Monitoring equipment control method with intelligent voice interaction function and monitoring equipment

Abstract

The embodiment of the invention provides a monitoring equipment control method with an intelligent voice interaction function and monitoring equipment, through the method, a user can trigger inquiry interaction by himself / herself to start instruction language monitoring equipment to perform voice interaction, and the monitoring equipment performs voice interaction according to historical monitoring information of the user and actual voice inquiry information of the user. Intelligent voice interaction with a user is realized, and the method does not need to depend on a monitoring result measured in real time. Meanwhile, on this basis, further temporary inquiry analysis is carried out in combination with the voice inquiry information of the user, and the reply content is further expanded when the user confirms, so that a comprehensive intelligent voice interaction function is provided for the user, and the use experience of the user is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of voice interaction technology, and more specifically, to a monitoring device control method and a monitoring device with intelligent voice interaction function. Background Art

[0002] Current physiological index (blood pressure, blood sugar) testing equipment can be roughly divided into two types: medical testing equipment and home testing equipment. For current home testing equipment, it can often only interact with the user in a simple voice after the user finishes the test, based on the user's test results and the pre-set response content for the test results. The user cannot obtain relevant precautions according to actual needs, and the user experience is poor. Summary of the invention

[0003] In view of the above problems, the embodiments of the present application provide a monitoring device control method and a monitoring device with intelligent voice interaction function, aiming to improve the user experience through intelligent voice interaction.

[0004] In a first aspect, an embodiment of the present application provides a method for controlling a monitoring device with an intelligent voice interaction function, which is applied to the monitoring device; the method comprises:

[0005] In response to the inquiry interaction start instruction, obtaining the user's voice inquiry information and the first historical monitoring information;

[0006] Performing a preliminary inquiry analysis based on the voice inquiry information and the first historical monitoring information, generating a first inquiry reply text, and playing a first reply voice to the user based on the first inquiry reply text;

[0007] Performing a temporary inquiry analysis based on the voice inquiry information to generate an interactive inquiry text, and after the first reply voice is played, playing the inquiry voice to the user based on the interactive inquiry text; the interactive inquiry text is used to inquire the user whether the inquiry reply text needs to be expanded;

[0008] When the user instructs to expand the inquiry reply text, a second inquiry reply text is generated according to the current inquiry time, the voice inquiry information, the first historical monitoring information and the positioning information of the monitoring device, and a second reply voice is played to the user based on the second inquiry reply text.

[0009] In a possible implementation, performing a preliminary inquiry analysis based on the voice inquiry information and the first historical monitoring information to generate a first inquiry reply text includes:

[0010] Determine whether there is second historical monitoring information in the first historical monitoring information; the second historical monitoring information is used to represent monitoring information for the user within a preset past time period in response to the query interaction start instruction;

[0011] When it is determined that the second historical monitoring information does not exist in the first historical monitoring information, predicting a current physiological parameter index of the user based on the first historical monitoring information;

[0012] A preliminary inquiry analysis is performed based on the current physiological parameter index and the basic vital sign information of the user to generate the first inquiry response text.

[0013] In a possible implementation manner, after determining whether the second historical monitoring information exists in the first historical monitoring information, the method further includes:

[0014] When it is determined that the second historical monitoring information exists in the first historical monitoring information, performing distortion judgment on the second historical monitoring information to obtain a distortion judgment result;

[0015] If there is at least one second historical monitoring information in the distortion judgment result in which the distortion judgment result is negative, a preliminary inquiry analysis is performed based on the second historical monitoring information and the voice inquiry information to generate the first inquiry reply text;

[0016] If the distortion judgment result does not include the second historical monitoring information with a negative judgment result, the second historical monitoring information is removed, and the first inquiry reply text is generated based on the first historical monitoring information after the second historical monitoring information is removed.

[0017] In a possible implementation, when the user instructs to extend the inquiry reply text, generating a second inquiry reply text according to the current inquiry time, the voice inquiry information, the first historical monitoring information, and the location information of the monitoring device includes:

[0018] Determining the inquiry category of the voice inquiry information;

[0019] The second inquiry reply text is generated according to the inquiry category, the current inquiry time, the first historical monitoring information, and the location information of the monitoring device.

[0020] In a possible implementation, the inquiry categories include: food-related inquiries;

[0021] The generating the second inquiry reply text according to the inquiry category, the current inquiry time, the first historical monitoring information, and the location information of the monitoring device includes:

[0022] When the inquiry category is the food inquiry, performing a geographical location analysis according to the positioning information of the monitoring device to obtain a first geographical category for the user;

[0023] Performing accent-based regional location analysis according to the voice query information to obtain a second regional category;

[0024] Predicting a first dining food category of the user within a preset future time period according to the first regional category, the second regional category and the current inquiry time;

[0025] Based on the first historical monitoring information and the basic vital sign information of the user, a second dining food category is screened out from the first dining food category, and the second inquiry reply text is generated based on the second dining food category; the second dining food category is used to characterize the food category that is not suitable for the user.

