Intelligent control method for electric lifting massage mattress based on voice recognition

Through voice recognition technology, the problem of inaccurate adjustment of traditional electric lift massage mattresses is solved, and intelligent lift control is realized, providing accurate adjustment and user-independent operation convenience.

CN119091875AInactive Publication Date: 2024-12-06SUZHOU HUIERKANG MEDICAL INSTR CO LTD
View PDF 13 Cites 0 Cited by

Patent Information

Application Number
CN202411118655.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional electric lift massage mattresses require manual adjustment of the foot of the bed, which leads to inaccurate adjustment and cannot achieve accurate adjustment of the bed body.

Method used

Using an intelligent control method based on voice recognition, voice data is collected through the data sensing unit, voice recognition and processing is performed, control signals are generated, and the lift control unit is activated to realize intelligent control of the electric lift massage mattress.

Benefits of technology

It realizes accurate adjustment of electric lift massage mattresses and user-independent operation, improving the accuracy and convenience of adjustment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119091875A_ABST
    Figure CN119091875A_ABST
Patent Text Reader

Abstract

The invention discloses an intelligent control method for an electric lifting massage mattress based on voice recognition, and relates to the related field of voice recognition, and the method comprises the steps: collecting voice data of a target user through a data sensing unit, synchronizing multiple pieces of voice information to a voice recognition unit through a data processing unit, and transmitting the voice information to a voice recognition unit; the method comprises the following steps: acquiring a plurality of identification instructions to carry out endpoint detection on a massage mattress, generating an endpoint positioning data set, activating a lifting control unit according to the plurality of identification instructions, generating a lifting control signal, and cooperatively controlling a first lifting control module and a second lifting control module through the lifting control signal to carry out electric lifting intelligent control on the massage mattress. The technical problem that bed legs of a traditional electric lifting massage mattress need to be manually adjusted, and adjustment is not accurate is solved, and the technical effects that intelligent lifting control is conducted on the electric lifting massage mattress through voice recognition, accurate adjustment is achieved, and user autonomous operation is achieved are achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field related to speech recognition, and in particular to an intelligent control method for an electric lifting massage mattress based on speech recognition. Background Art

[0002] With the improvement of living standards and the enhancement of health awareness, people's demand for massage mattresses is increasing. Traditional massage mattresses have certain limitations in terms of adjustment flexibility and comfort. When adjusting the height and angle of the bed, traditional massage beds often require manual adjustment of the bed legs or the use of complex mechanical structures such as hydraulic pumps. The operation is inconvenient and the adjustment range is limited. There is a technical problem that the bed cannot be accurately adjusted. Summary of the invention

[0003] The present application provides an intelligent control method for an electric lifting massage mattress based on voice recognition, thereby solving the technical problem that traditional electric lifting massage mattresses require manual adjustment of the bed legs and suffer from inaccurate adjustment. The application achieves the technical effect of intelligently controlling the lifting of the electric lifting massage mattress through voice recognition, realizing accurate adjustment and user-independent operation.

[0004] The present application provides an intelligent control method for an electric lifting massage mattress based on voice recognition, comprising: collecting voice data of a target user through a data sensing unit to obtain multiple voice information; synchronizing the multiple voice information to a voice recognition unit through a data processing unit to obtain multiple recognition instructions; performing endpoint detection on the massage mattress according to the multiple recognition instructions to generate an endpoint positioning data set; based on the endpoint positioning data set, activating a lifting control unit according to the multiple recognition instructions to generate a lifting control signal, the lifting control unit comprising a first lifting control module and a second lifting control module; and intelligently controlling the electric lifting of the massage mattress by collaboratively controlling the first lifting control module and the second lifting control module through the lifting control signal.

[0005] In a possible implementation, the voice data of the target user is collected through a data sensing unit to obtain multiple voice information, and the following processing is performed: the massage mattress is monitored in real time, and a pressure area is generated according to the monitoring results; when the pressure area of ​​the massage mattress is greater than or equal to a preset pressure area, a pressure signal is generated; the pressure signal is sent to the data sensing unit to start the voice sensing module; the voice data of the target user is collected through the voice sensing module to obtain the multiple voice information.

