Intelligent voice remote control

Abstract

The present invention discloses an intelligent voice remote controller. After the user issues an adaptive adjustment instruction, the present invention adjusts the state of the voice remote controller to an adaptive state. In the adaptive state, the application scene is divided into functional sub-areas, the optimum temperature range is determined in combination with season information, the temperature adjustment values ​​of different functional sub-areas are calculated, and the air quality index corresponding to the air pollutant content is calculated according to sensor data. The air conditioning equipment parameters are adjusted according to the temperature and regional information, and the air purification equipment power and window switch state are adjusted according to the air quality index, thereby realizing active perception and adaptive adjustment of the environment and user needs.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent remote controllers, and in particular to an intelligent voice remote controller. Background Art

[0002] With the rapid development of science and technology, the concept of smart home has gradually gained popularity, and people's demand for intelligent control of home appliances is growing; traditional remote controls often have single functions, and the operation method is limited to button operation. They lack flexibility and convenience and are difficult to meet the diverse needs of modern life.

[0003] Although the existing intelligent voice remote control has improved the convenience of operation to a certain extent, it still has the following shortcomings:

[0004] The existing intelligent voice remote controller has insufficient intelligent judgment ability and can only passively execute user commands, lacking the ability to actively perceive and adaptively adjust the environment and user needs;

[0005] In addition, when command recognition fails and no feedback is given to the command, the processing capability of the smart voice remote control in the prior art is poor and the problem cannot be solved promptly and effectively, thus affecting the user experience.

[0006] To this end, an intelligent voice remote control is launched. Summary of the invention

[0007] In view of this, the present invention provides an intelligent voice remote controller to solve the problems raised by the above background technology.

[0008] The object of the present invention can be achieved by the following technical solutions: an intelligent voice remote controller, comprising a control component, wherein the control component comprises a setting module, a use module, an intelligent module and an abnormal module;

[0009] Setting module: pair the voice remote control with the device to be controlled in the current application scenario;

[0010] Usage module: When the user uses the voice remote control at the current time point, the user activates the voice function of the voice remote control and issues a command through the preset wake-up word. The voice remote control recognizes the current command and executes it according to the command content. If the user command recognition fails during this process, the corresponding adjustment steps are executed;

[0011] Intelligent module: When the user issues an adaptive adjustment command, the state of the voice remote controller is adjusted to the adaptive state. Before the user issues a corresponding termination command to terminate the adaptive state of the voice remote controller, the device parameters in the current application scenario are adaptively adjusted through the voice remote controller;

[0012] Abnormal module: During the use of the voice remote control, if a command issued at a certain time does not receive feedback, the retrieval signaling is triggered and the corresponding processing steps are executed.

[0013] In some embodiments, if the user instruction recognition fails, corresponding adjustment steps are performed, specifically:

[0014] Convert the user's command into text, and input the converted text content into a pre-built historical database for matching; extract the text content corresponding to the historical voice command from the historical database;

[0015] Divide the converted text content into multiple groups of words; match the divided multiple groups of words with a preset basic word set, and remove the words that are successfully matched; and calculate the similarity between the remaining words and the words corresponding to the historical voice instructions, convert the remaining words and the words corresponding to the historical voice instructions into vector representations respectively, and calculate the cosine similarity between each group of vectors;

[0016] Determine the part of speech of each of the remaining words, preset a similarity value matrix corresponding to the similarities between different parts of speech, combine the parts of speech of each of the remaining words with the words corresponding to the historical voice instructions and input them into the matrix, set different part of speech combinations to correspond to a part of speech similarity, thereby obtaining the part of speech similarity between each of the remaining words and the words corresponding to the historical voice instructions.

[0017] In some embodiments, if the user instruction recognition fails, the corresponding adjustment steps are executed, further comprising:

[0018] The weight coefficients corresponding to the cosine similarity and the part-of-speech similarity are set respectively, and the cosine similarity and the part-of-speech similarity of each group of the remaining words are multiplied by the corresponding weight coefficients respectively, and then the sum is calculated to obtain the single group similarity between each word in the remaining words and the word corresponding to the historical voice instruction; the single group similarities are averaged to obtain the overall similarity between the historical voice instruction and the current remaining words;

[0019] Historical voice commands are sorted according to overall similarity, and the historical voice commands with the highest similarity are selected as possible commands, which are verified with the user through the voice remote control; if the verification fails, the parameters of voice recognition are adjusted based on preset adjustment rules, and a voice recognition failure message is popped up to the user to remind the user to issue the command again and use the adjusted parameters for voice recognition.