[0026] In a possible implementation, the inquiry categories include: sports-related inquiries;

[0027] The generating the second inquiry reply text according to the inquiry category, the current inquiry time, the first historical monitoring information, and the location information of the monitoring device includes:

[0028] When the inquiry category is the sports inquiry, obtaining real-time weather data of the user's current geographical location according to the positioning information of the monitoring device, and determining a possible exercise location based on the real-time weather data and the current geographical location;

[0029] Performing exercise preference analysis according to the voice inquiry information to determine the user's exercise preference, and generating exercise recommendation items based on the exercise preference, the available exercise locations, the first historical monitoring information, and the user's basic vital sign information;

[0030] Based on the exercise recommendation item, the second inquiry response text is generated.

[0031] In a possible implementation, predicting the first dining food category of the user in a preset future time period according to the first regional category, the second regional category, and the current inquiry time includes:

[0032] Determining whether the first regional category is the same as the second regional category;

[0033] When it is determined that the first regional category is different from the second regional category, predicting the first dining food category according to a second weight coefficient of the second regional category, a first weight coefficient of the first regional category, and the current inquiry time; the second weight coefficient is greater than the first weight coefficient;

[0034] When it is determined that the first regional category is the same as the second regional category, the first meal food category is predicted according to any one of the second regional category and the regional category and the current inquiry time.

[0035] In a possible implementation, the interactive inquiry text includes: food interactive inquiry text and exercise interactive inquiry text;

[0036] The performing temporary inquiry analysis according to the voice inquiry information to generate an interactive inquiry text comprises:

[0037] When the inquiry category is a food inquiry, the food interactive inquiry text is generated; the food interactive inquiry text is used to inquire the user whether more inedible food categories need to be reported;

[0038] When the inquiry category is a sports inquiry, the sports interaction inquiry text is generated; the sports interaction inquiry text is used to inquire the user whether more recommended sports items need to be reported.

[0039] In a possible implementation, after playing the inquiry voice to the user based on the interactive inquiry text, the method further includes:

[0040] When the user's response to the played inquiry voice is negative, the voice interaction with the user is terminated.

[0041] In a second aspect, an embodiment of the present application provides a monitoring device with an intelligent voice interaction function, including:

[0042] An acquisition module, configured to acquire the user's voice inquiry information and first historical monitoring information in response to an inquiry interaction start instruction;

[0043] a first question-and-answer module, configured to perform a preliminary inquiry analysis based on the voice inquiry information and the first historical monitoring information, generate a first inquiry reply text, and play a first reply voice to the user based on the first inquiry reply text;

[0044] A second question-and-answer module is used to perform temporary inquiry analysis based on the voice inquiry information, generate an interactive inquiry text, and play an inquiry voice to the user based on the interactive inquiry text after the first reply voice is played; the interactive inquiry text is used to inquire the user whether the inquiry reply text needs to be expanded;

[0045] The third question and answer module is used to generate a second inquiry reply text according to the current inquiry time, the voice inquiry information, the first historical monitoring information and the positioning information of the monitoring device when the user instructs to expand the inquiry reply text, and play a second reply voice to the user based on the second inquiry reply text.

[0046] The technical solution provided by the embodiment of the present application may include the following beneficial effects: The embodiment of the present application provides a monitoring device control method and a monitoring device with intelligent voice interaction function, the method comprising: in response to the query interaction start instruction, obtaining the user's voice query information and the first historical monitoring information, performing preliminary query analysis according to the voice query information and the first historical monitoring information, generating a first query reply text, and playing the first reply voice to the user based on the first query reply text. Perform temporary query analysis according to the voice query information, generate an interactive query text, and after the first reply voice is played, play the query voice to the user based on the interactive query text. When the user's reply to playing the query voice is confirmation, generate a second query reply text according to the current query time, the voice query information, the first historical monitoring information and the positioning information of the monitoring device, and play the second reply voice to the user based on the second query reply text. In this way, the user can trigger the query interaction start instruction language monitoring device to perform voice interaction by himself, and the monitoring device realizes intelligent voice interaction with the user through the user's historical monitoring information and the user's actual voice query information, without relying on the real-time measurement of the monitoring results. At the same time, on this basis, further temporary inquiry analysis will be carried out in combination with the user's voice inquiry information, and when the user confirms, the reply content will be further expanded, thereby providing users with comprehensive intelligent voice interaction functions and effectively improving the user experience.

