Artificial intelligence-based verbal skill generation method and device, computer equipment and medium

Abstract

The invention belongs to the field of artificial intelligence and the field of digital medical treatment, and relates to an artificial intelligence-based verbal skill generation method, which comprises the following steps of: obtaining a speaking content sent by a client, and performing information identification on the speaking content based on an identification rule to obtain key information; chronic disease type information of the customer is acquired; processing the key information and the chronic disease type information based on a data analysis model to obtain an analysis result; extracting target information related to the chronic disease type from the analysis result; generating suggestion information based on the target information; obtaining a target verbal skill template corresponding to the chronic disease type information; and filling the target verbal skill template based on the analysis result, the target information and the suggestion information to obtain a target verbal skill. The invention further provides a verbal skill generation device based on artificial intelligence, computer equipment and a storage medium. In addition, the invention also relates to a block chain technology, and the target verbal skill can be stored in a block chain. The generation efficiency of the target verbal skill is improved, and the data accuracy of the target verbal skill is ensured.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence development technology and the field of digital medicine, and in particular to an artificial intelligence-based speech generation method, device, computer equipment and storage medium. Background Art

[0002] With the accelerated development of the global population aging and the continuous advancement of medical technology leading to a significant increase in life expectancy, chronic diseases have become one of the main challenges affecting the health and quality of life of the elderly. Among them, metabolic diseases such as diabetes and hypertension are particularly prominent. At the same time, chronic diseases caused by unhealthy lifestyles such as obesity and fatty liver are also showing a high incidence.

[0003] Traditionally, chronic disease management relies on community health service agencies, which mainly rely on medical staff to establish health records for chronic disease patients, regularly monitor physiological indicators such as blood pressure and blood sugar, and generate management scripts that match chronic disease patients based on personal experience. This management method is not only inefficient, but also prone to low data accuracy due to human factors. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose a method, device, computer equipment and storage medium for generating speech based on artificial intelligence, so as to solve the technical problems that the existing chronic disease management method is inefficient and easily leads to low data accuracy due to human factors, in which medical staff establish health records for chronic disease patients, regularly monitor physiological indicators such as blood pressure and blood sugar, and generate management speech matching chronic disease patients based on personal experience.

[0005] In order to solve the above technical problems, the embodiment of the present application provides a method for generating speech based on artificial intelligence, which adopts the following technical solution:

[0006] Obtain the speech content sent by the customer, and perform information recognition on the speech content based on the preset recognition rules to obtain the corresponding key information;

[0007] Obtaining information on the type of chronic diseases of the customer;

[0008] Processing the key information and the chronic disease type information based on a preset data analysis model to obtain corresponding analysis results;

[0009] Extracting target information related to the chronic disease type information from the analysis results;

[0010] generating corresponding suggestion information based on the target information;

[0011] Obtain a target speech template corresponding to the chronic disease type information;

[0012] The target speech template is filled based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech.

[0013] Furthermore, the step of obtaining a target speech template corresponding to the chronic disease type information specifically includes:

[0014] Call the preset speech template library;

[0015] Generate corresponding search conditions based on the chronic disease type information;

[0016] Searching the speech template library based on the search condition to obtain a first speech template matching the search condition;

[0017] Determine whether there are multiple first speech templates;

[0018] If so, the first speech is screened based on a preset screening algorithm to obtain the target speech template corresponding to the chronic disease type information.

[0019] Furthermore, the step of screening the first speech based on a preset screening algorithm to obtain the target speech template corresponding to the chronic disease type information specifically includes:

[0020] Obtaining customer feedback scores for each of the first speech templates, and obtaining execution rates of each of the first speech templates;

[0021] Calculating the matching score of each of the first speech templates;

[0022] Generate a template score for each of the first speech templates based on the customer feedback score, the execution rate, and the matching score;

[0023] Filter out the designated template score with the highest score from all the template scores;

[0024] Obtain a second speech template corresponding to the specified template score, and use the second speech template as the target speech template.

[0025] Furthermore, the step of generating a template score for each of the first speech templates based on the customer feedback score, the execution rate, and the matching score specifically includes:

[0026] Call the preset weight setting algorithm;

[0027] Generate a first weight corresponding to the customer feedback factor, a second weight corresponding to the execution rate factor, and a third weight corresponding to the matching factor based on the weight setting algorithm;

[0028] Call the preset score calculation formula;

[0029] Based on the first weight, the second weight and the third weight, the customer feedback score, execution rate and matching score of each of the first speech templates are calculated using the score calculation formula to obtain the template score of each of the first speech templates.

[0030] Furthermore, the step of filling the target speech template based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech specifically includes:

[0031] Determine an analysis result filling area, a related information filling area, and a suggestion information filling area in the target speech template;

[0032] Fill the analysis result into the analysis result filling area, fill the target information into the related information filling area, and fill the suggestion information into the suggestion information filling area to obtain a corresponding first speech;

[0033] Calling a preset audit component to audit the first speech;

[0034] If the first speech is approved, generating the target speech based on the first speech;

[0035] If the first speech fails to pass the review, error information of the first speech is obtained, and the first speech is modified based on the review component to obtain a corresponding second speech;

[0036] The target speech is generated based on the second speech.