[0006] In a possible implementation, the multiple voice information are synchronized to a voice recognition unit through a data processing unit to obtain multiple recognition instructions, and the following processing is performed: the multiple voice information are subjected to noise reduction processing to generate multiple standard voice data; the multiple standard voice data are synchronized to the data processing unit for segmentation to generate multiple voice segments; the multiple voice segments are traversed by the voice recognition unit to perform keyword recognition, the multiple voice segments are screened according to the recognition results, and multiple voice keywords are determined; multiple control signals are generated based on the multiple voice keywords, and the multiple control signals are converted into the multiple recognition instructions.

[0007] In a possible implementation, the speech recognition unit traverses the multiple speech segments to perform keyword recognition, the multiple speech segments are screened according to the recognition results, and multiple speech keywords are determined, and the following processing is performed: based on the speech recognition unit traversing the multiple speech segments to perform feature extraction and generate multiple speech features; based on the multiple speech features, the multiple speech segments are subjected to text recognition to generate a candidate text sequence; the context information of the multiple speech segments is extracted to perform keyword recognition on the candidate text sequence to determine a key text sequence; and the key text sequence is added to the recognition result.

[0008] In a possible implementation, multiple control signals are generated based on the multiple voice keywords, the multiple control signals are converted into the multiple recognition instructions, and the following processing is performed: mapping and associating the multiple voice keywords with the lifting and lowering control information of the massage mattress to generate a mapping association network; traversing the multiple voice keywords as indexes according to the mapping association network to generate multiple control signals; parsing the multiple control signals to determine multiple massage mattress lifting control information and multiple massage mattress lowering control signals; adding the multiple massage mattress lifting control information and the multiple massage mattress lowering control signals to the multiple recognition instructions.

[0009] In a possible implementation, based on the endpoint positioning data set, the lifting and lowering control unit is activated according to the multiple recognition instructions, a lifting and lowering control signal is generated, and the following processing is performed: a keyword library is constructed based on the endpoint positioning data set, and the multiple voice keywords are matched with the keyword library to generate a keyword matching result; based on the keyword matching result, it is determined whether the multiple recognition instructions are control instructions; if the keyword matching result is that the multiple voice keywords can match the keyword library, then the multiple recognition instructions are control instructions, and the control instructions include a first control instruction and / or a second control instruction; the lifting and lowering control unit is activated by the control instructions to perform lifting and lowering control on the first lifting and lowering control module and the second lifting and lowering control module, and a lifting and lowering control signal is generated.

[0010] In a possible implementation, the lifting control unit is activated by the control instruction to perform lifting control on the first lifting control module and the second lifting control module, a lifting control signal is generated, and the following processing is performed: when the control instruction is the first control instruction, a first lifting control signal is generated to perform lifting control on the first lifting control module; when the control instruction is the second control instruction, a second lifting control signal is generated to perform lifting control on the second lifting control module; when the control instruction is the first control instruction and the second control instruction, a linkage lifting control signal is generated to perform linkage lifting control on the first lifting control module and the second lifting control module; the first lifting control signal, the second lifting control signal, and the linkage lifting control signal are added to the lifting control signal.

[0011] In a possible implementation, when the control instruction is the first control instruction and the second control instruction, a linkage lifting control signal is generated to perform linkage lifting control on the first lifting control module and the second lifting control module, and the following processing is performed: when the control instruction is the first control instruction and the second control instruction, a linkage lifting control signal is generated; based on the linkage lifting control signal, it is determined whether the lifting control of the first control instruction and the second control instruction is consistent. If the first control instruction and the second control instruction are both ascending control or descending control, it is deemed that the lifting control of the first control instruction and the second control instruction is consistent; when the first control instruction and the second control instruction are both ascending control, first collaborative control information is generated, and the first lifting control module and the second lifting control module are controlled to perform bidirectional retreat through the first collaborative control information; when the first control instruction and the second control instruction are both descending control, second collaborative control information is generated, and the first lifting control module and the second lifting control module are controlled to perform bidirectional forward movement through the second collaborative control information; the first collaborative control information and the second collaborative control information are added to the linkage lifting control signal.