[0047] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0049] Figure 1 It is a flow chart of a method for controlling a monitoring device with an intelligent voice interaction function provided in an embodiment of the present application;

[0050] Figure 2It is a flowchart of an inquiry analysis method provided in an embodiment of the present application;

[0051] Figure 3 This is a flow chart of a method for generating a second inquiry reply text when the inquiry category is a food inquiry provided in an embodiment of the present application;

[0052] Figure 4 This is a flow chart of a method for generating a second inquiry reply text when the inquiry category is a sports inquiry provided by an embodiment of the present application;

[0053] Figure 5 It is a structural schematic diagram of a monitoring device with intelligent voice interaction function provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of this application more clear, the application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings. It should be noted that the embodiments described in the embodiments of this application are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0055] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0056] As can be seen from the previous article, current physiological index (blood pressure, blood sugar, etc.) testing equipment can be roughly divided into two types: medical testing equipment and home testing equipment. For current home testing equipment, it can often only conduct simple voice interaction with users after the user finishes the test, based on the user's test results and pre-set response content for the test results. Users cannot obtain answers to relevant questions, health advice or precautions according to actual needs, and the user experience is poor.

[0057] In order to solve the above problems, the embodiment of the present application provides a monitoring device control method and monitoring device with intelligent voice interaction function, the method comprising: in response to the query interaction start instruction, obtaining the user's voice query information and the first historical monitoring information, performing preliminary query analysis according to the voice query information and the first historical monitoring information, generating a first query reply text, and playing the first reply voice to the user based on the first query reply text. Perform temporary query analysis according to the voice query information, generate an interactive query text, and after the first reply voice is played, play the query voice to the user based on the interactive query text. When the user instructs to expand the query reply text, generate a second query reply text according to the current query time, the voice query information, the first historical monitoring information and the location information of the monitoring device, and play the second reply voice to the user based on the second query reply text. In this way, the user can trigger the query interaction start instruction by himself, so as to interact with the monitoring device by voice, and the monitoring device realizes intelligent voice interaction with the user through the user's historical monitoring information and the user's actual voice query information, without relying on the real-time monitoring results. At the same time, on this basis, further temporary inquiry analysis will be carried out in combination with the user's voice inquiry information, and when the user confirms, the reply content will be further expanded, thereby providing users with comprehensive intelligent voice interaction functions and effectively improving the user experience.

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0059] It should be noted that the monitoring device in the embodiment of the present application can be any monitoring device that can monitor human physiological parameters, such as a blood pressure meter, a blood glucose meter, a uric acid meter, a blood lipid meter, an electrocardiogram monitor, a thermometer, and a blood oximeter, etc. This embodiment does not limit the specific type of the monitoring device.

[0060] See also Figure 1 , which is a flow chart of a method for controlling a monitoring device with an intelligent voice interaction function provided by an embodiment of the present application, and specifically includes the following steps:

[0061] S101: In response to an inquiry interaction start instruction, obtaining user's voice inquiry information and first historical monitoring information.

[0062] The voice interaction function of the monitoring device is enabled based on an inquiry interaction start instruction from the outside. Among them, the inquiry interaction start instruction can be triggered by a dedicated inquiry trigger button on the monitoring device, and the user can generate an inquiry interaction start instruction by pressing the inquiry trigger button, thereby activating the voice interaction function of the monitoring device. The inquiry interaction start instruction can also be generated based on specific voice information, such as a specific name for the monitoring device. At the same time, the inquiry interaction start instruction can also be automatically generated after the user completes the monitoring. For example, when the user monitors the physiological parameters (blood pressure, blood sugar, etc.), the monitoring device automatically generates an inquiry interaction start instruction to establish voice interaction with the user. This embodiment does not limit the generation method of the inquiry interaction start instruction.

[0063] After the monitoring device responds to the inquiry interaction start instruction, the monitoring device turns on the voice receiving device on the device, thereby obtaining the user's voice inquiry information and the user's first historical monitoring information in real time.

[0064] The user's first historical monitoring information includes the physiological index parameters monitored by the user each time in the past certain period of time. If it is a home monitoring device, a monitoring device is often used by only one user, and the historical monitoring information stored in the monitoring device is often for a single user. In one possible implementation, if you want to store historical monitoring information of multiple users, the user can enter his or her own specific identification name through the interactive interface of the monitoring device before monitoring, so that the monitoring device automatically stores the monitoring information in the storage structure corresponding to the identification name after completing the monitoring. The monitoring information can also be uploaded to the cloud server through a network connection, so as to fully store the user's multiple monitoring information in the past time.

[0065] S102: Performing a preliminary inquiry analysis based on the voice inquiry information and the first historical monitoring information, generating a first inquiry reply text, and playing a first reply voice to the user based on the first inquiry reply text.

[0066] After obtaining the user's voice inquiry information and the first historical monitoring information, the monitoring device predicts the user's physical condition indicators based on the user's physiological parameters represented in the first historical monitoring information, and generates corresponding reply content based on the user's inquiry needs represented in the voice inquiry information.

[0067] In this process, a large language model can be pre-set in the monitoring device to analyze the semantic information and inquiry requirements in the user's voice inquiry information, and combined with the physiological indicator parameters (such as blood pressure, blood sugar, uric acid, etc.) represented by the user in the first historical monitoring information, the user's current physical indicator condition is predicted, and then the first inquiry reply text is generated. On the other hand, the large language model can also be set in a server in the cloud, and the monitoring device sends the voice inquiry information and the first historical monitoring information to the server through a network connection, and the server completes the actual language analysis and question-answering work.