[0020] In some embodiments, the device parameters in the current application scenario are adaptively adjusted through the voice remote control, specifically:

[0021] T1: Read the sensor data pre-deployed in the current application scenario; the deployed sensors include temperature sensors, air quality sensors, and human body sensing sensors; the sensor data includes temperature and air pollutant content;

[0022] T2: Divide the current application scenario into functional sub-areas; obtain the season information at the current time point and determine the optimal temperature range; match the read temperature value with the determined optimal temperature range. If the match fails, take the middle value of the optimal temperature range as the adjustment target value, obtain the functional sub-area corresponding to the current user position through the human body sensing sensor, and set the temperature adjustment value corresponding to different functional sub-areas; add the temperature adjustment value of the functional sub-area and the adjustment target value, and use the accumulated result as the air-conditioning equipment adjustment value at the current time point. Based on the air-conditioning equipment adjustment value at the current time point, adjust the air-conditioning equipment parameters of the current application scenario through the voice remote control.

[0023] In some embodiments, adaptively adjusting device parameters in the current application scenario through the voice remote controller further includes:

[0024] T3: Divide the content of air pollutants according to the types of air pollutants to obtain the content of various types of air pollutants. Based on the number of deployment points of air quality sensors corresponding to the current application scenario, average the content of the same type of air pollutants at different deployment points to obtain the mean value of the content of various types of air pollutants in the current application scenario; the type of air pollutant is represented by i, where i=1,2......,h, h is the total number of air pollutant types, and the reference thresholds corresponding to the content of different types of air pollutants are preset; the mean value of the content of various types of air pollutants is marked as Ki, and the reference thresholds corresponding to the content of different types of air pollutants are marked as Ui; using the formula The average value Ki of the content of various types of air pollutants is weighted to obtain the air quality index Y of the current application scenario; where ai is the preset impact weight factor corresponding to different types of air pollutants;

[0025] A threshold range corresponding to the air quality index Y is preset. If the air quality index Y of the current application scenario is within the preset threshold range, the air purification device is turned on by the voice remote control, and a power range of the air purification device corresponding to the threshold range is set. After the air purification device is turned on, the power of the air purification device is adjusted to the corresponding power range;

[0026] If the air quality index Y of the current application scenario is higher than the preset threshold range, a power range of an air purification device corresponding to the range above the threshold is set, and the power of the air purification device is adjusted to the corresponding power range after the air purification device is turned on by the voice remote control. After the adjustment is completed, the air quality index Yt of each layout point is calculated, where t represents the number of each layout point, where t=1,2,..., g,g is the total number of layout points; the layout point corresponding to the highest air quality index Yt is marked, and the window closest to the marked layout point is opened by the voice remote control.

[0027] In some embodiments, adaptively adjusting device parameters in the current application scenario through the voice remote controller further includes:

[0028] Steps T1 to T3, when the voice remote controller is in an adaptive state, based on a preset adjustment time interval, after reaching the preset adjustment time interval, the device parameters in the current application scenario are adaptively adjusted.

[0029] In some embodiments, the retrieval signaling is triggered and corresponding processing steps are performed, specifically:

[0030] P1: Identify the specific device that does not provide feedback, and analyze the network connection and pairing status;

[0031] P1-101: Send a network detection data packet to the connected network device; if no response is received, it is determined that the network connection is abnormal;

[0032] P1-102: Check the pairing status between the specific device and the voice remote control that does not provide feedback. If the pairing status changes from paired to unpaired, it is considered a pairing abnormality.

[0033] P2: If the analysis result of the network connection and pairing status is abnormal, send the same command again after the set interval; if the device still does not respond, execute step P3;

[0034] P3: Cut off the power of the corresponding device that has not responded through the voice remote control, wait for the set time and then reconnect the power of the device to restart. After restarting, send the control command again. If the device still has no feedback, execute step P4;

[0035] P4: Randomly select a device in the current application scenario to send a random command. If the selected device responds normally and gives feedback, the corresponding device that does not give feedback is marked as a faulty device and sent to the user's mobile terminal; if the selected device also does not give feedback, the self-check signaling of the voice remote control is triggered, and step P5 is executed;

[0036] P5: After the self-check signaling is triggered, the preset self-diagnosis program is started; according to the result of the self-diagnosis program, if a fault is displayed, the fault information is sent to the user's mobile terminal.