[0012] One or more technical solutions provided in this application have at least the following technical effects or advantages: The intelligent control method of an electric lifting massage mattress based on voice recognition provided in the present application solves the technical problem that traditional electric lifting massage mattresses require manual adjustment of the bed legs and the result of inaccurate adjustment. The method achieves the technical effect of intelligently controlling the lifting of the electric lifting massage mattress through voice recognition, realizing accurate adjustment and user-independent operation. BRIEF DESCRIPTION OF THE DRAWINGS In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the method according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes. Figure 1 A flow chart of an intelligent control method for an electric lifting massage mattress based on voice recognition provided in an embodiment of the present application; Figure 2 A schematic flow chart of obtaining multiple recognition instructions of the intelligent control method for an electric lifting massage mattress based on voice recognition provided in an embodiment of the present application. DETAILED DESCRIPTION

[0013] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0014] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0015] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0016] The present application embodiment provides an intelligent control method for an electric lifting massage mattress based on voice recognition, such as Figure 1 As shown, the method includes: Step A100, collects the voice data of the target user through the data sensing unit to obtain multiple voice information; in a possible implementation, step A100 further includes step A110, real-time monitoring of the massage mattress, and generating a pressure area according to the monitoring result; pressure sensors can be embedded in key areas of the massage mattress (such as the back, legs, etc.), and the pressure distribution on the surface of the mattress can be sensed by the pressure sensors, and then the pressure data is collected regularly or continuously and sent to the data processing unit, the data processing unit receives the data from each sensor, and calculates the overall pressure area on the mattress through an algorithm, in order to prevent the user from being injured by external interference or voice when starting the massage before lying down completely, and the pressure area can be accurately estimated by integrating the sensor data and dividing the area, and step A120 is executed, when the massage When the pressure area of ​​the mattress is greater than or equal to the preset pressure area, a pressure signal is generated; it means comparing the calculated pressure area with the preset pressure area threshold, and the preset pressure area can be set according to user preference, mattress design or health guidelines. When the pressure area is greater than or equal to the preset pressure area, the data processing unit generates a pressure signal, and the pressure signal is used to indicate that the mattress is currently in a high pressure state, and step A130 is executed to send the pressure signal to the data sensing unit to start the voice sensing module; it means that the data sensing unit receives the pressure signal from the data processing unit, and according to the received pressure signal, the data sensing unit activates the voice sensing module and prepares to collect voice data, and step A140 is executed to collect the voice data of the target user through the voice sensing module to obtain the multiple voice information. The voice data of the target user is collected through the activated voice sensing module. The collection process may include a microphone array for capturing voices from multiple directions to improve the accuracy and clarity of the collection. The voice sensing module converts the collected raw voice data into a digital signal and may perform preliminary processing (such as denoising, enhancement, etc.), and then extracts multiple voice information (such as voice content, speaker recognition, etc.), which is further processed or stored according to system requirements to provide users with a more personalized and comfortable massage experience.

[0017] Execute step A200, synchronize the multiple voice information to the voice recognition unit through the data processing unit, and obtain multiple recognition instructions; in a possible implementation, Figure 2As shown, step A200 further includes step A210, performing noise reduction processing on the multiple voice information to generate multiple standard voice data; each voice information can be pre-processed initially, and the pre-processing can include removing DC offset, pre-emphasis, etc. to improve the signal-to-noise ratio of the signal, and then using a noise estimation algorithm (such as spectral subtraction, Wiener filtering, deep learning, etc.) to estimate the background noise in the voice signal, and performing noise reduction processing on the voice signal according to the noise estimation result. The noise reduction algorithm will adjust the components in the signal according to the characteristics of the signal and the noise to reduce or eliminate the interference of the noise, and finally perform post-processing such as smoothing processing, bandwidth limiting, etc. on the voice signal after noise reduction, thereby obtaining multiple standard voice data, and then perform step A220 to synchronize the multiple standard voice data to the data processing unit for segmentation to generate multiple voice segments; first, the multiple standard voice data after noise reduction are synchronized to the data processing unit through an appropriate data transmission protocol, and the data processing unit can be a dedicated server, embedded system or cloud service, and in the data processing unit, feature extraction is performed on the received standard voice data. Feature extraction usually includes the energy, fundamental frequency, spectral characteristics of the speech signal. According to the extracted features, the speech segmentation algorithm is applied, that is, the continuous speech data stream is segmented into multiple speech segments based on time thresholds or statistical models, machine learning and other methods, and a unique identifier is assigned to each generated speech segment, and its start time, duration and other metadata are recorded to facilitate subsequent indexing and retrieval, providing a clearer and more orderly data basis for subsequent tasks such as speech analysis, recognition or emotion calculation.

