Voice air conditioner control system and method compatible with multiple types

Abstract

The invention provides a voice air conditioner control system and method compatible with multiple types, and aims to solve the technical problem that an existing voice air conditioner system is difficult to give consideration to rapid multi-type adaptation and reliable voice interaction. The system comprises a voice module and an MCU controller, and the voice module stores a full-function voice recognition model and is in communication connection with the MCU controller. The method mainly comprises the steps that after the air conditioner is powered on, the voice module actively inquires function support information of the MCU controller and dynamically shields functions which are not supported by a host in a model of the voice module according to the information; and when a user voice instruction is recognized, judging and executing corresponding control logic according to the function support information. According to the invention, the same voice module can adapt to models with different functions, the accuracy and reliability of voice interaction are ensured while rapid adaptation is realized and the development cost is reduced, and the user experience is the same as that of a customized version.

Description

Technical Field

[0001] This invention relates to the field of air conditioning control technology, and in particular to a voice-controlled air conditioning system and method compatible with multiple models. Background Technology

[0002] Voice control technology, as an important human-computer interaction method in the field of smart home appliances, has made significant progress in the application of air conditioning products in recent years. Its functions have gradually upgraded from basic command recognition such as turning on / off and adjusting temperature to interactive experiences that support multi-turn dialogue and provide personalized responses. Currently, mainstream voice air conditioning control solutions in the industry mainly focus on optimizing the voice interaction experience of specific models, emphasizing the enhancement of user-friendliness between the user and the corresponding air conditioner through more natural semantic understanding and more human-like voice feedback mechanisms.

[0003] However, the aforementioned one-device-one-adapter voice customization mode has the following technical problems in practical applications:

[0004] On the one hand, because the function set of the voice module is deeply bound to the hardware functions of a specific air conditioner model, each new model with differentiated functions requires almost complete redevelopment or significant modification of the corresponding voice software based on the function list of that model. This process not only brings high R&D costs and long cycles, severely restricting the rapid iteration of product lines and market response speed, but also places a huge burden on subsequent maintenance, upgrades, and quality control.

[0005] On the other hand, to reduce the aforementioned customized development costs, a universal voice software capable of recognizing all potential air conditioning function commands was developed and directly applied to different models. However, when the user's voice command exceeds the actual functional range of the model, the system will still recognize it, potentially leading to invalid control or, more seriously, issuing a voice response that is inconsistent with the facts—for example, replying "Activating left and right airflow," when the model does not actually have this function. This situation, where the system can recognize commands but cannot execute them correctly or provides inaccurate feedback, negatively impacts the user experience and the product's professional image.

[0006] Therefore, how to construct a voice-controlled air conditioning system that is compatible with the functions of multiple models, enabling it to quickly adapt to different models with varying functions, while ensuring that the voice interaction behavior strictly matches the actual functional capabilities of any model to avoid recognition misalignment and feedback deviation, has become a technical challenge that urgently needs to be solved in this field. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a voice-controlled air conditioning system and method compatible with multiple models, which solves the problem that existing voice-controlled air conditioning systems struggle to simultaneously achieve rapid multi-model adaptation and reliable voice interaction.

[0008] On one hand, according to an embodiment of the present invention, a voice-controlled air conditioning system compatible with multiple models includes a voice module and an MCU controller;

[0009] The voice module is communicatively connected to the MCU controller;

[0010] The voice module is used to store a voice recognition model containing all air conditioning functions, and to send a function query command to the MCU controller after power-on; it is also used to block functions that are not supported in the stored voice recognition model according to the function support information returned by the MCU controller; and when a user voice command is recognized, it determines whether the corresponding function is supported according to the function support information and executes the corresponding control process.

[0011] The MCU controller is used to respond to the function query command and return the function support information supported by the device to the voice module.

[0012] Preferably, the voice module includes:

[0013] The query module is used to send the function query command and receive the function support information;

[0014] The tagging management module is used to mark unsupported functions as disabled based on the function support information.

[0015] The instruction processing module is used to generate a corresponding control instruction and send it to the MCU controller if the function corresponding to the user's voice instruction is not marked as disabled when the user's voice instruction is recognized.