[0037] Furthermore, after the step of filling the target speech template based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech, the method further includes:

[0038] Acquire customer information of the customer, and acquire the education status of the customer based on the customer information;

[0039] Determine whether the educational situation meets the preset educational conditions;

[0040] If yes, the target speech is split into sentences to obtain a corresponding third speech;

[0041] Performing language polishing on the third speech to obtain a corresponding fourth speech;

[0042] Adding emotional elements to the fourth speech to obtain a corresponding fifth speech;

[0043] The fifth script is sent to the customer.

[0044] Furthermore, the step of sending the fifth speech script to the customer specifically includes:

[0045] Determine a corresponding sending time based on the customer information;

[0046] Identify target communication channels;

[0047] Based on the sending time and the target communication channel, the fifth script is sent to the customer.

[0048] In order to solve the above technical problems, the embodiment of the present application also provides a speech generation device based on artificial intelligence, which adopts the following technical solution:

[0049] The recognition module is used to obtain the speech content sent by the customer and perform information recognition on the speech content based on the preset recognition rules to obtain the corresponding key information;

[0050] The first acquisition module is used to obtain the chronic disease type information of the customer;

[0051] A first processing module is used to process the key information and the chronic disease type information based on a preset data analysis model to obtain corresponding analysis results;

[0052] An extraction module, used to extract target information related to the chronic disease type information from the analysis results;

[0053] A first generating module, configured to generate corresponding suggestion information based on the target information;

[0054] A second acquisition module is used to acquire a target speech template corresponding to the chronic disease type information;

[0055] The second generation module is used to fill in the target speech template based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech.

[0056] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0057] Obtain the speech content sent by the customer, and perform information recognition on the speech content based on the preset recognition rules to obtain the corresponding key information;

[0058] Obtaining information on the type of chronic diseases of the customer;

[0059] Processing the key information and the chronic disease type information based on a preset data analysis model to obtain corresponding analysis results;

[0060] Extracting target information related to the chronic disease type information from the analysis results;

[0061] generating corresponding suggestion information based on the target information;

[0062] Obtain a target speech template corresponding to the chronic disease type information;

[0063] The target speech template is filled based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech.

[0064] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0065] Obtain the speech content sent by the customer, and perform information recognition on the speech content based on the preset recognition rules to obtain the corresponding key information;

[0066] Obtaining information on the type of chronic diseases of the customer;

[0067] Processing the key information and the chronic disease type information based on a preset data analysis model to obtain corresponding analysis results;

[0068] Extracting target information related to the chronic disease type information from the analysis results;

[0069] generating corresponding suggestion information based on the target information;

[0070] Obtain a target speech template corresponding to the chronic disease type information;

[0071] The target speech template is filled based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech.

[0072] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0073] This application first obtains the speech content sent by the customer, and performs information recognition on the speech content based on preset recognition rules to obtain corresponding key information; then obtains the chronic disease type information of the customer; and processes the key information and the chronic disease type information based on a preset data analysis model to obtain corresponding analysis results; then extracts target information related to the chronic disease type information from the analysis results; subsequently generates corresponding suggestion information based on the target information; further obtains a target speech template corresponding to the chronic disease type information; finally, fills in the target speech template based on the analysis results, the target information and the suggestion information to obtain the corresponding target speech. After obtaining the speech content sent by the customer, this application will automatically and intelligently identify the speech content based on the recognition rules to collect the corresponding key information, and then process the key information and the customer's chronic disease type information based on the use of the data analysis model to obtain the analysis result, and then extract the target information related to the chronic disease type information from the analysis result, and generate the corresponding suggestion information based on the target information, and finally automatically fill in the target speech template corresponding to the chronic disease type information based on the analysis result, the target information and the suggestion information, so as to quickly and accurately generate the target speech that matches the speech content sent by the customer, thereby eliminating the need for medical staff to manually collect physiological indicators and generate management speech based on personal experience, effectively improving the generation efficiency of the target speech, and ensuring the data accuracy of the generated target speech. It is conducive to providing customers with more accurate and effective health management suggestions based on the subsequent use of target speech, thereby improving the customer's user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0075] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0076] Figure 2 A flowchart of an embodiment of a method for generating speech based on artificial intelligence according to the present application;

[0077] Figure 3 It is a structural schematic diagram of an embodiment of an artificial intelligence-based speech generation device according to the present application;

[0078] Figure 4 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0079] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0080] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0081] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0082] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0083] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0084] The terminal device 101 can be various electronic devices with a display screen and supporting web browsing. In addition to the laptop computer 1011, the tablet computer 1012 or the mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop computer, a desktop computer, etc.

[0085] The server 103 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal device 101 .

[0086] It should be noted that the artificial intelligence-based speech generation method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the artificial intelligence-based speech generation device is generally set in the server / terminal device.

[0087] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0088] Continue to refer Figure 2 , showing a flow chart of an embodiment of the method for generating speech based on artificial intelligence according to the present application. According to different needs, the order of the steps in the flowchart can be changed, and some steps can be omitted. The method for generating speech based on artificial intelligence provided in the embodiment of the present application can be applied to any scenario where speech generation is required, and the method for generating speech based on artificial intelligence can be applied to products in these scenarios, for example, in the field of digital medicine, corresponding management speech is generated for the speech content provided by the customer. The method for generating speech based on artificial intelligence comprises the following steps:

[0089] Step S201, obtaining the speech content sent by the customer, and performing information recognition on the speech content based on a preset recognition rule to obtain corresponding key information.