[0018] Execute step A230, traverse the multiple voice segments to perform keyword recognition through the voice recognition unit, screen the multiple voice segments according to the recognition results, and determine multiple voice keywords; in a possible implementation, step A230 further includes step A231, based on the voice recognition unit traversing the multiple voice segments to perform feature extraction, and generate multiple voice features; by using the voice recognition unit to traverse each voice segment, for each voice segment, calculate its acoustic features, the acoustic features may include but are not limited to MFCC (Mel frequency cepstral coefficient), PLP (perceptual linear prediction), spectrogram, energy envelope, etc., the acoustic features are key information in capturing voice signals, execute step A232, perform semantic recognition on the multiple voice segments according to the multiple voice features, and generate text candidate sequences; semantic recognition of multiple voice segments based on the obtained voice features can be achieved by constructing a voice recognition model, and the specific steps are: first, the extracted voice features are input into the trained voice recognition model, and the voice recognition model is based on the acoustic feature mapping The speech recognition model is obtained by training and learning of the text sequence, and the speech recognition model also has a built-in decoder, which is used to convert the continuous acoustic feature sequence into the most likely text sequence. For each speech segment, the decoding process will generate one or more text candidate sequences, and the text candidate sequence is used to characterize the speech recognition model's understanding of the speech content. Step A233 is executed to extract the context information of the multiple speech segments to perform keyword recognition on the text candidate sequence to determine the key text sequence; based on the multiple speech segments, the context information of each speech segment is extracted, and the context information may include but is not limited to the position of the speech segment, that is, the position in the time series, the surrounding speech content, the user's historical behavior or preference, etc., and then the extracted context information is used to perform semantic understanding and recognition of keywords on the text candidate sequence through rule-based keyword matching, statistical-based language model scoring, or combined with deep learning models. Text sequences containing important information or keywords are screened out from the text candidate sequences as the final key text sequence, and step A234 is executed to add the key text sequence to the recognition result.

[0019] Execute step A240, generate multiple control signals based on the multiple voice keywords, and convert the multiple control signals into the multiple recognition instructions. In a possible implementation, step A240 further includes step A241, mapping and associating the multiple voice keywords with the lifting control information of the massage mattress to generate a mapping association network; first, clarify all voice keywords related to the lifting control of the massage mattress that the user may use, such as "back rise", "back fall", "leg rise", "leg fall", etc., define the lifting control information of the massage mattress corresponding to the keyword set, including specific rising or falling actions, adjustment amplitude, speed and other parameters, and further map and associate each keyword with one or more items in the control information set. This mapping can be directly one keyword corresponding to one control action, or it can be based on a combination of multiple keywords corresponding to one control action, or the mapping relationship is determined considering context information, and a mapping association network is constructed according to the mapping relationship. The mapping association network can be based on if-else statement rules or a machine learning model, and is ultimately used to generate corresponding control signals according to the input voice keywords. Execute step A242, and traverse the multiple voice keywords as indexes according to the mapping association network to generate multiple control signals; traversing the multiple recognized voice keywords according to the mapping association network means that for each keyword, the control information matching it is searched in the mapping association network, and a corresponding control signal is generated according to the matching result, the signal may include the rise and fall instructions of the massage mattress, and specific adjustment parameters, such as height change, speed and other parameters, and execute step A243 to parse the multiple control signals to determine multiple massage mattress rise control information and multiple massage mattress fall control signals; and parse the generated control signals to classify them into massage mattress rise control information and massage mattress fall control signals, and extract specific adjustment parameters from the control signals, the multiple massage mattress rise control information includes the rise control information of the back of the massage mattress and / or the rise control information of the legs, and the multiple massage mattress fall control signals include the fall control information of the back of the massage mattress and / or the fall control information of the legs, and execute step A244 to add the multiple massage mattress rise control information and the multiple massage mattress fall control signals to the multiple recognition instructions. Integrating the analyzed multiple massage mattress rising control information and multiple massage mattress falling control signals into the recognition instruction not only improves the user's convenience, but also enhances the intelligence level of the massage mattress.