[0037] In some embodiments, if it is displayed in step P5 that there is no fault, a pop-up window "Do you need to evaluate the recent performance of the voice remote control?" is displayed on the user's mobile terminal. After the user confirms, a comprehensive evaluation is performed on the performance parameters of the voice remote control within the set time period. Based on the results of the comprehensive evaluation, the performance level of the voice remote control within the set time period is determined and sent to the user's mobile terminal; the performance parameters include the command sending success rate, voice recognition accuracy and response efficiency.

[0038] In some embodiments, the performance parameters of the voice remote control within a set time period are comprehensively evaluated, specifically:

[0039] Count the total number of commands issued by the voice remote control within a set time period, and obtain the feedback results of the device corresponding to each command. If the feedback result matches the command, mark the command as a successful command; if the feedback result does not match the command, mark the command as a failed command; parse the failed commands. If a set number of failed commands are concentrated on a certain group of devices, extract the number of failed commands of the corresponding device from the total number of failed commands and round it to an integer of one;

[0040] Then the success rate of voice remote control command sending within the set time period is F1, and the formula Calculated, where b1 represents the total number of instructions issued, b2 represents the total number of successful instructions, and b3 represents the number of failed instructions for the corresponding device;

[0041] Count the total number of user commands received by the voice remote control within a set time period, extract the total number of commands successfully recognized by the voice remote control at one time, and calculate the proportion of the total number of commands successfully recognized at one time in the total number of user commands received to obtain the voice recognition accuracy F2 of the voice remote control within the set time period;

[0042] Record the time interval from when each command is sent from the voice remote control to when the device feedback is received within the set time period, mark the time interval as the command response time, calculate the average of the command response time of each command within the set time period, and use the calculated average as the response efficiency F3 of the voice remote control within the set time period;

[0043] According to the formula , comprehensively calculate the command sending success rate F1, voice recognition accuracy F2 and response efficiency F3 of the voice remote control within the set time period, and obtain the performance evaluation index Fr of the voice remote control within the set time period; wherein w1, w2 and w3 represent the influencing weight factors of the command sending success rate F1, the voice recognition accuracy F2 and the response efficiency F3 respectively;

[0044] The preset performance evaluation index Fr corresponds to each group of index value ranges, and each group of index value ranges corresponds to a performance level; the performance evaluation index Fr of the voice remote control within the set time period is matched with the preset groups of index value ranges to obtain the performance level of the voice remote control within the set time period.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] The present invention adjusts the state of the voice remote control to an adaptive state after the user issues an adaptive adjustment instruction. In the adaptive state, the application scene is divided into functional sub-areas, the optimal temperature range is determined in combination with season information, the temperature adjustment values ​​of different functional sub-areas are calculated, and the air quality index corresponding to the air pollutant content is calculated according to sensor data. The air conditioning equipment parameters are adjusted according to the temperature and regional information, and the air purification equipment power and window switch state are adjusted according to the air quality index, thereby realizing active perception and adaptive adjustment of the environment and user needs.

[0047] The present invention improves the ability of the voice remote control to handle abnormal situations by identifying the specific device that has not given feedback, analyzing the network connection and pairing status, and taking a series of measures such as resending instructions, restarting the device, and marking the faulty device if abnormal.

[0048] The present invention determines the performance level by evaluating the performance of the remote control. The evaluation indicators include the success rate of command sending, the accuracy rate of voice recognition and the response efficiency. The total number of commands, the number of successful commands, the total number of user commands received, the number of commands successfully recognized at one time and the command response time data are counted, and the performance evaluation index is comprehensively calculated. Finally, the performance level is obtained by matching with the preset index value range, thereby reflecting the recent performance of the voice remote control.

[0049] The present invention sorts and verifies possible instructions through text conversion, matching with a historical database, word division, similarity calculation and comprehensive evaluation. If the verification fails, the voice recognition parameters can be adjusted to re-recognize the instructions, thereby improving the processing capability of the voice remote control. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Further details, features and advantages of the present application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0051] Figure 1 It is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0052] Several embodiments of the present application will be described in more detail below with reference to the accompanying drawings so that those skilled in the art can implement the present application. The present application can be embodied in many different forms and purposes and should not be limited to the embodiments described herein. These embodiments are provided to make the present application comprehensive and complete, and to fully convey the scope of the present application to those skilled in the art. The embodiments do not limit the present application.

[0053] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the relevant art and / or the context of this specification, and will not be interpreted in an idealized or overly formal sense unless explicitly defined as such herein.