[0090] In this embodiment, the electronic device (for example, Figure 1The server / terminal device shown in the figure) can obtain the speech content sent by the customer through a wired connection or a wireless connection. It should be noted that the above-mentioned wireless connection method may include but is not limited to 3G / 4G / 5G connection, Wi F i connection, Bluetooth connection, Wi MAX connection, Zigbee connection, UWB (ultra wi deband) connection, and other wireless connection methods currently known or to be developed in the future. This application can be applied to the business scenario of generating corresponding management speech for the speech content provided by the customer in the field of digital medicine. The above-mentioned customers may be specific chronic disease personnel. After the health manager adds the WeChat of a specific chronic disease person through the enterprise WeChat, an enterprise WeChat management group is established, and popular science information is sent to the customer to guide the customer to speak every day. Among them, the speech content may include the customer's diet information, exercise information, measurement (blood sugar, blood pressure, weight, waist circumference) information, etc., and the speech content may be in the form of text or pictures. And set up clear sending rules in the enterprise WeChat management group, such as fixed formats or tags (such as "#Diet#", "#Exercise#", "#Measurement#") to mark the content sent by the customer for automatic identification and classification.

[0091] In addition, the circle robot function of WeChat for Business can be used to automatically capture and classify the above-mentioned speeches to ensure the accuracy and completeness of the data. Specifically, configure the circle robot in the WeChat for Business background, and set it to automatically listen to and capture the speeches in the management group. Then use the natural language processing (NLP) and rule engine functions of the circle robot to automatically identify and classify the collected information. For example, the content is classified into different categories such as diet, exercise, and measurement according to preset tags or keywords. After that, for key data (such as measurement values), the circle robot can perform preliminary data verification to ensure that the values ​​are within a reasonable range, such as whether blood sugar values ​​and blood pressure values ​​are within the normal or expected range. The classified and verified data will be subsequently stored in the company's database or cloud storage service to ensure the accuracy and completeness of the data.

[0092] In addition, the process of identifying the speech content based on the preset recognition rules to obtain the corresponding key information includes: 1) If the speech content is in the form of a picture, use OCR (optical character recognition) technology to identify the text information in the picture, such as the nutrition information table on the food packaging, the readings on the measuring instrument, etc. In addition, for food pictures, image recognition technology (such as deep learning models) can be used to use a pre-trained model for food recognition to identify the type of meal, portion and cooking method. 2) If the speech content is in text form, use NLP technology for word segmentation, part-of-speech tagging and semantic understanding to extract key information such as food names, exercise types, measurement values, etc.

[0093] Step S202, obtaining the chronic disease type information of the customer.

[0094] In this embodiment, a chronic disease type label (such as diabetes, hypertension) is maintained for each customer in the database. By obtaining the customer's information, such as the customer's name, the chronic disease type label that matches the customer information can be queried from the database and used as the customer's chronic disease type information.

[0095] Step S203: Process the key information and the chronic disease type information based on a preset data analysis model to obtain corresponding analysis results.

[0096] In this embodiment, the above-mentioned data analysis model can specifically adopt the GPT model. By starting a pre-configured GPT model instance, and loading the pre-trained professional knowledge base in the field of chronic disease management into the GPT model. The above-mentioned knowledge base may at least include management guidelines, dietary recommendations, exercise guidance, psychological support and other contents for various chronic diseases. Specifically, by inputting the key information and the chronic disease type information into the data analysis model, a basic analysis of the key information and the chronic disease type information is performed through the data analysis model to understand the customer's health status, living habits, dietary preferences and other aspects, and output the corresponding analysis results.

[0097] Step S204, extracting target information related to the chronic disease type information from the analysis results.

[0098] In this embodiment, the attention mechanism within the GPT model can also be used to capture key information points in the input data. Specifically, the most critical information points that match the above-mentioned chronic disease type information and need to be conveyed to customers can be further extracted from the analysis results, such as eating habits that need to be improved, recommended types of exercise, and emergency response measures, so as to obtain the above-mentioned target information.

[0099] Step S205: Generate corresponding suggestion information based on the target information.

[0100] In this embodiment, after obtaining the target information, the GPT model can be further used to generate specific health management suggestions, i.e., the above-mentioned suggestion information, based on the above target information and combined with health management knowledge and experience. For example, if the customer's blood sugar level is high, it can be suggested that he reduce sugar intake, increase exercise, etc.

[0101] Step S206, obtaining a target speech template corresponding to the chronic disease type information.

[0102] In this embodiment, the specific implementation process of obtaining the target speech template corresponding to the chronic disease type information will be described in further detail in subsequent specific embodiments of the present application and will not be elaborated here.

[0103] Step S207, filling in the target speech template based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech.

[0104] In this embodiment, the target speech template is filled based on the analysis results, the target information and the suggestion information to obtain the specific implementation process of the corresponding target speech. This application will provide further detailed descriptions of this in subsequent specific embodiments and will not be elaborated here.