[0020] Execute step A300, perform endpoint detection on the massage mattress according to the multiple recognition instructions, and generate an endpoint positioning data set; in the present application, first classify the multiple recognition instructions received, distinguish the multiple recognition instructions related to the massage action of the massage mattress, especially those involving start, pause, continue, stop, rise, fall, etc., extract the specific parameters of the position to be controlled on the massage mattress from the relevant instructions, and clarify the definition of the starting point and the ending point of the massage action. The starting point can be the moment when the back or leg of the massage mattress starts to rise or fall, and the ending point is the moment when the back or leg of the massage mattress completely stops. In the actual operation of the massage mattress, according to the result of the endpoint detection algorithm, collect the starting point and the ending point data of the back or leg of the massage mattress, and the starting point and the ending point data can include timestamp, massage mode, intensity and other related information, and organize the collected data into a structured endpoint positioning data set. Each data point should contain the unique identifier of the massage action, the starting point timestamp, the ending point timestamp, the massage mode, the intensity and other fields, and on this basis, generate the endpoint positioning data set of the massage mattress, providing strong support for the intelligent control and personalized service of the massage mattress.

[0021] Execute step A400, based on the endpoint positioning data set, activate the lifting control unit according to the multiple recognition instructions to generate a lifting control signal, the lifting control unit includes a first lifting control module and a second lifting control module; In a possible implementation, step A400 further includes step A410, constructing a keyword library based on the endpoint positioning data set, matching the multiple voice keywords with the keyword library, and generating a keyword matching result; extracting voice keywords related to the massage mattress control from the endpoint positioning data set. These keywords should be words commonly used by users when initiating control instructions, such as "back up", "back down", "leg up", "leg down", etc. The extracted keywords are screened, and duplicates, irrelevant or low-frequency items are removed, and the most representative and control-significant keywords are retained. The screened keywords are organized into a keyword library according to a certain structure. The keyword library can be a simple list form or a complex structure containing information such as keyword weights, synonyms, and phrases.

[0022] The text contents in the received multiple recognition instructions are further preprocessed, such as word segmentation, stop word removal, word form restoration, etc., so as to match with the vocabulary in the keyword library, and multiple voice keywords are matched with the vocabulary in the keyword library using matching algorithms such as exact matching, fuzzy matching, and semantic-based matching. According to the output of the matching algorithm, a keyword matching result is generated, and the keyword matching result is used to characterize multiple voice keywords that match the vocabulary in the keyword library, as well as the strength and credibility of the match.

[0023] Execute step A420, based on the keyword matching result, determine whether the multiple recognition instructions are control instructions; execute step A430, if the keyword matching result is that the multiple voice keywords can match the keyword library, then the multiple recognition instructions are control instructions, and the control instructions include a first control instruction and / or a second control instruction; Based on the keyword matching results, it is determined whether the multiple recognition instructions include control instructions. If the multiple voice keywords are successfully matched with the control-related words in the keyword library, and these matching results logically constitute a valid control instruction, for example, "back rise" is followed by "start", then these recognition instructions are regarded as control instructions, and the control instructions are further determined according to the specific content of the instructions and the control system architecture of the massage mattress. The control instructions are classified into a first control instruction, i.e., an instruction for controlling the first lifting control module, and a second control instruction, i.e., an instruction for controlling the second lifting control module, or an instruction containing both.

[0024] Execute step A440, activate the lifting control unit through the control instruction to perform lifting control on the first lifting control module and the second lifting control module, and generate a lifting control signal. In a possible implementation, step A440 further includes step A441, when the control instruction is the first control instruction, generate a first lifting control signal to perform lifting control on the first lifting control module; execute step A442, when the control instruction is the second control instruction, generate a second lifting control signal to perform lifting control on the second lifting control module; Sending the generated lifting control signal to the lifting control unit to activate its control over the first lifting control module and the second lifting control module means that the control unit will adjust the lifting state of the massage mattress according to the received signal to meet the user's control needs.

