisition module, a sound source localization module, a user intent parsing module, a spatial model management module, an ambient light detection module, a control strategy generation module, and a lighting control execution module; The sound source localization module is connected to the voice acquisition module, the user intent parsing module is connected to the voice acquisition module, the control strategy generation module is connected to the sound source localization module, the user intent parsing module, the spatial model management module, and the ambient light detection module, respectively, and the lighting control execution module is connected to the control strategy generation module; The voice acquisition module is used to acquire the user's voice control commands; The sound source localization module is used to determine the first area where the sound source that issued the voice control command is located. The user intent parsing module is used to recognize voice control commands and parse out the user intent; The spatial model management module is used to store predefined indoor space area division information, as well as the adjacency relationship information between each area; The ambient light detection module is used to detect the ambient light illuminance in each area in real time. The control strategy generation module is used to generate brightness adjustment commands for the lamps based on user intent, first area information and ambient illuminance. The lighting control execution module is used to send brightness adjustment commands to the corresponding lamps to control the lamps to adjust to the target brightness.
2. The voice recognition-based lighting control system as described in claim 1, characterized in that, The user intent parsing module includes a speech recognition unit and a semantic understanding unit; the semantic understanding unit and the speech recognition unit are connected. The speech recognition unit is used to recognize the text content of voice control commands; The semantic understanding unit is used to parse the user's intent based on the text content of the recognized voice control command.
3. The voice recognition-based lighting control system as described in claim 2, characterized in that, The user intent includes turning lights on, turning lights off, and the target area targeted by voice control commands.
4. The voice recognition-based lighting control system as described in claim 3, characterized in that, The control strategy generation module includes a data receiving unit and a command generation unit; the command generation unit and the data receiving unit are connected. The data receiving unit is used to receive user intent, first area information, and ambient light intensity. The command generation unit is used to generate brightness adjustment commands for the lamps.
5. The voice recognition-based lighting control system as described in claim 4, characterized in that, The command generation unit generates brightness adjustment commands for lamps, including: when the parsed user intent is an on command without a specific area, based on the spatial model management module, obtaining at least one second area adjacent to the first area, and combining the ambient light intensity of the first and second areas, generating on and brightness adjustment commands for lamps in the first and second areas.
6. The voice recognition-based lighting control system as described in claim 5, characterized in that, The command generation unit also generates a brightness adjustment command for the lamps, which further includes: when the parsed user intent contains a specific area reference, ignoring the first area, generating a control command for controlling the lamps within the specific area.
7. The voice recognition-based lighting control system as described in claim 6, characterized in that, The command generation unit also generates brightness adjustment commands for lighting fixtures, including: when the parsed user intent is a global control instruction, generating control commands for controlling lighting fixtures in all indoor areas.