[0105] After obtaining the speech content sent by the customer, this application will automatically and intelligently identify the speech content based on the recognition rules to collect the corresponding key information, and then process the key information and the customer's chronic disease type information based on the use of the data analysis model to obtain the analysis result, and then extract the target information related to the chronic disease type information from the analysis result, and generate the corresponding suggestion information based on the target information, and finally automatically fill in the target speech template corresponding to the chronic disease type information based on the analysis result, the target information and the suggestion information, so as to quickly and accurately generate the target speech that matches the speech content sent by the customer, thereby eliminating the need for medical staff to manually collect physiological indicators and generate management speech based on personal experience, effectively improving the generation efficiency of the target speech, and ensuring the data accuracy of the generated target speech. It is conducive to providing customers with more accurate and effective health management suggestions based on the subsequent use of target speech, thereby improving the customer's user experience.

[0106] In some optional implementations, step S206 includes the following steps:

[0107] Call the preset speech template library.

[0108] In this embodiment, a rich speech template library is pre-built, which includes speech templates corresponding to different chronic disease types, health management scenarios and suggestion types. The above speech template library supports flexible search and filtering functions.

[0109] Generate corresponding search conditions based on the chronic disease type information.

[0110] In this embodiment, corresponding search conditions can be set according to the above chronic disease type information. For example, for patients with hypertension, templates containing keywords such as "lowering blood pressure" and "diet control" can be searched. Alternatively, the above chronic disease type information can be directly used as the search condition.

[0111] The speech template library is searched based on the search condition to obtain a first speech template that matches the search condition.

[0112] In this embodiment, the above-mentioned search condition can be used to perform a match in the speech template library to obtain a first speech template that matches the search condition.

[0113] Determine whether there are multiple first speech templates.

[0114] In this embodiment, the number of the first speech templates may be 1 or more. If the number of the first speech template is 1, the first speech template is directly used as the target speech template.

[0115] If so, the first speech is screened based on a preset screening algorithm to obtain the target speech template corresponding to the chronic disease type information.

[0116] In this embodiment, the above-mentioned specific implementation process of screening and processing the first speech based on the preset screening algorithm to obtain the target speech template corresponding to the chronic disease type information will be further described in detail in the subsequent specific embodiments of this application and will not be elaborated in detail here.

[0117] This application calls a preset speech template library; then generates corresponding search conditions based on the chronic disease type information; then searches the speech template library based on the search conditions to obtain a first speech template that matches the search conditions; then determines whether the number of the first speech templates is multiple; if so, the first speech is screened based on a preset screening algorithm to obtain the target speech template corresponding to the chronic disease type information. This application generates corresponding search conditions based on the chronic disease type information, then searches the preset speech template library based on the search conditions to obtain a first speech template that matches the search conditions. If it is detected that the number of the first speech templates is multiple, the first speech will be intelligently screened based on the preset screening algorithm to obtain the target speech template corresponding to the chronic disease type information. By using the screening algorithm, the target speech template that best matches the customer's chronic disease type information can be accurately screened from the speech template library, effectively improving the accuracy and applicability of the target speech template. The target speech template obtained can be subsequently filled in based on the analysis results, the target information and the suggestion information to obtain the corresponding target speech, which can effectively improve the accuracy of the generated target speech.

[0118] In some optional implementations of this embodiment, the screening process of the first speech based on a preset screening algorithm to obtain the target speech template corresponding to the chronic disease type information includes the following steps:

[0119] Obtain the customer feedback score of each of the first speech templates, and obtain the execution rate of each of the first speech templates.

[0120] In this embodiment, the historical use effect data of each speech template contained in the speech template library can be obtained, and the specified historical use effect data of each first speech template can be extracted from the historical use effect data, and then the customer feedback score of each first speech template can be screened out from the specified historical use effect data, and the execution rate of each first speech template can be obtained. Among them, the above-mentioned customer feedback score is the average evaluation score returned by the customer after receiving the reply speech generated based on the first speech template. The above-mentioned execution rate is the number of times the first speech template is used. In addition, the above-mentioned historical use effect data may at least include the customer feedback score and execution rate of the speech template.

[0121] Calculate the matching score of each of the first speech templates.

[0122] In this embodiment, the matching scores of the first speech templates can be obtained by calculating the matching degree between the chronic type information and the keywords contained in the first speech templates.

[0123] Based on the customer feedback score, the execution rate and the matching score, a template score for each of the first speech templates is generated.

[0124] In this embodiment, the specific implementation process of generating the template score of each of the first speech templates based on the customer feedback score, the execution rate and the matching score will be described in further detail in subsequent specific embodiments of the present application and will not be elaborated on here.

[0125] The designated template score with the highest score is selected from all the template scores.

[0126] In this embodiment, numerical comparison may be performed on all obtained template scores to screen out the designated template score with the highest score from all the template scores.

[0127] Obtain a second speech template corresponding to the specified template score, and use the second speech template as the target speech template.

[0128] In this embodiment, the second speech template corresponding to the specified template score can be selected from all the first speech templates.