[0025] Furthermore, when the control instruction is the first control instruction, it means that if the control instruction contains keywords or phrases related to the first lifting control module, that is, "back up" or "back down", the instruction is identified as the first control instruction. For the first control instruction, the system will generate a first lifting control signal according to the lifting direction, height change, speed and other parameters specified in the instruction. The first lifting control signal is used to activate the first lifting control module and perform lifting operations according to the specified parameters.

[0026] When the control instruction is the second control instruction, it means that if the control instruction contains keywords or phrases related to the second lifting control module, that is, "legs up" and "legs down", the instruction is identified as the second control instruction, and for the second control instruction, the system will generate a second lifting control signal. This signal will be for the second lifting control module, and also contain necessary parameters such as lifting direction, height change, speed, etc.

[0027] Execute step A443, when the control instruction is the first control instruction and the second control instruction, generate a linkage lifting control signal to perform linkage lifting control on the first lifting control module and the second lifting control module; in a possible implementation method, step A443 further includes step A4431, when the control instruction is the first control instruction and the second control instruction, generate a linkage lifting control signal; based on the received first control instruction and the second control instruction, generate a linkage lifting control signal, the linkage lifting control signal is the basis for subsequent control operations, and contains common control instructions for the two lifting control modules.

[0028] Execute step A4432, and based on the linked lifting control signal, determine whether the lifting control of the first control instruction and the second control instruction is consistent. If the first control instruction and the second control instruction are both rising control or falling control, then the lifting control of the first control instruction and the second control instruction are deemed to be consistent. Then compare the lifting direction in the first control instruction and the second control instruction, that is, whether the rising or falling is the same, and determine whether the lifting control in the first control instruction and the second control instruction is consistent. If the first control instruction and the second control instruction are both rising control or both falling control, then the lifting control is deemed to be consistent. If the lifting control is inconsistent, that is, one of the control instructions is rising and the other control instruction is falling, it can be processed according to the preset rules, and there is no two-way retreat or two-way advance during the processing.

[0029] Execute step A4433, when the first control instruction and the second control instruction are both ascending control, generate first collaborative control information, and control the first lifting control module and the second lifting control module to perform bidirectional fallback through the first collaborative control information; When the first control instruction and the second control instruction are both ascending controls, the first collaborative control information is generated. The first collaborative control information will guide the first lifting control module and the second lifting control module to perform a two-way retract operation, which means that the first lifting control module and the second lifting control module perform an ascending operation at the same time, moving from an initial position or a higher position, and the connection between the first lifting control module and the second lifting control module retracts at the same time, reducing the target user's abdominal squeezing feeling.

[0030] Execute step A4434. When the first control instruction and the second control instruction are both descending control instructions, generate second collaborative control information, and control the first lifting control module and the second lifting control module to move forward in both directions through the second collaborative control information. Execute step A4435 to add the first collaborative control information and the second collaborative control information to the linkage lifting control signal.

[0031] When the first control instruction and the second control instruction are both descending controls, the second collaborative control information is generated, and the second collaborative control information is used to control the two lifting control modules to perform a two-way forward operation, that is, the first lifting control module and the second lifting control module simultaneously perform a descending operation, moving from an initial position or a lower position, and the connection between the first lifting control module and the second lifting control module simultaneously moves forward, and the forward distance is equal to the above-mentioned retreat distance, so as to better adjust the overall height of the mattress.

[0032] Finally, according to the specific content of the control instruction, the first collaborative control information or the second collaborative control information is added to the linked lifting control signal. The linked lifting control signal will simultaneously include independent control instructions for the two lifting control modules and the collaborative control information between them, thereby realizing precise and collaborative control of the two lifting control modules and providing a more personalized and comfortable massage experience.

[0033] Execute step A444 to add the first lifting control signal, the second lifting control signal, and the linkage lifting control signal to the lifting control signal.