[0068] It should be noted that in this step, in order to ensure the accuracy of the user's physical condition prediction, the first historical monitoring information often covers a longer historical period. However, if the user has conducted a physical monitoring within a short period of time before triggering the query interaction, then compared with the historical monitoring information that has been in the past for a long time, the data of the current monitoring can actually more accurately reflect the user's physical condition indicators. Therefore, in order to ensure the accuracy of the original data in this step, after obtaining the first historical monitoring information, it is necessary to determine whether there is a second historical monitoring information that has been measured in a short period of time in the first historical monitoring information, and use it to predict the physical condition after confirming that the second historical monitoring information is reliable, so as to ensure the accuracy of the user's physical condition prediction. Next, the judgment and execution process of this part will be introduced in conjunction with the specific embodiment drawings.

[0069] See also Figure 2 , which is a flow chart of an inquiry analysis method provided in an embodiment of the present application, comprising Figure 2 It can be seen that after obtaining the first historical monitoring information, it is necessary to determine whether there is second historical monitoring information therein. The second historical monitoring information is used to characterize the information monitored by the user through the monitoring device within the preset past time period in response to the query interaction start execution. Taking the preset past time period of six hours as an example, the second historical monitoring information is the physiological parameters measured by the monitoring device within six hours before the user triggers the query trigger button (issuing the query interaction start instruction). If there is no monitoring information within the first six hours in the user's first historical monitoring information, it can be determined that there is no second historical monitoring information in the first historical monitoring information. At this time, the user's current physiological parameter indicators are predicted based on the monitoring information of the user in a longer period of time.

[0070] In actual application scenarios, the prediction of physiological parameter indicators can be completed through parameter prediction models. The parameter prediction model predicts the user's current physiological parameter indicators by analyzing multiple sets of monitoring information predicted by the user in different time periods in the past, and by the changing patterns and fluctuation trends of the parameters in the multiple sets of monitoring information. For example, taking uric acid value as an example of monitoring information, for users with obesity problems, if the user's uric acid value monitored each time in the past period is gradually decreasing, it indicates that the user's obesity problem may be alleviated, and his uric acid value is in a stable decreasing range. In this way, by analyzing the changing patterns of uric acid values, the user's current uric acid value (i.e., the current physiological parameter indicator) can be predicted.

[0071] Furthermore, based on the predicted current physiological parameter indicators and combined with the user's basic vital signs information (such as age, gender, weight, etc.), a preliminary inquiry analysis is performed through a language model to generate a first inquiry response text for the user's voice inquiry information.

[0072] On the contrary, if it is determined that the second historical monitoring information exists in the first historical monitoring information, before applying the second historical monitoring information for preliminary inquiry analysis, it is necessary to first perform a distortion judgment on the second historical monitoring to determine the data reliability of the second historical monitoring information. It is understandable that in the actual physiological parameter monitoring scenario of the user, sometimes the user may take multiple measurements in a short period of time. At this time, it is necessary to screen out the true and reliable monitoring information from the monitoring information of multiple measurements.

[0073] For example, if it is determined that there are multiple sets of monitoring information (i.e., the second historical monitoring information) in the past six hours, then the multiple sets of monitoring information are subjected to distortion judgment to obtain corresponding distortion judgment results. As long as it is confirmed that there is at least one piece of monitoring information with a distortion judgment result of no in the distortion judgment result, the monitoring information can be used as the basic data for the preliminary inquiry analysis, and the first inquiry reply text can be generated based on the monitoring information and the voice inquiry information. Correspondingly, if it is found that the distortion judgment result of each piece of monitoring information in the multiple sets of monitoring information within six hours is yes, it indicates that the monitoring information in the past six hours is all unreliable information. At this time, the distorted monitoring information is removed from the first historical monitoring information, and the above-mentioned step of predicting the physiological parameter index is performed based on the first historical monitoring information after the second historical monitoring information is removed.

[0074] In one possible implementation, the distortion judgment of the second historical monitoring information can be completed based on a pre-set monitoring information range table. Taking uric acid value as an example, the uric acid value of men can be set between 150-420 μmol / L. If the user's uric acid value is measured to be much larger than this interval, it can be determined that the monitoring information is distorted, thereby ensuring the reliability of the original data. Similarly, the same distortion judgment and data elimination method for the second historical monitoring information can also be used and applied to the data processing of the first historical monitoring information to improve the data reliability of the first historical monitoring information.

[0075] S103: Perform temporary inquiry analysis based on the voice inquiry information to generate an interactive inquiry text, and after the first reply voice is played, play the inquiry voice to the user based on the interactive inquiry text; the interactive inquiry text is used to inquire the user whether the inquiry reply text needs to be expanded.