[0054] See also Figure 1 As shown, an intelligent voice remote controller includes a control component, wherein the control component includes a setting module, a use module, an intelligent module and an abnormal module;

[0055] The setting module is used to pair the voice remote control with the device to be controlled in the current application scenario; the pairing connection is performed via Bluetooth, Wi-Fi or a specific wireless communication protocol, and the devices include but are not limited to TVs, air conditioners, and cameras, etc.;

[0056] Ability to quickly identify and connect various devices, providing a basis for subsequent control operations;

[0057] The use module is used to activate the voice function of the voice remote control and issue a command through a preset wake-up word when the user uses the voice remote control at the current time point. The voice remote control recognizes the current command and executes it according to the command content. If the user command recognition fails during this process, the corresponding adjustment steps are executed;

[0058] Specifically:

[0059] The user's command is converted into text using speech-to-text technology, and the converted text content is input into a pre-built historical database for matching; the historical database includes the voice commands issued by the user within a set time period, and is continuously updated and deleted; the text content corresponding to the historical voice commands is extracted from the historical database;

[0060] Use the word segmentation technology in natural language processing to divide the converted text content into multiple groups of words; match the multiple groups of divided words with a preset basic word set, and eliminate the words that match successfully; preset a basic word set, which contains some common and highly general words, such as "de", "ba", "yidian", "yixie", etc., which appear in many instructions but have little impact on the specific meaning of the instructions; and calculate the similarity between the remaining words and the words corresponding to the historical voice instructions, use the pre-trained word vector model to convert the remaining words and the words corresponding to the historical voice instructions into vector representations respectively, and calculate the cosine similarity between each group of vectors;

[0061] Use the cosine similarity formula to calculate the cosine value of the angle between the converted content vector and the historical content vector; the closer the cosine value is to 1, the closer the directions of the two vectors are, that is, the higher the similarity degree of the text content in the semantic space;

[0062] Determine the part of speech of each word in the remaining words, preset a similarity value matrix corresponding to the similarity between different parts of speech, combine the part of speech of each word in the remaining words with the words corresponding to the historical voice instructions and input them into the matrix, and set that each different part of speech combination corresponds to a part of speech similarity, so as to obtain the part of speech similarity between each word in the remaining words and the words corresponding to the historical voice instructions;

[0063] The initial similarity value between nouns can be set to 0.8, between verbs to 0.7, between a noun and a verb to 0.5, etc. This initial value matrix can be adjusted and optimized according to the actual situation;

[0064] Set the weight coefficients corresponding to the cosine similarity and the part of speech similarity respectively, multiply the corresponding cosine similarity and part of speech similarity of each group in the remaining words by the corresponding set weight coefficients respectively, and then sum to obtain the single-group similarity between each word in the remaining words and the words corresponding to the historical voice instructions; take the average of each single-group similarity to obtain the overall similarity between the historical voice instruction and the current remaining words;

[0065] Sort the historical voice instructions according to the overall similarity, select the historical voice instruction with the highest similarity as the possible instruction, and verify it with the user through the voice remote control;

[0066] Suppose the historical voice instruction "Turn on the light in the living room" has the highest similarity among all historical instructions; the voice remote control verifies with the user, such as sending the voice "Do you want to turn on the light in the living room?" If the user confirms, perform the corresponding operation;

[0067] If the verification fails, the parameters of speech recognition are adjusted based on the preset adjustment rules, including but not limited to microphone sensitivity gain, increasing noise reduction parameters, and adjusting echo cancellation parameters; and a pop-up message indicating speech recognition failure is displayed to the user, reminding the user to issue the command again, and use the adjusted parameters for speech recognition;

[0068] For example, "We are sorry that we failed to accurately recognize your command. Please give the command again." This will let the user know immediately that the recognition failed and guide the user to try again;

[0069] The intelligent module is used for the user to issue adaptive adjustment instructions; the user can trigger the adaptive state through a specific voice command, such as "turn on adaptive mode". When the voice remote controller recognizes this command, the adaptive state is immediately started; and the state of the voice remote controller is adjusted to the adaptive state. Before the user issues a corresponding abort command to abort the adaptive state of the voice remote controller, the device parameters in the current application scenario are adaptively adjusted through the voice remote controller;

[0070] Specifically:

[0071] T1: Read the sensor data pre-deployed in the current application scenario; the deployed sensors include temperature sensors, air quality sensors, and human body sensing sensors; the sensor data includes temperature and air pollutant content;