[0034] In the control system of the massage mattress, when it is necessary to simultaneously process the independent control signals of the first lifting control module and the second lifting control module, i.e., the first lifting control signal and the second lifting control signal, and the linkage control signal between them, i.e., the linkage lifting control signal, after adding each control signal to the set, the lifting control signal is integrated and verified, which may include verifying the integrity and correctness of the signal, and the logical consistency between them. It ensures that all signals are valid and can be executed, so as to flexibly handle the independent and linkage control requirements between multiple lifting control modules and provide a richer and more personalized massage experience.

[0035] Next, step A500 is executed to coordinately control the first lifting control module and the second lifting control module through the lifting control signal to perform intelligent control on the electric lifting of the massage mattress.

[0036] In the electric lifting intelligent control system of the massage mattress, through a carefully designed lifting control signal, the coordinated control of the first lifting control module and the second lifting control module can be achieved, so as to accurately adjust different areas of the massage mattress to achieve the comfort and massage effect required by the user.

[0037] First, the control command of the target user is received. The control command is input through voice recognition and converted into a signal recognizable by the system. The received control command is parsed to determine the intention of the target user, such as rising, falling, stopping, etc. The corresponding lifting control signal is generated according to the parsing result. The lifting control signal includes not only independent control instructions for the first lifting control module, i.e., the back lifting control module of the massage mattress, and the second lifting control module, i.e., the leg lifting control module of the massage mattress, but also a linkage control signal to achieve coordinated action between the two modules.

[0038] During the lifting action, the real-time feedback signal of the lifting control module, such as the data of the position sensor, force sensor, etc., is fed back and received. The feedback signal is used to monitor the accuracy and stability of the lifting action and make adjustments as needed. At the same time, when the lifting action reaches the preset target position or meets the user's specific needs, the control signal is stopped and the next user instruction is waited for. At this time, the massage mattress will remain in the current state to provide users with a comfortable massage experience.

[0039] The embodiments of the present application solve the technical problem that the traditional electric lift massage mattress requires manual adjustment of the bed legs, resulting in inaccurate adjustment. The electric lift massage mattress can be intelligently raised and lowered by voice recognition to achieve the technical effect of precise adjustment and user-independent operation.

[0040] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. An intelligent control method for an electric lifting massage mattress based on voice recognition, characterized in that: The method comprises: The data sensor unit collects the voice data of the target user to obtain multiple voice information; Synchronizing the plurality of voice information to the voice recognition unit through the data processing unit to obtain a plurality of recognition instructions; Performing endpoint detection on the massage mattress according to the multiple recognition instructions to generate an endpoint positioning data set; Based on the endpoint positioning data set, activating a lifting control unit according to the multiple recognition instructions to generate a lifting control signal, the lifting control unit comprising a first lifting control module and a second lifting control module; The first lifting control module and the second lifting control module are cooperatively controlled by the lifting control signal to intelligently control the electric lifting of the massage mattress.

2. The intelligent control method of the electric lifting massage mattress based on voice recognition according to claim 1, characterized in that: The data sensing unit collects the voice data of the target user to obtain a plurality of voice information, and the method includes: Monitor the massage mattress in real time and generate the pressure area based on the monitoring results; When the pressure-bearing area of ​​the massage mattress is greater than or equal to a preset pressure-bearing area, a pressure signal is generated; Sending the pressure signal to the data sensing unit to start the voice sensing module; The voice data of the target user is collected through the voice sensing module to obtain the multiple voice information.

3. The intelligent control method of the electric lifting massage mattress based on voice recognition according to claim 1, characterized in that: The plurality of voice information are synchronized to the voice recognition unit through the data processing unit to obtain a plurality of recognition instructions, the method comprising: Performing noise reduction processing on the plurality of voice information to generate a plurality of standard voice data; Synchronizing the plurality of standard voice data to the data processing unit for segmentation to generate a plurality of voice segments; The speech recognition unit traverses the plurality of speech segments to perform keyword recognition, and screens the plurality of speech segments according to the recognition results to determine a plurality of speech keywords; A plurality of control signals are generated based on the plurality of voice keywords, and the plurality of control signals are converted into the plurality of recognition instructions.