[0076] After studying the user inquiry scenarios, the applicant found that in most inquiry scenarios, the user's inquiries are often some immediate needs. For example, when the user is about to participate in a dinner party, due to the large number of food types involved in the dinner party scene, the user needs to know in advance the types of food that he cannot eat. Or when the user has plans to go out for exercise, the user needs to know the sports he can currently participate in. Therefore, in order to improve the comprehensiveness and intelligence of the inquiry response, after the preliminary inquiry analysis of step S103, the monitoring device needs to further perform a temporary inquiry analysis on the user's voice inquiry information, and generate an interactive inquiry text by analyzing the actual needs in the user's voice inquiry information and use it to play the inquiry voice to the user to inquire whether the user needs to further expand the content of the reply.

[0077] The interactive inquiry text is determined based on the inquiry category of the voice inquiry information. In the embodiment of the present application, the inquiry category is divided into food inquiry and sports inquiry. When the inquiry category is food inquiry, a food interactive inquiry text is generated, and when the inquiry category is sports inquiry, a sports interactive inquiry text is generated.

[0078] For example, if the user's voice inquiry information is: "What foods can I not eat?", the user's inquiry category is food-related inquiry. At this time, if you want to further ask the user whether you need to announce more inedible food categories, the corresponding food-related interactive inquiry text is: "Do you need to announce more inedible food categories?". Similarly, the user's sports inquiry can be "What sports can I currently do?", and its corresponding sports interactive inquiry text is "Do you need to announce more recommended sports projects?".

[0079] S104: When the user instructs to expand the inquiry reply text, a second inquiry reply text is generated according to the current inquiry time, the voice inquiry information, the first historical monitoring information and the positioning information of the monitoring device, and a second reply voice is played to the user based on the second inquiry reply text.

[0080] When the user instructs to expand the inquiry reply text, it indicates that the user needs more reply suggestions. At this time, the inquiry reply is expanded according to the current inquiry time, voice inquiry information, first historical monitoring information and the positioning information of the monitoring device, and a second inquiry reply text is generated, and the second reply voice is broadcast to the user. Among them, the voice inquiry information includes the inquiry text of the user at the time of inquiry and the actual inquiry voice. Through the user's actual inquiry voice, the region to which the user's accent belongs can be identified, and then the food category that the user may be involved in can be determined. This part will be introduced in detail later.

[0081] On the contrary, if the user's answer to the play inquiry is negative, the voice interaction process with the user is ended.

[0082] In one possible implementation, the user instructs the expansion of the inquiry reply text by confirming the played inquiry voice, or by triggering the expansion of the inquiry reply text through a detailed voice instruction such as "Please provide me with more detailed suggestions." This embodiment does not limit the method in which the user instructs the expansion of the inquiry reply text.

[0083] It should be noted that, as mentioned above, the inquiry analysis process can be completed by the language model built into the monitoring device or by the cloud server. When the language model is set on the cloud server, the monitoring device needs to upload the current inquiry timestamp, the device's GPS location information, the voice inquiry text content, and the actual inquiry voice to the server so that the server can expand the reply text.

[0084] As can be seen from step S103, the interactive inquiry text needs to be determined based on the inquiry category corresponding to the user's voice inquiry information. Similarly, the second inquiry reply text used to expand the reply also needs to be determined based on the inquiry category of the voice inquiry information.

[0085] Next, in conjunction with the accompanying drawings of specific embodiments, the reply text expansion methods corresponding to the two inquiry categories of food-related inquiries and sports-related inquiries will be introduced respectively.

[0086] See also Figure 3 , which is a flow chart of a method for generating a second inquiry reply text when the inquiry category is a food inquiry provided in an embodiment of the present application, which specifically includes the following steps:

[0087] S1041: When the inquiry category is the food inquiry, performing a geographical location analysis according to the positioning information of the monitoring device to obtain a first geographical category for the user;

[0088] S1042: Performing accent-based regional location analysis according to the voice query information to obtain a second regional category;

[0089] S1043: Predicting a first dining food category of the user within a preset future time period according to the first regional category, the second regional category, and the current inquiry time.

[0090] When the user's voice inquiry category is food-related, in order to ensure accurate analysis of the food categories that may appear in the user's future dining scenes, it is necessary to determine the user's first geographical category based on the GPS positioning information of the monitoring device, and analyze the geographical location of the user's accent based on the actual voice in the voice inquiry information to obtain the second geographical category.

[0091] On this basis, the user's current query time can be used to further determine the upcoming holidays (such as Qingming Festival, Dragon Boat Festival, Mid-Autumn Festival, etc.). By integrating the user's actual geographical location, the regional background of the user's hometown, and the current holiday trends, the user's first meal food category in the preset future period can be predicted, thus achieving comprehensive and accurate dining scene prediction.