[0016] Preferably, the instruction processing module is further configured to:

[0017] If the recognized user voice command corresponds to a special function, a control command is directly generated and sent to the MCU controller. The system then waits for the execution result feedback from the MCU controller and triggers the corresponding voice broadcast based on the execution result feedback. The special function refers to a function with different enumeration values ​​on different models.

[0018] Preferably, the voice module is connected to the MCU controller via serial communication. The function query command, function support information and control command are transmitted through a predefined serial protocol, which includes fields for indicating whether a function is supported.

[0019] Preferably, the serial port protocol also defines a field for transmitting specific enumeration values ​​for the special function.

[0020] Preferably, the speech recognition model is an offline speech model trained using a speech vocabulary list that includes all air conditioning functions.

[0021] Preferably, the voice module supports activation via a preset or user-defined wake word.

[0022] On the other hand, according to embodiments of the present invention, a voice-controlled air conditioner method compatible with multiple models is also provided, applied to a voice-controlled air conditioner control system compatible with multiple models as described in the above technical solution, the method comprising:

[0023] When the air conditioner is powered on, the voice module sends a function query command to the MCU controller and receives function support information returned by the MCU controller.

[0024] The voice module blocks functions that are not supported in the voice recognition model based on the function support information; when the voice module recognizes a user's voice command, it determines whether the corresponding function is supported based on the function support information and executes the corresponding control process.

[0025] Preferably, the step of determining whether the corresponding function is supported based on the function support information and executing the corresponding control flow includes:

[0026] If the function corresponding to the user's voice command is supported and is not a special function, the voice module retrieves the corresponding voice response and sends the corresponding control command to the MCU controller; if the function corresponding to the user's voice command is not supported, the voice module does not perform voice playback and does not send a command to the MCU controller.

[0027] Preferably, the step of determining whether the corresponding function is supported based on the function support information and executing the corresponding control flow further includes:

[0028] If the function corresponding to the user voice command is a special function, the voice module directly sends a control command to the MCU controller; the voice module waits for and receives the execution result feedback, which contains specific enumeration values, returned by the MCU controller after execution;

[0029] The voice module, based on the execution result feedback, broadcasts the voice response corresponding to the function of the specific enumeration value.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] By proactively querying the actual functional support information of the air conditioner's main unit upon power-up, the voice module dynamically and in real-time blocks any parts of its full-featured voice recognition model that are not supported. This achieves the universality of the voice module, enabling rapid adaptation to different models with varying functions without modifying hardware or software. This solves the problem of balancing universality and interactive reliability in voice systems. It not only significantly reduces the development costs and time required for multi-model adaptation and achieves standardized and platform-based deployment of the voice module, but also ensures that voice commands issued by users on any model are recognized, responded to, and fed back in accordance with the actual hardware functions of that model. This avoids interactive failures such as recognition without execution or inaccurate feedback, thus providing an accurate and reliable user experience while ensuring rapid product iteration. Attached Figure Description

[0032] Figure 1 A flowchart illustrating the steps of a voice-controlled air conditioner method compatible with multiple models provided by the present invention;

[0033] Figure 2 This is a schematic diagram of a voice-controlled air conditioning system compatible with multiple models, provided by the present invention. Detailed Implementation

[0034] The technical solutions of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0035] As described in the background section, existing voice-activated air conditioning systems face the challenge of balancing versatility, development efficiency, and interactive reliability. This invention aims to provide a solution that allows the same voice system to be adaptively applied to different air conditioning models with varying functions, while ensuring accurate and reliable interaction.

[0036] like Figure 1 and Figure 2 As shown in the figure, this embodiment of the invention proposes a voice-controlled air conditioning system compatible with multiple models, including a voice module and an MCU controller; the voice module is communicatively connected to the MCU controller; the voice module is used to store a voice recognition model containing all air conditioning functions, and send a function query command to the MCU controller after power-on; it is also used to mask functions not supported in the stored voice recognition model according to the function support information returned by the MCU controller; and when a user voice command is recognized, it determines whether the corresponding function is supported according to the function support information and executes the corresponding control process; the MCU controller is used to respond to the function query command and return the function support information supported by the device to the voice module.