[0072] T2: Divide the current application scenario into various functional sub-areas, including living room, bedroom and kitchen, etc. The application scenario is divided into different functional sub-areas through pre-configured house layout information or manual settings by the user; obtain the season information of the current time point and determine the optimal temperature range; different seasons have different optimal temperature ranges, which are determined according to general human comfort and energy-saving principles; match the read temperature value with the determined optimal temperature range, if the match fails, take the middle value of the optimal temperature range as the adjustment target value, obtain the functional sub-area corresponding to the current user position through the human body sensing sensor, and set the temperature adjustment value corresponding to different functional sub-areas; add the temperature adjustment value of the functional sub-area and the adjustment target value, and use the accumulated result as the air-conditioning equipment adjustment value at the current time point. Based on the air-conditioning equipment adjustment value at the current time point, adjust the air-conditioning equipment parameters of the current application scenario through the voice remote control;

[0073] For example, if the temperature adjustment value of the kitchen is a negative value, the temperature adjustment value of the kitchen is accumulated with the adjustment target value as the temperature adjustment value; if the temperature adjustment value of the bedroom is a positive value, the temperature adjustment value of the bedroom is accumulated with the adjustment target value as the temperature adjustment value;

[0074] Users can set specific temperature adjustment values ​​for different functional sub-areas according to their preferences;

[0075] By dividing the application scenarios into different functional sub-areas and combining seasonal information, user location and user-defined temperature adjustment values, the most suitable temperature can be provided for each area. For example, in the bedroom, the user can adjust the temperature slightly higher according to their preferences to create a comfortable sleeping environment; while in the kitchen, since cooking activities generate heat, the temperature adjustment value can be set to a negative value to keep the kitchen relatively cool. This personalized temperature adjustment can greatly improve the user's comfort in different areas.

[0076] T3: Divide the content of air pollutants according to the types of air pollutants to obtain the content of various types of air pollutants. Based on the number of deployment points of air quality sensors corresponding to the current application scenario, average the content of the same type of air pollutants at different deployment points to obtain the mean value of the content of various types of air pollutants in the current application scenario; the type of air pollutant is represented by i, where i=1,2......,h, h is the total number of air pollutant types, and the reference thresholds corresponding to the content of different types of air pollutants are preset; the corresponding reference thresholds are preset based on user needs and health status; the mean value of the content of various types of air pollutants is marked as Ki, and the reference thresholds corresponding to the content of different types of air pollutants are marked as Ui; using the formula The average value Ki of the content of various types of air pollutants is weighted to obtain the air quality index Y of the current application scenario; where ai is the preset impact weight factor corresponding to different types of air pollutants;

[0077] A threshold range corresponding to the air quality index Y is preset. If the air quality index Y of the current application scenario is within the preset threshold range, the air purification device is turned on by the voice remote control, and a power range of the air purification device corresponding to the threshold range is set. After the air purification device is turned on, the power of the air purification device is adjusted to the corresponding power range;

[0078] If the air quality index Y of the current application scenario is higher than the preset threshold range, a power range of an air purification device corresponding to the range above the threshold is set, and the power of the air purification device is adjusted to the corresponding power range after the air purification device is turned on by the voice remote control. After the adjustment is completed, the air quality index Yt of each layout point is calculated, where t represents the number of each layout point, where t=1,2,..., g,g is the total number of layout points; the layout point corresponding to the highest air quality index Yt is marked, and the window closest to the marked layout point is opened by the voice remote control;

[0079] If the window is not a smart window, a ventilation signal is sent to the user's mobile terminal via the voice remote control to notify the user to go to the location of the window to open it;

[0080] Monitor and classify the content of air pollutants, calculate the air quality index based on the average content of different types of pollutants, and adjust the power of the air purification equipment according to the index; this ensures that the indoor air always maintains a good quality level and reduces the impact of air pollution on user health. In this process, the air quality index of different layout points is targeted and differentiated, and the windows closest to the corresponding layout points are opened according to the results of the differentiation, thereby improving ventilation quality and purification efficiency;

[0081] T4: Steps T1 to T3, when the voice remote controller is in an adaptive state, based on a preset adjustment time interval, after reaching the preset adjustment time interval, adaptively adjust the device parameters in the current application scenario;

[0082] The exception module is used to trigger the retrieval signaling and execute the corresponding processing steps if a command issued once does not receive feedback during the use of the voice remote control;

[0083] Specifically:

[0084] P1: Identify the specific device that does not provide feedback, and analyze the network connection and pairing status;

[0085] P1-101: Send a network detection data packet to the connected network device; the network detection data packet is the Ping command, and the network device is a router or other; if no response is received, it is determined that the network connection is abnormal;

[0086] P1-102: Check the pairing status between the specific device and the voice remote control that does not provide feedback. If the pairing status changes from paired to unpaired, it is considered a pairing abnormality.