[0092] In this step, based on specific holidays and meal times, it is necessary to predict dining scenes based on the similarities and differences between the first regional category and the second regional category. When the first regional category is different from the second regional category, it is necessary to predict the user's dining scene based on the weight coefficients corresponding to the two regional categories. Among them, because the user's eating habits are more affected by their subjective factors, the second weight coefficient determined based on accent analysis (i.e., the second weight coefficient corresponding to the second regional category) is greater than the first weight coefficient determined based on positioning information (i.e., the first weight coefficient corresponding to the first regional category).

[0093] Specifically, first, according to the first regional category and the second regional category, the types of food that may be involved in the two regional categories are predicted respectively. For example, assuming that the first regional category is the north, the staple foods involved are steamed buns, noodles and dumplings. In the predicted probability of food occurrence for the north, the probability of steamed buns is 50%, the probability of noodles is 30%, and the probability of dumplings is 20%. The second regional category is the south, and the staple foods involved are rice and porridge. The probability of rice is 60%, and the probability of porridge is 40%. On this basis, assuming that the first weight coefficient is 0.4 and the second weight coefficient is 0.6, the second weight coefficient and the first weight coefficient are weighted with the probability of occurrence of each food, respectively, to obtain the probability that each food category may eventually appear, and prioritize them, select the food categories that rank high, and use them as the first dining food category. In this example, the final probability of steamed buns is 20%, the probability of noodles is 12%, the probability of dumplings is 8%, the probability of rice is 36%, and the probability of porridge is 24%. The top three food categories are selected as the first dining food category, namely: rice, porridge and steamed buns.

[0094] On the contrary, if the first and second regional categories are the same, then any one of them can be used as the regional type to predict the category of dining food. For example, if the location information of the monitoring device is located in Guangdong (i.e. the first regional category is Guangdong), the user's accent recognition result is also Guangdong (i.e. the second regional category is Guangdong), and the inquiry time is the afternoon of Mid-Autumn Festival. The common food list in Guangdong during the Mid-Autumn Festival will be analyzed, and then the user's possible dining scene will be predicted to obtain the first dining food category.

[0095] S1044: Based on the first historical monitoring information and the basic vital sign information of the user, a second dining food category is screened out from the first dining food category, and the second inquiry reply text is generated based on the second dining food category; the second dining food category is used to characterize the food category that is not suitable for the user.

[0096] After predicting the user's future dining scenarios, based on the user's first historical monitoring information (uric acid level, blood pressure) and the user's basic vital signs information (age, weight, medical history, etc.), the food categories that the user is currently inedible are screened out from the first dining food category and determined as the second dining food category.

[0097] In one possible implementation, in order to ensure the conciseness of the voice broadcast, the second meal food category can also be divided into food types, for example, beef dishes can be classified as beef, and dishes containing seafood can be classified as seafood. When the second meal food category is broadcast, the specific cuisine name is not broadcast, only the actual food category is broadcast, so that the voice response is more intuitive and concise, and the user experience is optimized.

[0098] See also Figure 4 , which is a flow chart of a method for generating a second inquiry reply text when the inquiry category is a sports inquiry provided by an embodiment of the present application, and specifically includes the following steps:

[0099] S1045: When the inquiry category is the sports inquiry, obtain the real-time weather data of the user's current geographical location according to the positioning information of the monitoring device, and determine a possible exercise location based on the real-time weather data and the current geographical location.

[0100] The sports recommendation for users needs to refer not only to the actual physical condition of the user, but also to the weather of the day and the configuration of sports facilities around the user. Therefore, when the inquiry category is sports inquiry, it is necessary to obtain the real-time weather data of the user's current geographical location based on the GPS positioning information of the monitoring device, and determine the user's available sports locations based on the real-time weather data and the sports settings around the current geographical location.

[0101] For example, when the real-time weather data indicates rainy days and it is determined that there are indoor basketball courts and outdoor tennis courts in the surrounding geographical locations, the user's available exercise location is determined to be the indoor basketball court.

[0102] S1046: performing exercise preference analysis according to the voice inquiry information to determine the user's exercise preference, and generating exercise recommendation items based on the exercise preference, the available exercise locations, the first historical monitoring information, and the user's basic vital sign information;

[0103] S1047: Generate the second inquiry response text based on the recommended exercise item.

[0104] Furthermore, after determining the current possible exercise locations around the user, it is necessary to further analyze the exercise preferences based on the user's voice inquiry information, and at the same time combine the user's exercise preferences, possible exercise locations, the first historical monitoring information, and the user's basic vital signs information to generate the final exercise recommendation items as the second inquiry response text.

[0105] The user's voice inquiry information about sports items usually includes the user's sports preferences. For example, when the user asks "Can I go for a run at 3 pm?", it can be determined that the user's sports preference is aerobic exercise. When generating sports recommendations, the user's physical condition and the aerobic exercise items that can be carried out in the current surrounding environment can be combined to output sports recommendations to the user.