[0037] The detailed working process of this embodiment is as follows: When the air conditioner is powered on, the voice module first sends a function query command to the MCU controller connected to it. After receiving the command, the MCU controller retrieves all the functions supported by its own hardware, generates a list of its own function support information, such as a status list indicating whether each function is supported or not, and returns it to the voice module. Upon receiving this information, the voice module immediately dynamically adjusts its internally stored full-function voice recognition model, logically masking the voice recognition entries corresponding to functions marked as unsupported by the MCU controller, making these entries invalid in subsequent recognition processes. At this point, the voice module has completed an adaptive configuration for the specific model, and its effective recognition range is precisely matched with the actual capabilities of the MCU controller.

[0038] After initialization, the system enters standby mode. When the user speaks a voice command, the voice module first recognizes it using its pre-adapted recognition model. Upon successful recognition, the voice module does not execute the command immediately. Instead, it first queries the function support information acquired and stored during initialization to determine if the function requested by the command is supported by the current MCU controller. If supported, the voice module performs two operations: first, it plays a confirmation voice message corresponding to the function to the user, such as "The temperature has been adjusted for you"; second, it sends an accurate control command to the MCU controller to drive it to perform the corresponding action. If unsupported, the voice module neither plays a voice message nor sends any command to the MCU controller. From the user's perspective, the air conditioner does not respond to the command, thus avoiding erroneous responses.

[0039] The technical solution proposed in this embodiment allows the same hardware and software version of the voice module to be activated on air conditioner models with different functions at any time. After power-on, the voice module uses an automatic query and adaptive masking mechanism to dynamically match its function set with the actual capabilities of the current host, eliminating the need to redevelop or flash different software for each new model. This significantly shortens the development cycle and reduces production costs and maintenance complexity. It fundamentally eliminates the problem of recognizing but failing to execute or providing erroneous feedback. On any model, the voice commands successfully triggered by the user are always functions supported by the hardware of that model, and the system's voice response and control actions remain consistent, providing users with a stable and reliable interactive experience.

[0040] The MCU controller is the main control unit in the air conditioner unit. It establishes a communication connection with the voice module through a physical serial port and exchanges data according to a predefined serial communication protocol. The voice module is an independent hardware module integrating an offline voice recognition chip, memory, and related peripheral circuits. Internally, it stores an offline voice recognition model trained using a voice vocabulary table containing all possible functions of the air conditioner. This model is pre-programmed to recognize all preset commands such as "cooling," "heating," "up and down airflow," "left and right airflow," "power mode," and "sleep mode." The voice module allows users to activate its voice monitoring function using preset wake-up words or user-defined wake-up words.

[0041] After the air conditioner is powered on, the voice module immediately begins initialization. Its internal query module sends a function query command to the MCU controller via serial port. Upon receiving this command, the MCU controller generates function support information based on the hardware configuration of the unit. This information, in the serial port protocol of this embodiment, includes a function attribute table, where a "supported" field is defined for each function. For example, a basic model might return: Cooling - Supported, Heating - Supported, Left / Right Swing - Not Supported, Power Mode - Supported. The MCU controller then replies this information table to the voice module via serial port.

[0042] Upon receiving this response, the voice module's tagging management module begins its work. By parsing the "supported" field in the protocol, it marks those functions identified as unsupported (such as "left and right airflow") as disabled in its internal voice recognition model. Under this masking, the relevant terms still exist in the model, but the instruction processing module will ignore the triggering of these terms in subsequent recognition. Through the above process, the effective recognition range of the voice module is perfectly aligned with the actual capabilities of the air conditioning unit.

[0043] After initialization, the user can activate the voice module by saying a wake word and then issue control commands. Once the voice module recognizes the user's voice command, it enters different processing flows depending on the command type:

[0044] For functions marked as disabled (unsupported), suppose a user says "Turn on left and right air swing" to the basic model mentioned above. After recognizing this command, the instruction processing module will first check the status of the marking management module and find that the left and right air swing function has been marked as disabled. At this time, the system will not make any voice announcements, nor will it send any control commands to the MCU controller. From the user's perspective, the air conditioner does not recognize this command, thus avoiding the erroneous feedback that it understands but cannot perform the action.

