User interaction method and device, electronic equipment and computer readable storage medium

Abstract

The invention provides a user interaction method and device, electronic equipment and a computer readable storage medium. The method comprises the steps of obtaining user behavior data and user portrait data of a target user; according to the user behavior data and the user portrait data, obtaining an emotion category of the target user for the target object; according to the emotion category and the user portrait data, generating a target verbal-skill term, the target verbal-skill term corresponding to the emotion category and being used for guiding the emotion category of the user to be converted into a forward emotion category; and interacting with the target user based on the target verbal skill term. According to the embodiment of the invention, interaction with the target user is carried out, the effect of improving the user experience can be realized, the interaction content is ensured to be matched with the user demand, and the accuracy of interaction response is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a user interaction method and device, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the continuous advancement of computer technology, intelligent voice interaction systems have been widely used in many fields, such as intelligent voice customer service, intelligent marketing, smart speakers, etc. In the current intelligent voice interaction system, generally an intelligent voice robot interacts with the user by voice. Among them, the intelligent voice robot is composed of multiple technologies such as natural language processing technology (NLP, Natural Language Processing), speech recognition technology and speech synthesis technology. Through language recognition technology, the intelligent voice robot can convert the user's voice into text, and through natural speech processing technology, the intelligent voice robot can understand and parse the text. Finally, through speech synthesis technology, the intelligent voice robot can convert text into natural speech output to achieve voice interaction with the user.

[0003] In the process of interaction between intelligent voice robots and users, how to improve the communication effect becomes a difficult point. When existing intelligent voice robots interact with users, they often select matching words from preset words and broadcast them according to the questions raised by users. This interaction method usually has the problems of low efficiency and poor user experience. Summary of the invention

[0004] The present disclosure provides a user interaction method and device, an electronic device, and a computer-readable storage medium.

[0005] In a first aspect, the present disclosure provides a user interaction method, the method comprising:

[0006] Obtain user behavior data and user portrait data of target users;

[0007] Obtaining the emotion category of the target user toward the target object according to the user behavior data and the user portrait data;

[0008] Generate target speech terms according to the emotion category and the user portrait data, wherein the target speech terms correspond to the emotion category and are used to guide the user's emotion category to convert to a positive emotion category;

[0009] Interact with the target user based on the target speech terms.

[0010] In a second aspect, the present disclosure provides a user interaction device, the user interaction device comprising:

[0011] An acquisition unit, used to acquire user behavior data and user portrait data of target users;

[0012] An emotion category recognition unit, used to obtain the emotion category of the target user towards the target object according to the user behavior data and the user portrait data;

[0013] A speech generating unit, used to generate target speech terms according to the emotion category and the user portrait data, wherein the target speech terms correspond to the emotion category and are used to guide the user's emotion category to convert to a positive emotion category;

[0014] An interaction unit is used to interact with the target user based on the target speech terms.

[0015] In a third aspect, the present disclosure provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores one or more computer programs executable by the at least one processor, and one or more of the computer programs are executed by the at least one processor so that the at least one processor can execute the user interaction method of the first aspect mentioned above.

[0016] In a fourth aspect, the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the user interaction method of the first aspect when executed by a processor.

[0017] According to the embodiments provided by the present disclosure, when an electronic device integrated with an intelligent voice robot interacts with a user, with the user's authorization, the target user's behavior data and user portrait data are obtained, and the target user's emotion category for the target object is obtained based on the user behavior data and the user portrait data; thereafter, the target speech terms are generated based on the emotion category and the user portrait data, and the target user is interacted with based on the target speech terms. Since the target speech terms correspond to the target user's emotion category for the target object and can guide the user's emotion to convert to a positive emotion category, the intelligent voice robot can sensitively respond to the user's emotions, provide more considerate services, and enhance the emotional connection with the user by interacting with the user based on the embodiments of the present disclosure, thereby achieving the effect of improving the user experience, ensuring that the interactive content matches the user's needs, and improving the accuracy of the interactive response.

[0018] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. The above and other features and advantages will become more apparent to those skilled in the art by describing detailed example embodiments with reference to the accompanying drawings, in which:

[0020] Figure 1 A flowchart of a user interaction method provided by an embodiment of the present disclosure;

[0021] Figure 2 A flowchart for determining emotion categories provided by an embodiment of the present disclosure;

[0022] Figure 3 A flow chart for generating target speech terms provided in an embodiment of the present disclosure;

[0023] Figure 4 A schematic diagram of a user interaction process provided by an embodiment of the present disclosure;

[0024] Figure 5 A block diagram of a user interaction device provided in an embodiment of the present disclosure;

[0025] Figure 6 A block diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the technical solutions of the present disclosure, the exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0027] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.

[0028] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0029] The terms used herein are only used to describe specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a" and "the" are also intended to include plural forms, unless the context clearly indicates otherwise. It will also be understood that when the terms "including" and / or "made of" are used in this specification, the presence of the features, wholes, steps, operations, elements and / or components is specified, but the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof is not excluded. "Connected" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0030] It should be noted that the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in the technical solution of this disclosure are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals. The use of user data in this technical solution complies with relevant national laws and regulations (for example, the "Information Security Technology Personal Information Security Specification", etc.). For example: corresponding prescribed measures are taken for personal information access control; the display of personal information is subject to prescribed restrictions; the purpose of using personal information does not exceed the scope of direct or reasonable association; when using personal information, clear identity orientation is eliminated to avoid precise positioning of specific individuals.