[0106] The embodiment of the present application provides a monitoring device control method with intelligent voice interaction function, which includes: in response to the query interaction start instruction, obtaining the user's voice query information and the first historical monitoring information, performing preliminary query analysis according to the voice query information and the first historical monitoring information, generating a first query reply text, and playing the first reply voice to the user based on the first query reply text. Perform temporary query analysis according to the voice query information, generate an interactive query text, and after the first reply voice is played, play the query voice to the user based on the interactive query text. When the user's reply to playing the query voice is confirmation, a second query reply text is generated according to the current query time, the voice query information, the first historical monitoring information and the positioning information of the monitoring device, and the second reply voice is played to the user based on the second query reply text. In this way, the user can trigger the query interaction start instruction language monitoring device to perform voice interaction by himself, and the monitoring device realizes intelligent voice interaction with the user through the user's historical monitoring information and the user's actual voice query information, without relying on the real-time measurement of the monitoring results. At the same time, on this basis, further temporary inquiry analysis will be carried out in combination with the user's voice inquiry information, and when the user confirms, the reply content will be further expanded, thereby providing users with comprehensive intelligent voice interaction functions and effectively improving the user experience.

[0107] The following is an introduction to a monitoring device with an intelligent voice interaction function provided in an embodiment of the present application. The monitoring device with an intelligent voice interaction function described below and the monitoring device control method with an intelligent voice interaction function described above can refer to each other.

[0108] See also Figure 5 , which is a schematic diagram of the structure of a monitoring device with intelligent voice interaction function provided by an embodiment of the present application, specifically including the following modules:

[0109] The acquisition module 100 is used to obtain the user's voice inquiry information and the first historical monitoring information in response to the inquiry interaction start instruction;

[0110] A first question-and-answer module 200, configured to perform a preliminary inquiry analysis based on the voice inquiry information and the first historical monitoring information, generate a first inquiry reply text, and play a first reply voice to the user based on the first inquiry reply text;

[0111] The second question-and-answer module 300 is used to perform temporary inquiry analysis based on the voice inquiry information, generate an interactive inquiry text, and play an inquiry voice to the user based on the interactive inquiry text after the first reply voice is played; the interactive inquiry text is used to inquire the user whether the inquiry reply text needs to be expanded;

[0112] The third question and answer module 400 is used to generate a second inquiry reply text according to the current inquiry time, the voice inquiry information, the first historical monitoring information and the positioning information of the monitoring device when the user instructs to expand the inquiry reply text, and play a second reply voice to the user based on the second inquiry reply text.

[0113] In a possible implementation, the first question-and-answer module 200 is specifically configured to:

[0114] Determine whether there is second historical monitoring information in the first historical monitoring information; the second historical monitoring information is used to represent monitoring information for the user within a preset past time period in response to the query interaction start instruction;

[0115] When it is determined that the second historical monitoring information does not exist in the first historical monitoring information, predicting a current physiological parameter index of the user based on the first historical monitoring information;

[0116] A preliminary inquiry analysis is performed based on the current physiological parameter index and the basic vital sign information of the user to generate the first inquiry response text.

[0117] In a possible implementation, the third question-and-answer module 400 is specifically configured to:

[0118] Determining the inquiry category of the voice inquiry information;

[0119] The second inquiry reply text is generated according to the inquiry category, the current inquiry time, the first historical monitoring information, and the location information of the monitoring device.

[0120] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same and similar parts between the various embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the method and the monitoring device, since they are basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The method and the monitoring device described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components indicated as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. A person of ordinary skill in the art can understand and implement it without creative work.

[0121] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for controlling a monitoring device with intelligent voice interaction function, characterized in that: Applied to the monitoring device; the method comprises: In response to the inquiry interaction start instruction, obtaining the user's voice inquiry information and the first historical monitoring information; Performing a preliminary inquiry analysis based on the voice inquiry information and the first historical monitoring information, generating a first inquiry reply text, and playing a first reply voice to the user based on the first inquiry reply text; Performing a temporary inquiry analysis based on the voice inquiry information to generate an interactive inquiry text, and after the first reply voice is played, playing the inquiry voice to the user based on the interactive inquiry text; the interactive inquiry text is used to inquire the user whether the inquiry reply text needs to be expanded; When the user instructs to expand the inquiry reply text, a second inquiry reply text is generated according to the current inquiry time, the voice inquiry information, the first historical monitoring information and the positioning information of the monitoring device, and a second reply voice is played to the user based on the second inquiry reply text.

2. The method according to claim 1, characterized in that The performing preliminary inquiry analysis according to the voice inquiry information and the first historical monitoring information to generate a first inquiry reply text includes: Determine whether there is second historical monitoring information in the first historical monitoring information; the second historical monitoring information is used to represent monitoring information for the user within a preset past time period in response to the query interaction start instruction; When it is determined that the second historical monitoring information does not exist in the first historical monitoring information, predicting a current physiological parameter index of the user based on the first historical monitoring information; A preliminary inquiry analysis is performed based on the current physiological parameter index and the basic vital sign information of the user to generate the first inquiry response text.