[0045] For supported and non-special routine functions, assuming the user says "Turn on the cooling," the instruction processing module checks and finds that the function is not disabled. The system immediately retrieves and plays the corresponding confirmation voice from the voice library, such as "The cooling has been turned on for you," and sends the corresponding control command to the MCU controller. The MCU controller then executes the cooling operation, achieving precise control.

[0046] Special functions refer to those functions that have different enumeration values ​​on different models. Suppose a high-end model A supports both vertical and horizontal oscillation (enumeration values ​​1 and 2), while a mid-range model B only supports vertical oscillation (enumeration value 1). However, the user might only say "Turn on oscillation" for both. When the voice module recognizes the "Turn on oscillation" command, the instruction processing module determines it's a special function. Instead of broadcasting the command, it directly generates a general oscillation control command and sends it to the MCU controller via serial port. Upon receiving the command, the MCU controller executes the default oscillation action preset for its model. For example, model A executes vertical oscillation, and model B also executes its unique vertical oscillation. After execution, it returns the actual executed enumeration value (e.g., model A responds with 1, model B responds with 1) to the voice module via the enumeration value field defined for special functions in the serial port protocol. The voice module's instruction processing module waits for and receives this execution result feedback, then triggers the broadcast and corresponding voice message based on the specific enumeration value, such as "Version 1 has been turned on for you." This ensures that no matter what the user says, the final voice feedback is strictly consistent with the actual actions performed by the hardware, thus solving the problem of feedback accuracy in models with multiple meanings of a single word.

[0047] The multi-model compatible voice-controlled air conditioning system proposed in this embodiment enables the same voice module to be adapted to different models with varying functions without software modification, achieving hardware universality and significantly reducing R&D costs and time. Simultaneously, by dynamically disabling unsupported functions based on host capabilities during initialization, incorrect identification and invalid responses are eliminated, ensuring reliable interaction. Furthermore, a "execute first, feedback later" strategy is adopted for special functions, ensuring precise matching between voice broadcast content and actual hardware actions, significantly improving the intelligence of the interaction and the consistency of the user experience.

[0048] Please refer to the reference. Figure 1The flowchart shown illustrates that, according to another embodiment of the present invention, a voice-controlled air conditioner method compatible with multiple models is also provided, applied to a voice-controlled air conditioner control system compatible with multiple models as described in the above technical solution. The method includes: powering on the air conditioner; the voice module sending a function query command to the MCU controller and receiving function support information returned by the MCU controller; the voice module disabling functions not supported in the voice recognition model based on the function support information; and when the voice module recognizes a user voice command, determining whether the corresponding function is supported based on the function support information and executing the corresponding control process.

[0049] After configuration, the system enters voice monitoring ready state. The user activates the voice module using a wake word and then issues control commands. Once the voice module recognizes the user's command, it enters the corresponding control flow:

[0050] For commonly supported functions: if the detected command, such as turning on the cooling function, is not disabled (i.e., supported), the voice module immediately retrieves the preset confirmation voice and announces, "Cooling has been turned on for you," while simultaneously sending the corresponding control command to the MCU controller via the serial port. The MCU controller executes the operation, completing a fast and accurate human-machine interaction.

[0051] For unsupported functions: If the detected command, such as activating left or right air swing, is disabled (i.e., unsupported), the voice module will neither make any voice announcements nor send any commands to the MCU controller. From the user's perspective, the air conditioner will not respond to the command, thus avoiding erroneous feedback from the system.

[0052] For special functions: If the recognized command corresponds to a special function, the voice module does not first broadcast it, but directly sends a general control command for that function to the MCU controller. Then, the voice module waits for and receives the execution result feedback from the MCU controller after executing the action. This feedback contains the specific enumeration value or status when this model executes this function. Finally, based on the received specific enumeration value, the voice module broadcasts the precisely corresponding voice. That is, control is executed first, and then the result is broadcast, ensuring that the content of the voice feedback matches the actual operation performed by the hardware on any model.