[0031] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless explicitly defined as such herein.

[0032] In the relevant technology, answer texts or answer voices corresponding to different questions are generally preset. When the intelligent voice robot interacts with the user, it generally receives the user's question first, and then searches for matching texts or matching voices from the preset answer texts or answer voices based on the user's question. If the matching text or matching voice cannot be found, the intelligent voice robot often cannot accurately respond to the user's question, thereby affecting the interaction effect and bringing a bad experience to the user.

[0033] In order to match the interactive content with user needs and improve the user experience, you can also consider grouping users and then setting corresponding scripts for different user groups. For example, you can divide them into user group 1, user group 2, ..., user group n, where n is a positive integer, and preset robot scripts corresponding to the n user groups. During the interaction between the intelligent voice robot and the user, the corresponding user group is selected according to the user group to which the user belongs to perform the interaction, thereby improving the matching of the interactive content with user needs and further improving the user experience.

[0034] However, although this method can improve the matching between interactive content and user needs at a coarse-grained level, it often cannot achieve more detailed personalized interactions, for example, it cannot achieve a personalized interaction method for each user. That is, in this user interaction method, the optimization of the speech is relatively simple, and the speech corresponding to different user groups needs to rely on offline artificial intelligence trainers to summarize and refine, and to find the best solution through systematic serial comparison experiments. This process often takes a long time, for example, it may take up to 10 days or longer.

[0035] In view of this, in the embodiments of the present disclosure, in order to improve the accuracy of the interactive response and improve the user experience, it is considered that when interacting with the user, the user behavior data, such as user attribute data or historical voice data, usually reflects the user's emotional tendency towards the target object, for example, whether it is an emotional tendency of the positive category or a negative category, and, in the process of real-time interaction with the user, the user's emotional tendency towards the target object will usually change in real time. For this reason, in the embodiments of the present disclosure, with the authorization of the user, the emotional category of the target user is determined through the user behavior data and the user portrait data, and the target speech terms are generated according to the emotional category and the user portrait data. Since the target speech terms correspond to the emotional category and can guide the user's emotional category to convert to a positive emotional category, therefore, based on the target speech terms, the interaction with the target user can be carried out in real time during the interaction with the target user. The emotional changes of the user towards the target object involved in the interaction process can be perceived, and more considerate services can be provided to the target user, and the emotional connection with the target user can be enhanced, thereby achieving the effect of accurately providing interactive content to the user according to the user's needs and improving the user experience.

[0036] Please see Figure 1 , which is a flow chart of a user interaction method provided by an embodiment of the present disclosure. Figure 1As shown, the method can be applied to electronic devices, more specifically, it can be applied to electronic devices integrated with intelligent voice robots, the intelligent voice robot can be composed of at least one of the machine learning models such as emotion category recognition model, resource recommendation model, speech prediction model and strategy prediction model, the intelligent voice robot can be used for voice interaction processing with the target user, for example, it can be used to obtain the voice content of the target user, and interact with the target user based on the target speech terms corresponding to the target user, so as to ensure that the interactive content matches the user's needs, and improve the user experience by improving the accuracy of the interactive response. In actual implementation, the electronic device can be a server, or it can also be a terminal device, which is not specifically limited here.

[0037] like Figure 1 As shown, the user interaction method provided by the embodiment of the present disclosure includes the following steps S101-S104, which are described in detail below.

[0038] Step S101, obtaining user behavior data and user portrait data of the target user.

[0039] User behavior data may include user attribute data, for example, user gender, age, interests and hobbies, etc. User behavior data may also include the user's behavior towards the target object, for example, whether the user pays attention to the target object or has paid attention to or used the competing product of the target object.

[0040] Among them, the target object can be a physical object or a virtual object. The target object can be, for example, a physical commodity or a virtual product. For example, in the financial field, the target object can be a virtual financial product. In actual implementation, the target object can be set as needed and is not specifically limited here.

[0041] It should be noted that, in actual implementation, before implementing step S101, the method also includes: sending an authorization request to the target user, where the authorization request is used to request the use of the user behavior data and user portrait data of the target user; upon receiving the authorization result corresponding to the authorization request and the authorization result indicates that the target user agrees to use the user behavior data and user portrait data, executing the step of obtaining the user behavior data and user portrait data of the target user.

[0042] That is, in the embodiments of the present disclosure, the acquisition and use of the user behavior data and user portrait data require user authorization. For example, during interaction with the user, such as voice interaction, before preparing to broadcast the formal voice content, a voice prompt content may be played first to solicit the user's authorization to use their behavior data and to authorize their portrait and use their portrait data. For example, the user may be prompted to click a corresponding button to determine whether to authorize, and after obtaining the user's authorization, the user behavior data is acquired and the user portrait data of the user is generated for use. Of course, this is only an illustrative explanation. In actual implementation, other forms may also be used to solicit the user's authorization to use the user behavior data and the user portrait data, which is not specifically limited here.