[0129] The present application obtains the customer feedback score of each of the first speech templates and the execution rate of each of the first speech templates; then calculates the matching score of each of the first speech templates; then generates the template score of each of the first speech templates based on the customer feedback score, the execution rate and the matching score; subsequently selects the designated template score with the highest score from all the template scores; finally obtains the second speech template corresponding to the designated template score, and uses the second speech template as the target speech template. Since the target speech template obtained is the speech template with the best comprehensive score corresponding to the customer feedback score, the execution rate and the matching score among all the first speech templates, the determination accuracy and intelligence of the target speech template are effectively improved. The target speech template obtained is subsequently filled based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech, which can effectively improve the accuracy of the generated target speech.

[0130] In some optional implementations, generating a template score for each of the first speech templates based on the customer feedback score, the execution rate, and the matching score includes the following steps:

[0131] Call the preset weight setting algorithm.

[0132] In this embodiment, there is no specific limitation on the selection of the above-mentioned weight setting algorithm, and specifically, any one of principal component analysis, entropy method, AHP hierarchical analysis method, factor analysis method, standardization method, etc. can be adopted.

[0133] Based on the weight setting algorithm, a first weight corresponding to the customer feedback factor, a second weight corresponding to the execution rate factor, and a third weight corresponding to the matching factor are generated.

[0134] In this embodiment, weight calculation processing can be performed on all factors corresponding to the speech template (customer feedback factor, execution rate factor and matching factor) according to the processing steps of calculating weights matched in the above-mentioned weight setting algorithm, so as to obtain a first weight corresponding to the customer feedback factor, a second weight corresponding to the execution rate factor, and a third weight corresponding to the matching factor.

[0135] Call the preset score calculation formula.

[0136] In this embodiment, the above score calculation formula may specifically include: S=a*X+b*Y+c*Z, where S is the algorithm score, X is the customer feedback score, a is the first weight, Y is the execution rate, b is the second weight, Z is the matching score, and c is the third weight.

[0137] Based on the first weight, the second weight and the third weight, the customer feedback score, execution rate and matching score of each of the first speech templates are calculated using the score calculation formula to obtain the template score of each of the first speech templates.

[0138] In this embodiment, the template score of each of the first speech templates can be obtained by substituting the first weight, the second weight, the third weight, the customer feedback score, the execution rate and the matching score of each of the first speech templates into the corresponding positions in the above-mentioned score calculation formula for calculation and processing.

[0139] This application calls a preset weight setting algorithm; then generates a first weight corresponding to the customer feedback factor, a second weight corresponding to the execution rate factor, and a third weight corresponding to the matching factor based on the weight setting algorithm; then calls a preset score calculation formula; subsequently, based on the first weight, the second weight, and the third weight, the customer feedback score, execution rate, and matching score of each of the first speech templates are calculated and processed using the score calculation formula to obtain the template score of each of the first speech templates. This application generates a first weight corresponding to the customer feedback factor, a second weight corresponding to the execution rate factor, and a third weight corresponding to the matching factor based on the weight setting algorithm, and then calculates and processes the customer feedback score, execution rate, and matching score of each of the first speech templates based on the score calculation formula, so that the template score of each of the first speech templates can be quickly and accurately generated, effectively ensuring the data accuracy of the template score of each of the first speech templates generated. It is beneficial to subsequently screen out the final target speech template from all the template scores according to the template scores of each of the first speech templates, thereby improving the accuracy of determining the target speech template.

[0140] In some optional implementations, step S207 includes the following steps:

[0141] Determine the analysis result filling area, related information filling area and suggestion information filling area in the target speech template.

[0142] In this embodiment, the above-mentioned target speech template is a template written according to the actual speech construction requirements for chronic disease management. The target speech template includes a fixed part (such as a greeting, a closing remark, etc.) and a variable part, and the variable part includes at least an analysis result filling area, a related information filling area, and a suggestion information filling area. Among them, the above-mentioned analysis result filling area is a pre-set area for inputting analysis results, the above-mentioned related information filling area is a pre-set area for inputting relevant information matching the analysis results, and the above-mentioned suggestion information filling area is a pre-set area for inputting suggestion information.

[0143] The analysis result is filled into the analysis result filling area, the target information is filled into the related information filling area, and the suggestion information is filled into the suggestion information filling area to obtain a corresponding first speech.

[0144] In this embodiment, according to the position information corresponding to the analysis result filling area, the relevant information filling area and the suggested information filling area in the target speech template, the analysis result can be filled into the analysis result filling area, the target information can be filled into the relevant information filling area, and the suggested information can be filled into the suggested information filling area, thereby obtaining the corresponding first speech.

[0145] The preset audit component is called to audit the first speech.

[0146] In this embodiment, the audit component is specifically a pre-built integrated natural language processing (NLP) technology, which is a functional module that can automatically identify and correct common grammatical errors, spelling errors, etc. Among them, a set of NLP-based audit rule bases are also established, including common grammatical error patterns, improper vocabulary collocations, logical contradictions and other detection rules. By inputting the first speech into the audit component, the audit component analyzes the first speech sentence by sentence according to the rule base and marks potential errors or problems.

[0147] If the first speech passes the review, the target speech is generated based on the first speech.

[0148] In this embodiment, after the audit component audits the first speech, if it is detected that there is no error in the first speech, it is determined that there is no error in the grammar, logic, expression, etc. of the first speech, and the first speech can be directly used as the above-mentioned target speech.

[0149] If the first speech fails to pass the review, error information of the first speech is obtained, and the first speech is modified based on the review component to obtain a corresponding second speech.