[0053] The control method provided in this embodiment dynamically integrates a general-purpose, full-function voice system with the specific hardware capabilities of air conditioning models through the aforementioned process. This allows a single control method to drive the system on a variety of products, achieving not only standardized and efficient development but also ensuring the accuracy, reliability, and consistency of voice interaction across all models. This fundamentally solves the problem of balancing adaptation costs and interaction quality in large-scale applications.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A voice-controlled air conditioning system compatible with multiple models, characterized in that, Includes a voice module and an MCU controller; The voice module is communicatively connected to the MCU controller; The voice module is used to store a voice recognition model containing all air conditioning functions and to send a function query command to the MCU controller after power-on. It is also used to mask functions that are not supported in the stored speech recognition model based on the function support information returned by the MCU controller; And when a user's voice command is recognized, the system determines whether the corresponding function is supported based on the function support information and executes the corresponding control process. The MCU controller is used to respond to the function query command and return the function support information supported by the device to the voice module.

2. The voice-controlled air conditioning system compatible with multiple models according to claim 1, characterized in that, The voice module includes: The query module is used to send the function query command and receive the function support information; The tagging management module is used to mark unsupported functions as disabled based on the function support information. The instruction processing module is used to generate a corresponding control instruction and send it to the MCU controller if the function corresponding to the user's voice instruction is not marked as disabled when the user's voice instruction is recognized.

3. The voice-controlled air conditioning system compatible with multiple models according to claim 2, characterized in that, The instruction processing module is also used for: If the recognized user voice command corresponds to a special function, a control command is directly generated and sent to the MCU controller. The system then waits for the execution result feedback from the MCU controller and triggers the corresponding voice broadcast based on the execution result feedback. The special function refers to a function with different enumeration values ​​on different models.

4. A voice-controlled air conditioning system compatible with multiple models according to any one of claims 1 to 3, characterized in that, The voice module is connected to the MCU controller via serial communication. The function query command, function support information and control command are transmitted through a predefined serial protocol, which includes fields for indicating whether a function is supported.

5. A voice-controlled air conditioning system compatible with multiple models according to claim 4, characterized in that, The serial port protocol also defines fields for transmitting specific enumeration values ​​for the special functions.

6. A voice-controlled air conditioning system compatible with multiple models according to claim 1, characterized in that, The speech recognition model is an offline speech model trained using a speech vocabulary list that includes all air conditioning functions.

7. A voice-controlled air conditioning system compatible with multiple models according to claim 1, characterized in that, The voice module supports activation via preset or user-defined wake words.

8. A voice-controlled air conditioner method compatible with multiple models, applied to a voice-controlled air conditioner control system compatible with multiple models as described in any one of claims 1 to 7, characterized in that, The method includes: When the air conditioner is powered on, the voice module sends a function query command to the MCU controller and receives function support information returned by the MCU controller. The voice module blocks functions that are not supported in the voice recognition model based on the function support information; when the voice module recognizes a user's voice command, it determines whether the corresponding function is supported based on the function support information and executes the corresponding control process.

9. A voice-controlled air conditioner method compatible with multiple models according to claim 8, characterized in that, The step of determining whether the corresponding function is supported based on the function support information and executing the corresponding control flow includes: If the function corresponding to the user's voice command is supported and is not a special function, the voice module retrieves the corresponding voice response and sends the corresponding control command to the MCU controller; if the function corresponding to the user's voice command is not supported, the voice module does not perform voice playback and does not send a command to the MCU controller.

10. A voice-controlled air conditioner method compatible with multiple models according to claim 8, characterized in that, The step of determining whether the corresponding function is supported based on the function support information and executing the corresponding control process further includes: if the function corresponding to the user voice command is a special function, the voice module directly sends a control command to the MCU controller; the voice module waits for and receives the execution result feedback containing the specific enumeration value returned by the MCU controller after execution; the voice module broadcasts the voice response corresponding to the function of the specific enumeration value based on the execution result feedback.

Images