[0087] P2: If the analysis result of the network connection and pairing status is abnormal, send the same command again after a set time interval; the set time is 5s, for example; if the device still does not respond, execute step P3;

[0088] P3: Use the voice remote control to cut off the power of the corresponding device that has not responded, and wait for the set time before reconnecting the power of the device to restart. After restarting, send the control command again. If the device still has no feedback, execute step P4; the waiting time is set to 30s;

[0089] P4: Randomly select a device in the current application scenario to send a random command. If the selected device responds normally and gives feedback, the corresponding device that does not give feedback is marked as a faulty device and sent to the user's mobile terminal; if the selected device also does not give feedback, the self-check signaling of the voice remote control is triggered, and step P5 is executed;

[0090] P5: After the self-check signaling is triggered, the preset self-diagnosis program is started; the self-diagnosis program includes checking the hardware functions (such as whether the microphone, speaker, buttons, etc. are working properly) and the software status (such as whether the operating system is running normally and whether the application has errors); according to the results of the self-diagnosis program, if a fault is displayed, the fault information is sent to the user's mobile terminal;

[0091] If it is displayed that there is no fault, a pop-up window "Do you need to evaluate the recent performance of the voice remote control" will be displayed on the user's mobile terminal. After the user confirms, a comprehensive evaluation will be conducted on the performance parameters of the voice remote control within the set time period. Based on the results of the comprehensive evaluation, the performance level of the voice remote control within the set time period will be determined and sent to the user's mobile terminal; the performance parameters include the command sending success rate, voice recognition accuracy, and response efficiency;

[0092] Count the total number of commands issued by the voice remote control within a set time period, and obtain the feedback results of the device corresponding to each command. If the feedback result matches the command, mark the command as a successful command; if the feedback result does not match the command, mark the command as a failed command; parse the failed commands. If a set number of failed commands are concentrated on a certain group of devices, extract the number of failed commands of the corresponding device from the total number of failed commands and round it to an integer of one;

[0093] If the command sending failure rate is high for only one or two devices, there may be problems with these devices themselves or compatibility issues with the voice remote control. In this case, the failed commands corresponding to the device need to be normalized and recalculated.

[0094] Then the success rate of voice remote control command sending within the set time period is F1, and the formula Calculated, where b1 represents the total number of instructions issued, b2 represents the total number of successful instructions, and b3 represents the number of failed instructions for the corresponding device;

[0095] Count the total number of user commands received by the voice remote control within a set time period, extract the total number of commands successfully recognized by the voice remote control at one time, and calculate the proportion of the total number of commands successfully recognized at one time in the total number of user commands received to obtain the voice recognition accuracy F2 of the voice remote control within the set time period;

[0096] Record the time interval from when each command is sent from the voice remote control to when the device feedback is received within the set time period, mark the time interval as the command response time, calculate the average of the command response time of each command within the set time period, and use the calculated average as the response efficiency F3 of the voice remote control within the set time period;

[0097] According to the formula , comprehensively calculate the command sending success rate F1, voice recognition accuracy F2 and response efficiency F3 of the voice remote control within the set time period, and obtain the performance evaluation index Fr of the voice remote control within the set time period; wherein w1, w2 and w3 represent the influencing weight factors of the command sending success rate F1, the voice recognition accuracy F2 and the response efficiency F3 respectively;

[0098] The preset performance evaluation index Fr corresponds to each group of index value ranges, and each group of index value ranges corresponds to a performance level; the larger the performance evaluation index Fr is, the higher the corresponding matching performance level is; the performance evaluation index Fr of the voice remote control in the set time period is matched with each group of preset index value ranges, so as to obtain the performance level of the voice remote control in the set time period;

[0099] The analysis and calculation processes corresponding to the modules of the above-mentioned voice remote control can be sent to the home terminal pre-deployed in the current application scenario through the wireless network for analysis and calculation when the voice remote control is low on power, and then fed back to the corresponding module for direct execution.