[0150] In this embodiment, after the audit component has audited the first speech, if error information is detected in the first speech, the audit component will automatically perform corresponding modification processing on the first speech to correct grammatical errors and spelling errors in the first speech, thereby obtaining the corresponding second speech.

[0151] The target speech is generated based on the second speech.

[0152] In this embodiment, the second speech is used as the target speech.

[0153] The present application determines the analysis result filling area, the relevant information filling area and the suggestion information filling area in the target speech template; then fills the analysis result into the analysis result filling area, fills the target information into the relevant information filling area, and fills the suggestion information into the suggestion information filling area to obtain the corresponding first speech; then calls a preset review component to review the first speech; if the first speech passes the review, the target speech is generated based on the first speech; and if the first speech fails the review, the error information of the first speech is obtained, and the first speech is modified based on the review component to obtain the corresponding second speech; and then the target speech is generated based on the second speech. The present application obtains a first speech by filling the analysis result into the analysis result filling area in the target speech template, filling the target information into the relevant information filling area in the target speech template, and filling the suggestion information into the suggestion information filling area in the target speech template, and then intelligently audits the first speech based on the use of the audit component. If the first speech passes the audit, the target speech is generated based on the first speech. If the first speech fails the audit, the error information of the first speech is obtained, and the first speech is automatically modified based on the audit component to obtain the corresponding second speech and used as the final target speech, effectively ensuring the grammatical correctness, logical clarity and expression accuracy of the obtained target speech, which is conducive to providing customers with more accurate and effective health management suggestions based on the subsequent use of the target speech, thereby improving the customer's user experience.

[0154] In some optional implementations of this embodiment, after step S207, the electronic device may further perform the following steps:

[0155] The client information of the client is obtained, and the educational status of the client is obtained based on the client information.

[0156] In this embodiment, the customer information may refer to the customer's name information, and the preset customer database may be queried according to the customer's name information to obtain the customer's education status. The customer database is a pre-built database storing customer data of each customer. The customer data at least includes customer name, gender, education status, income and other profile data. The education status may include elementary school, junior high school, high school, university, graduate student, doctoral student, etc.

[0157] Determine whether the educational situation meets the preset educational conditions.

[0158] In this embodiment, the above-mentioned educational condition is that the education level is lower than high school education. If the customer's education level is lower than high school education, it is determined that the customer's acceptance level is low, and the language expression of the dialogue can be adjusted. Keep the language concise and easy to understand, thereby improving the customer's understanding of the dialogue, and thus improving the customer's reading experience.

[0159] If so, the target speech is subjected to sentence splitting processing to obtain a corresponding third speech.

[0160] In this embodiment, the sentence splitting process may include splitting complex sentences in the target speech into short sentences, ensuring that each sentence conveys a clear information point. Simple and easy-to-understand vocabulary is used, and industry terms or abbreviations are avoided unless the customer has clearly indicated that they are familiar with these terms. By performing sentence splitting on the target speech, it is possible to ensure that the information in the target speech is expressed directly and without detours, so that the customer can quickly grasp the key points.

[0161] The third speech is subjected to language polishing to obtain a corresponding fourth speech.

[0162] In this embodiment, the natural language generation capability of the GPT model can be used to perform language polishing and optimization on the dialogue to complete the language polishing processing of the third dialogue, thereby ensuring that the resulting dialogue is fluent and easy to understand while maintaining a humanistic communication style.

[0163] The emotional element is added to the fourth speech to obtain the corresponding fifth speech.

[0164] In this embodiment, positive and encouraging emotional elements can be appropriately incorporated into the fourth speech to enhance the customer's trust and willingness to execute, thereby obtaining the corresponding fifth speech.

[0165] The fifth script is sent to the customer.

[0166] In this embodiment, the specific implementation process of sending the fifth speech to the customer will be further described in detail in subsequent specific embodiments of the present application and will not be elaborated here.

[0167] This application obtains the customer information of the customer, and obtains the customer's education situation based on the customer information; then determines whether the education situation meets the preset education conditions; if so, performs sentence splitting processing on the target speech to obtain the corresponding third speech; then performs language polishing processing on the third speech to obtain the corresponding fourth speech; subsequently, performs emotional element addition processing on the fourth speech to obtain the corresponding fifth speech; and finally sends the fifth speech to the customer. After filling the target speech template based on the analysis results, the target information and the suggestion information to obtain the corresponding target speech, this application will further obtain the customer information of the customer, and obtain the customer's education situation based on the customer information. If it is detected that the education situation meets the preset education conditions, the target speech will be intelligently split, language polished and emotionally added to obtain the corresponding fifth speech, effectively ensuring that the generated fifth speech is accurately, clearly and personalized, so that the fifth speech is subsequently sent to the customer, which can provide the customer with more considerate and effective health management suggestions, thereby improving the customer's user experience.

[0168] In some optional implementations of this embodiment, sending the fifth speech script to the client includes the following steps:

[0169] A corresponding sending time is determined based on the customer information.

[0170] In this embodiment, the working time and sleeping time of the customer can be obtained according to the above customer information, and then the working time and sleeping time can be filtered from the time of one day to obtain the corresponding sending time.

[0171] Identify target communication channels.