3. The method according to claim 2, characterized in that After determining whether the second historical monitoring information exists in the first historical monitoring information, the method further includes: When it is determined that the second historical monitoring information exists in the first historical monitoring information, performing distortion judgment on the second historical monitoring information to obtain a distortion judgment result; If there is at least one second historical monitoring information in the distortion judgment result in which the distortion judgment result is negative, a preliminary inquiry analysis is performed based on the second historical monitoring information and the voice inquiry information to generate the first inquiry reply text; If the distortion judgment result does not include the second historical monitoring information with a negative judgment result, the second historical monitoring information is removed, and the first inquiry reply text is generated based on the first historical monitoring information after the second historical monitoring information is removed.

4. The method according to claim 1, characterized in that When the user instructs to extend the inquiry reply text, generating a second inquiry reply text according to the current inquiry time, the voice inquiry information, the first historical monitoring information, and the location information of the monitoring device, includes: Determining the inquiry category of the voice inquiry information; The second inquiry reply text is generated according to the inquiry category, the current inquiry time, the first historical monitoring information, and the location information of the monitoring device.

5. The method according to claim 4, characterized in that The inquiry categories include: food-related inquiries; The generating the second inquiry reply text according to the inquiry category, the current inquiry time, the first historical monitoring information, and the location information of the monitoring device includes: When the inquiry category is the food inquiry, performing a geographical location analysis according to the positioning information of the monitoring device to obtain a first geographical category for the user; Performing accent-based regional location analysis according to the voice query information to obtain a second regional category; Predicting a first dining food category of the user within a preset future time period according to the first regional category, the second regional category and the current inquiry time; Based on the first historical monitoring information and the basic vital sign information of the user, a second dining food category is screened out from the first dining food category, and the second inquiry reply text is generated based on the second dining food category; the second dining food category is used to characterize the food category that is not suitable for the user.

6. The method according to claim 4, characterized in that The inquiry categories include: sports inquiries; The generating the second inquiry reply text according to the inquiry category, the current inquiry time, the first historical monitoring information, and the location information of the monitoring device includes: When the inquiry category is the sports inquiry, obtaining real-time weather data of the user's current geographical location according to the positioning information of the monitoring device, and determining a possible exercise location based on the real-time weather data and the current geographical location; Performing exercise preference analysis according to the voice inquiry information to determine the user's exercise preference, and generating exercise recommendation items based on the exercise preference, the available exercise locations, the first historical monitoring information, and the user's basic vital sign information; Based on the exercise recommendation item, the second inquiry response text is generated.

7. The method according to claim 5, characterized in that The predicting, based on the first regional category, the second regional category, and the current query time, a first dining food category of the user in a preset future time period includes: Determining whether the first regional category is the same as the second regional category; When it is determined that the first regional category is different from the second regional category, predicting the first dining food category according to a second weight coefficient of the second regional category, a first weight coefficient of the first regional category, and the current inquiry time; the second weight coefficient is greater than the first weight coefficient; When it is determined that the first regional category is the same as the second regional category, the first meal food category is predicted according to any one of the second regional category and the regional category and the current inquiry time.

8. The method according to any one of claim 4, characterized in that: The interactive inquiry text includes: food interactive inquiry text and exercise interactive inquiry text; The performing temporary inquiry analysis according to the voice inquiry information to generate an interactive inquiry text comprises: When the inquiry category is a food inquiry, the food interactive inquiry text is generated; the food interactive inquiry text is used to inquire the user whether more inedible food categories need to be reported; When the inquiry category is a sports inquiry, the sports interaction inquiry text is generated; the sports interaction inquiry text is used to inquire the user whether more recommended sports items need to be reported.

9. The method according to claim 1, characterized in that: After playing the inquiry voice to the user based on the interactive inquiry text, the method further includes: When the user's response to the played inquiry voice is negative, the voice interaction with the user is terminated.

10. A monitoring device with intelligent voice interaction function, characterized in that: include: An acquisition module, configured to acquire the user's voice inquiry information and first historical monitoring information in response to an inquiry interaction start instruction; a first question-and-answer module, configured to perform a preliminary inquiry analysis based on the voice inquiry information and the first historical monitoring information, generate a first inquiry reply text, and play a first reply voice to the user based on the first inquiry reply text; A second question-and-answer module is used to perform temporary inquiry analysis based on the voice inquiry information, generate an interactive inquiry text, and play an inquiry voice to the user based on the interactive inquiry text after the first reply voice is played; the interactive inquiry text is used to inquire the user whether the inquiry reply text needs to be expanded; The third question and answer module is used to generate a second inquiry reply text according to the current inquiry time, the voice inquiry information, the first historical monitoring information and the positioning information of the monitoring device when the user instructs to expand the inquiry reply text, and play a second reply voice to the user based on the second inquiry reply text.

Images