[0100] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An intelligent voice remote controller, characterized in that: It includes a control component, wherein the control component includes a setting module, a use module, an intelligent module and an abnormal module; Setting module: pair the voice remote control with the device to be controlled in the current application scenario; Usage module: When the user uses the voice remote control at the current time point, the user activates the voice function of the voice remote control and issues a command through the preset wake-up word. The voice remote control recognizes the current command and executes it according to the command content. If the user command recognition fails during this process, the corresponding adjustment steps are executed; Intelligent module: When the user issues an adaptive adjustment command, the state of the voice remote controller is adjusted to the adaptive state. Before the user issues a corresponding termination command to terminate the adaptive state of the voice remote controller, the device parameters in the current application scenario are adaptively adjusted through the voice remote controller; Use the voice remote control to adaptively adjust the device parameters in the current application scenario. Specifically: T1: Read the sensor data pre-deployed in the current application scenario; the deployed sensors include temperature sensors, air quality sensors, and human body sensing sensors; the sensor data includes temperature and air pollutant content; T2: Divide the current application scenario into functional sub-areas; obtain the season information of the current time point and determine the optimal temperature range; match the read temperature value with the determined optimal temperature range. If the match fails, take the middle value of the optimal temperature range as the adjustment target value, obtain the functional sub-area corresponding to the current user position through the human body sensing sensor, and set the temperature adjustment values ​​corresponding to different functional sub-areas; The temperature adjustment value of the functional sub-area is accumulated with the adjustment target value, and the accumulated result is used as the adjustment value of the air-conditioning device at the current time point. Based on the adjustment value of the air-conditioning device at the current time point, the air-conditioning device parameters of the current application scenario are adjusted through the voice remote control; T3: Divide the content of air pollutants according to the types of air pollutants to obtain the content of various types of air pollutants. Based on the number of deployment points of air quality sensors corresponding to the current application scenario, average the content of the same type of air pollutants at different deployment points to obtain the mean value of the content of various types of air pollutants in the current application scenario; the type of air pollutant is represented by i, where i=1,2......,h, h is the total number of air pollutant types, and the reference thresholds corresponding to the content of different types of air pollutants are preset; the mean value of the content of various types of air pollutants is marked as Ki, and the reference thresholds corresponding to the content of different types of air pollutants are marked as Ui; using the formula The average value Ki of the content of various types of air pollutants is weighted to obtain the air quality index Y of the current application scenario; where ai is the preset impact weight factor corresponding to different types of air pollutants; A threshold range corresponding to the air quality index Y is preset. If the air quality index Y of the current application scenario is within the preset threshold range, the air purification device is turned on by the voice remote control, and a power range of the air purification device corresponding to the threshold range is set. After the air purification device is turned on, the power of the air purification device is adjusted to the corresponding power range; If the air quality index Y of the current application scenario is higher than the preset threshold range, a power range of an air purification device corresponding to the range above the threshold is set, and the power of the air purification device is adjusted to the corresponding power range after the air purification device is turned on by the voice remote control. After the adjustment is completed, the air quality index Yt of each layout point is calculated, where t represents the number of each layout point, where t=1,2,..., g,g is the total number of layout points; the layout point corresponding to the highest air quality index Yt is marked, and the window closest to the marked layout point is opened by the voice remote control; Steps T1 to T3, when the voice remote controller is in an adaptive state, based on a preset adjustment time interval, after reaching the preset adjustment time interval, adaptively adjusting the device parameters in the current application scenario; Abnormal module: During the use of the voice remote control, if a command issued at a certain time does not receive feedback, the retrieval signaling is triggered and the corresponding processing steps are executed.

2. The intelligent voice remote controller according to claim 1, characterized in that: If the user command recognition fails, the corresponding adjustment steps are executed, specifically: Convert the user's command into text, and input the converted text content into a pre-built historical database for matching; extract the text content corresponding to the historical voice command from the historical database; Divide the converted text content into multiple groups of words; match the divided multiple groups of words with a preset basic word set, and remove the words that are successfully matched; and calculate the similarity between the remaining words and the words corresponding to the historical voice instructions, convert the remaining words and the words corresponding to the historical voice instructions into vector representations respectively, and calculate the cosine similarity between each group of vectors; Determine the part of speech of each of the remaining words, preset a similarity value matrix corresponding to the similarities between different parts of speech, combine the parts of speech of each of the remaining words with the words corresponding to the historical voice instructions and input them into the matrix, set different part of speech combinations to correspond to a part of speech similarity, thereby obtaining the part of speech similarity between each of the remaining words and the words corresponding to the historical voice instructions.