[0172] In this embodiment, the above-mentioned target communication channels may include enterprise WeChat, SMS platform, email platform, etc.

[0173] Based on the sending time and the target communication channel, the fifth script is sent to the customer.

[0174] In this embodiment, when it is detected that the current time reaches the sending time, the fifth speech script is sent to the customer using the above communication channel.

[0175] This application determines the corresponding sending time based on the customer information; then determines the target communication channel; and subsequently sends the fifth speech to the customer based on the sending time and the target communication channel. After optimizing the target speech based on the customer's education situation to obtain the fifth speech, this application will also intelligently determine the corresponding sending time based on the customer's customer information, and then send the fifth speech to the customer based on the occurrence time and the target communication channel, effectively improving the push intelligence of the fifth speech and helping to ensure the customer's user experience.

[0176] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0177] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned target words, the above-mentioned target words can also be stored in a node of a blockchain.

[0178] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0179] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0180] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0181] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0182] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0183] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a speech generation device based on artificial intelligence, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0184] like Figure 3 As shown, the speech generation device 300 based on artificial intelligence described in this embodiment includes: a recognition module 301, a first acquisition module 302, a first processing module 303, an extraction module 304, a first generation module 305, a second acquisition module 306 and a second generation module 307. Among them:

[0185] The recognition module 301 is used to obtain the speech content sent by the customer, and perform information recognition on the speech content based on the preset recognition rules to obtain the corresponding key information;

[0186] The first acquisition module 302 is used to obtain the chronic disease type information of the customer;

[0187] The first processing module 303 is used to process the key information and the chronic disease type information based on a preset data analysis model to obtain corresponding analysis results;

[0188] An extraction module 304 is used to extract target information related to the chronic disease type information from the analysis results;

[0189] A first generating module 305, configured to generate corresponding suggestion information based on the target information;

[0190] The second acquisition module 306 is used to acquire a target speech template corresponding to the chronic disease type information;

[0191] The second generation module 307 is used to fill in the target speech template based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech.

[0192] In some optional implementations of this embodiment, the second acquisition module 306 includes:

[0193] The calling submodule is used to call the preset speech template library;

[0194] A first generating submodule, used to generate corresponding search conditions based on the chronic disease type information;

[0195] A search submodule, configured to search the speech template library based on the search condition to obtain a first speech template matching the search condition;

[0196] A judgment submodule, used for judging whether the number of the first speech template is multiple;

[0197] The screening submodule is used to screen the first speech based on a preset screening algorithm to obtain the target speech template corresponding to the chronic disease type information.

[0198] In some optional implementations of this embodiment, the screening submodule includes:

[0199] An acquisition unit, used for acquiring a customer feedback score of each of the first speech templates, and acquiring an execution rate of each of the first speech templates;

[0200] A calculation unit, used to calculate the matching score of each of the first speech templates;

[0201] A generating unit, configured to generate a template score for each of the first speech templates based on the customer feedback score, the execution rate, and the matching score;

[0202] A screening unit, used to screen out a designated template score with the highest score from all the template scores;

[0203] A determination unit is used to obtain a second speech template corresponding to a specified template score, and use the second speech template as the target speech template.

[0204] In some optional implementations of this embodiment, the generating unit includes:

[0205] A first calling subunit, used for calling a preset weight setting algorithm;

[0206] A generating subunit, configured to generate, based on the weight setting algorithm, a first weight corresponding to a customer feedback factor, a second weight corresponding to an execution rate factor, and a third weight corresponding to a matching factor;

[0207] The second calling subunit is used to call a preset score calculation formula;

[0208] A calculation subunit is used to calculate the customer feedback score, execution rate and matching score of each of the first speech templates based on the first weight, the second weight and the third weight using the score calculation formula to obtain the template score of each of the first speech templates.

[0209] In some optional implementations of this embodiment, the second generating module 307 includes:

[0210] The first determination submodule is used to determine the analysis result filling area, the related information filling area and the suggestion information filling area in the target speech template;

[0211] A filling submodule, used to fill the analysis result into the analysis result filling area, fill the target information into the related information filling area, and fill the suggestion information into the suggestion information filling area to obtain a corresponding first speech;

[0212] An audit submodule, used for calling a preset audit component to perform audit processing on the first speech;

[0213] A second generating submodule, configured to generate the target speech based on the first speech if the first speech passes the review;

[0214] A modification submodule, configured to obtain error information of the first speech if the first speech fails to pass the review, and modify the first speech based on the review component to obtain a corresponding second speech;

[0215] The third generating submodule is used to generate the target speech based on the second speech.

[0216] In some optional implementations of this embodiment, the speech generation device based on artificial intelligence further includes:

[0217] A third acquisition module is used to acquire the customer information of the customer, and acquire the education status of the customer based on the customer information;

[0218] A judgment module, used to judge whether the education situation meets the preset education conditions;

[0219] A second processing module is used for, if yes, performing sentence splitting processing on the target speech to obtain a corresponding third speech;

[0220] A third processing module is used to perform language polishing on the third speech to obtain a corresponding fourth speech;

[0221] A fourth processing module, used for adding emotional elements to the fourth speech to obtain a corresponding fifth speech;

[0222] A sending module is used to send the fifth speech script to the customer.