4. The intelligent control method of the electric lifting massage mattress based on voice recognition according to claim 3, characterized in that: The method comprises: traversing the plurality of voice segments by the voice recognition unit to perform keyword recognition, screening the plurality of voice segments according to the recognition results, and determining a plurality of voice keywords. Traversing the plurality of speech segments based on the speech recognition unit to perform feature extraction and generate a plurality of speech features; Performing semantic recognition on the multiple speech segments according to the multiple speech features to generate a text candidate sequence; Extracting context information of the multiple voice segments to perform keyword recognition on the text candidate sequence to determine a key text sequence; The key text sequence is added to the recognition result.

5. The intelligent control method of the electric lifting massage mattress based on voice recognition as claimed in claim 3, characterized in that: Based on the multiple voice keywords, multiple control signals are generated, and the multiple control signals are converted into the multiple recognition instructions. The method includes: Generate a mapping association network by mapping and associating the plurality of voice keywords with the lifting and lowering control information of the massage mattress; According to the mapping association network, the plurality of voice keywords are traversed as indexes to generate a plurality of control signals; Analyzing the multiple control signals to determine multiple massage mattress ascending control information and multiple massage mattress descending control signals; The plurality of massage mattress ascending control information and the plurality of massage mattress descending control signals are added to the plurality of recognition instructions.

6. The intelligent control method of the electric lifting massage mattress based on voice recognition according to claim 5, characterized in that: Based on the endpoint positioning data set, activating the lifting control unit according to the multiple recognition instructions to generate a lifting control signal, the method includes: Building a keyword library based on the endpoint positioning data set, matching the multiple voice keywords with the keyword library, and generating a keyword matching result; Based on the keyword matching result, determining whether the multiple recognition instructions are control instructions; If the keyword matching result is that the multiple voice keywords can match the keyword library, the multiple recognition instructions are control instructions, and the control instructions include a first control instruction and / or a second control instruction; The lifting control unit is activated by the control instruction to perform lifting control on the first lifting control module and the second lifting control module to generate a lifting control signal.

7. The intelligent control method of the electric lifting massage mattress based on voice recognition according to claim 6, characterized in that: The method includes: activating the lifting control unit through the control instruction to perform lifting control on the first lifting control module and the second lifting control module to generate a lifting control signal; When the control instruction is the first control instruction, generating a first lifting control signal to perform lifting control on the first lifting control module; When the control instruction is the second control instruction, generating a second lifting control signal to perform lifting control on the second lifting control module; When the control instruction is the first control instruction and the second control instruction, a linkage lifting control signal is generated to perform linkage lifting control on the first lifting control module and the second lifting control module; The first lifting control signal, the second lifting control signal, and the linkage lifting control signal are added to the lifting control signal.

8. The intelligent control method of the electric lifting massage mattress based on voice recognition according to claim 7, characterized in that: When the control instruction is the first control instruction and the second control instruction, a linkage lifting control signal is generated to perform linkage lifting control on the first lifting control module and the second lifting control module, and the method includes: When the control instruction is the first control instruction and the second control instruction, a linkage lifting control signal is generated; Based on the linkage lifting control signal, determining whether the lifting control of the first control instruction and the second control instruction is consistent; if the first control instruction and the second control instruction are both lifting control or descending control, it is considered that the lifting control of the first control instruction and the second control instruction is consistent; When the first control instruction and the second control instruction are both ascending control, first collaborative control information is generated, and the first lifting control module and the second lifting control module are controlled to perform bidirectional fallback through the first collaborative control information; When both the first control instruction and the second control instruction are descending control, second coordinated control information is generated, and the first lifting control module and the second lifting control module are controlled to move forward in both directions through the second coordinated control information; The first coordinated control information and the second coordinated control information are added to the linked lifting and lowering control signal.

Citation Information

Patent Citations

  • Voice control method and device of intelligent home system

    CN106847269A

  • Call control method for wearable device and wearable device

    CN106850963A

  • Intelligent hospital bed

    CN109846624A

  • Intelligent control system for vertical lifting of mattress

    CN109965584A

  • Voice control method for nursing bed

    CN110010122A