3. The intelligent voice remote controller according to claim 2, characterized in that: If the user command recognition fails, the corresponding adjustment steps are executed, which also include: The weight coefficients corresponding to the cosine similarity and the part-of-speech similarity are set respectively, and the cosine similarity and the part-of-speech similarity of each group of the remaining words are multiplied by the corresponding weight coefficients respectively, and then the sum is calculated to obtain the single group similarity between each word in the remaining words and the word corresponding to the historical voice instruction; the single group similarities are averaged to obtain the overall similarity between the historical voice instruction and the current remaining words; Historical voice commands are sorted according to overall similarity, and the historical voice commands with the highest similarity are selected as possible commands, which are verified with the user through the voice remote control; if the verification fails, the parameters of voice recognition are adjusted based on preset adjustment rules, and a voice recognition failure message is popped up to the user to remind the user to issue the command again and use the adjusted parameters for voice recognition.

4. The intelligent voice remote controller according to claim 1, characterized in that: Trigger retrieval signaling and execute corresponding processing steps, specifically: P1: Identify the specific device that does not provide feedback, and analyze the network connection and pairing status; P1-101: Send a network detection data packet to the connected network device; if no response is received, it is determined that the network connection is abnormal; P1-102: Check the pairing status between the specific device and the voice remote control that does not provide feedback. If the pairing status changes from paired to unpaired, it is considered a pairing abnormality. P2: If the analysis result of the network connection and pairing status is abnormal, send the same command again after the set interval; if the device still does not respond, execute step P3; P3: Cut off the power of the corresponding device that has not responded through the voice remote control, wait for the set time and then reconnect the power of the device to restart. After restarting, send the control command again. If the device still has no feedback, execute step P4; P4: Randomly select a device in the current application scenario to send a random command. If the selected device responds normally and gives feedback, the corresponding device that does not give feedback is marked as a faulty device and sent to the user's mobile terminal; if the selected device also does not give feedback, the self-check signaling of the voice remote control is triggered, and step P5 is executed; P5: After the self-check signaling is triggered, the preset self-diagnosis program is started; according to the result of the self-diagnosis program, if a fault is displayed, the fault information is sent to the user's mobile terminal.

5. The intelligent voice remote controller according to claim 4, characterized in that: If it is displayed in step P5 that there is no fault, a pop-up window "Do you need to evaluate the recent performance of the voice remote control?" is displayed on the user's mobile terminal. After the user confirms, a comprehensive evaluation is performed on the performance parameters of the voice remote control within the set time period. Based on the results of the comprehensive evaluation, the performance level of the voice remote control within the set time period is determined and sent to the user's mobile terminal; the performance parameters include the command sending success rate, voice recognition accuracy and response efficiency.

6. The intelligent voice remote controller according to claim 5, characterized in that: Comprehensively evaluate the performance parameters of the voice remote control within the set time period, specifically: Count the total number of commands issued by the voice remote control within a set time period, and obtain the feedback results of the device corresponding to each command. If the feedback result matches the command, mark the command as a successful command; if the feedback result does not match the command, mark the command as a failed command; parse the failed commands. If a set number of failed commands are concentrated on a certain group of devices, extract the number of failed commands of the corresponding device from the total number of failed commands and round it to an integer of one; Then the success rate of voice remote control command sending within the set time period is F1, and the formula Calculated, where b1 represents the total number of instructions issued, b2 represents the total number of successful instructions, and b3 represents the number of failed instructions for the corresponding device; Count the total number of user commands received by the voice remote control within a set time period, extract the total number of commands successfully recognized by the voice remote control at one time, and calculate the proportion of the total number of commands successfully recognized at one time in the total number of user commands received to obtain the voice recognition accuracy F2 of the voice remote control within the set time period; Record the time interval from when each command is sent from the voice remote control to when the device feedback is received within the set time period, mark the time interval as the command response time, calculate the average of the command response time of each command within the set time period, and use the calculated average as the response efficiency F3 of the voice remote control within the set time period; According to the formula , comprehensively calculate the command sending success rate F1, voice recognition accuracy F2 and response efficiency F3 of the voice remote control within the set time period, and obtain the performance evaluation index Fr of the voice remote control within the set time period; Where w1, w2 and w3 represent the influencing weight factors of the command sending success rate F1, the speech recognition accuracy F2 and the response efficiency F3 respectively; The preset performance evaluation index Fr corresponds to each group of index value ranges, and each group of index value ranges corresponds to a performance level; the performance evaluation index Fr of the voice remote control within the set time period is matched with the preset groups of index value ranges to obtain the performance level of the voice remote control within the set time period.

Images