[0223] In some optional implementations of this embodiment, the sending module includes:

[0224] A second determination submodule, configured to determine a corresponding sending time based on the customer information;

[0225] The third determination submodule is used to determine the target communication channel;

[0226] The sending submodule is used to send the fifth speech script to the customer based on the sending time and the target communication channel.

[0227] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0228] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital processors (DSPs), embedded devices, etc.

[0229] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0230] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device 4. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions based on the artificial intelligence speech generation method. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0231] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run the computer-readable instructions stored in the memory 41 or process data, such as running the computer-readable instructions of the artificial intelligence-based speech generation method.

[0232] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0233] The present application also provides another embodiment, namely, providing a computer-readable storage medium, which stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the above-mentioned artificial intelligence-based speech generation method.

[0234] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0235] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A method for generating speech based on artificial intelligence, characterized in that: The steps include: Obtain the speech content sent by the customer, and perform information recognition on the speech content based on the preset recognition rules to obtain the corresponding key information; Obtaining information on the type of chronic diseases of the customer; Processing the key information and the chronic disease type information based on a preset data analysis model to obtain corresponding analysis results; Extracting target information related to the chronic disease type information from the analysis results; generating corresponding suggestion information based on the target information; Obtain a target speech template corresponding to the chronic disease type information; The target speech template is filled based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech.

2. The method for generating speech based on artificial intelligence according to claim 1, characterized in that: The step of obtaining a target speech template corresponding to the chronic disease type information specifically includes: Call the preset speech template library; Generate corresponding search conditions based on the chronic disease type information; Searching the speech template library based on the search condition to obtain a first speech template matching the search condition; Determine whether there are multiple first speech templates; If so, the first speech is screened based on a preset screening algorithm to obtain the target speech template corresponding to the chronic disease type information.

3. The method for generating speech based on artificial intelligence according to claim 2, characterized in that: The step of screening the first speech based on a preset screening algorithm to obtain the target speech template corresponding to the chronic disease type information specifically includes: Obtaining customer feedback scores for each of the first speech templates, and obtaining execution rates of each of the first speech templates; Calculating the matching score of each of the first speech templates; Generate a template score for each of the first speech templates based on the customer feedback score, the execution rate, and the matching score; Filter out the designated template score with the highest score from all the template scores; Obtain a second speech template corresponding to the specified template score, and use the second speech template as the target speech template.

4. The method for generating speech based on artificial intelligence according to claim 3 is characterized in that: The step of generating a template score for each of the first speech templates based on the customer feedback score, the execution rate, and the matching score specifically includes: Call the preset weight setting algorithm; Generate a first weight corresponding to the customer feedback factor, a second weight corresponding to the execution rate factor, and a third weight corresponding to the matching factor based on the weight setting algorithm; Call the preset score calculation formula; Based on the first weight, the second weight and the third weight, the customer feedback score, execution rate and matching score of each of the first speech templates are calculated using the score calculation formula to obtain the template score of each of the first speech templates.

5. The method for generating speech based on artificial intelligence according to claim 1, characterized in that: The step of filling the target speech template based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech specifically includes: Determine an analysis result filling area, a related information filling area, and a suggestion information filling area in the target speech template; Fill the analysis result into the analysis result filling area, fill the target information into the related information filling area, and fill the suggestion information into the suggestion information filling area to obtain a corresponding first speech; Calling a preset audit component to audit the first speech; If the first speech is approved, generating the target speech based on the first speech; If the first speech fails to pass the review, error information of the first speech is obtained, and the first speech is modified based on the review component to obtain a corresponding second speech; The target speech is generated based on the second speech.

6. The method for generating speech based on artificial intelligence according to claim 1, characterized in that: After the step of filling the target speech template based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech, the method further includes: Acquire customer information of the customer, and acquire the education status of the customer based on the customer information; Determine whether the educational situation meets the preset educational conditions; If yes, the target speech is split into sentences to obtain a corresponding third speech; Performing language polishing on the third speech to obtain a corresponding fourth speech; Adding emotional elements to the fourth speech to obtain a corresponding fifth speech; The fifth script is sent to the customer.

7. The method for generating speech based on artificial intelligence according to claim 6, characterized in that: The step of sending the fifth speech to the customer specifically includes: Determine a corresponding sending time based on the customer information; Identify target communication channels; Based on the sending time and the target communication channel, the fifth script is sent to the customer.

8. A speech generation device based on artificial intelligence, characterized in that: include: The recognition module is used to obtain the speech content sent by the customer and perform information recognition on the speech content based on the preset recognition rules to obtain the corresponding key information; The first acquisition module is used to obtain the chronic disease type information of the customer; A first processing module is used to process the key information and the chronic disease type information based on a preset data analysis model to obtain corresponding analysis results; An extraction module, used to extract target information related to the chronic disease type information from the analysis results; A first generating module, configured to generate corresponding suggestion information based on the target information; A second acquisition module is used to acquire a target speech template corresponding to the chronic disease type information; The second generation module is used to fill in the target speech template based on the analysis result, the target information and the suggestion information to obtain the corresponding target speech.

9. A computer device, characterized in that: It includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, it implements the steps of the method for generating speech based on artificial intelligence as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the method for generating speech based on artificial intelligence as described in any one of claims 1 to 7.

Images