[0043] User portrait (Perona) data can be obtained by profiling the user based on historical interaction data with the target user with the user's authorization.

[0044] In the embodiments of the present disclosure, in order to improve the accuracy of user portrait data, during the process of profiling the user, fine-grained portrait processing can be performed in combination with the current real-time business scenario.

[0045] For example, in a product marketing scenario, the user portrait data can be used to profile the user through the user's value data at different promotion stages. In this implementation, with the user's authorization, the user's transaction data for the product at different promotion stages can be obtained, and the user can be profiled based on the RFM model and the transaction data, wherein the RFM model is a customer relationship management model that describes the user's value status through three indicators: the user's most recent consumption (R, Recency), consumption frequency (F, Frequency) and consumption amount (M, Monetary). For example, the product promotion stage can generally be divided into a pre-sale stage and a formal sales stage. The user transaction data in the pre-sale stage usually indicates the user's intention for the product, and the user transaction data in the formal sales stage can further clarify the user's intention for the product. Therefore, by profiling the user based on the user consumption data in different promotion stages, the accuracy and pertinence of the obtained user portrait data can be improved.

[0046] In the disclosed embodiments, the user portrait data may include, for example, user categories such as "promoting renewal of expired membership", "promoting first-order users", "promoting membership conversion", "providing care for high-value users", "consumed in the same theme promotion activities in the past two years but have not logged in in the past year", etc. Of course, here the method is applied to the product marketing scenario, and the target object is the product as an example for explanation. In actual implementation, the user portrait data can also be set accordingly according to different application scenarios, and no special limitation is made here.

[0047] Step S102, obtaining the emotion category of the target user towards the target object based on the user behavior data and the user portrait data.

[0048] In the embodiments of the present disclosure, the emotion category is used to characterize the target user's intention towards the target object. The emotion category may include a positive emotion category and a negative emotion category. The positive emotion category refers to a positive type of emotion, and the negative emotion category refers to a negative type of emotion. For example, if the target user continues to pay attention to an object, or continues to use an object, it often indicates that the target user has a positive interaction intention with respect to the object. If the target user has never paid attention to an object, or has reduced his or her attention to an object compared to historical data, it usually indicates that the target user has a negative interaction intention with respect to the object.

[0049] By determining the target user's emotional category based on user behavior data and user portrait data, the intelligent voice robot can be more sensitive to changes in user emotions during voice interaction with the user, actively respond to the user's negative emotions, or provide more attractive content when the user shows interest, thereby enhancing the emotional resonance between the robot and the user, and guiding the user's emotions to convert to positive emotional categories that are beneficial to the target object, thereby ensuring that the interactive content matches user needs and improving the user experience by improving the accuracy of the interactive response.

[0050] Step S103, generating target speech terms according to the emotion category and user portrait data, wherein the target speech terms correspond to the emotion category and are used to guide the user's emotion category to convert to a positive emotion category; and executing step S104, interacting with the target user based on the target speech terms.

[0051] The target sales talk can be in the form of, for example, "We have prepared a blind box of products full of unknowns and surprises for you, valued at 18,000 yuan and priced at only 7,000 yuan. Our exclusive customer service has pushed the link to you, and you can place an order if you like it." That is, based on the user's interest points, resource data corresponding to the target users can be generated, and the resource data can be formed into sales talk that can arouse the user's interest, and interaction with users can be carried out based on the sales talk to improve the conversion rate.

[0052] It can be seen that based on the user interaction method provided by the embodiment of the present disclosure, when an electronic device integrated with an intelligent voice robot interacts with a user, it obtains the user behavior data and user portrait data of the target user, and obtains the emotion category of the target user for the target object based on the user behavior data and the user portrait data; thereafter, based on the emotion category and the user portrait data, it generates target speech terms, and interacts with the target user based on the target speech terms. Since the target speech terms correspond to the emotion category of the target user for the target object, and can guide the user's emotions to convert to positive emotion categories, therefore, by interacting with the user based on the embodiment of the present disclosure, the intelligent voice robot can respond sensitively to the user's emotions, provide more considerate services, and enhance the emotional connection with the user, thereby achieving the effect of improving the user experience, ensuring that the interaction content matches the user's needs, and improving the accuracy of the interaction response.

[0053] Please see Figure 2 , which is a flow chart of determining emotion categories provided by an embodiment of the present disclosure. Figure 2 As shown, in some embodiments, obtaining the emotion category of the target user with respect to the target object according to the user behavior data and the user portrait data may include the following steps S201-S203:

[0054] Step S201, determining the target user category of the target user according to the user portrait data;

[0055] In actual implementation, the user ID of the target user may be obtained, and the category of the target user may be obtained by querying in the user portrait data based on the user ID.

[0056] For example, in a product marketing scenario, user portrait data can be used to divide users into user categories such as "promoting renewal of expired membership", "promoting first-order users", "promoting membership conversion", "providing care for high-value users", and "consumers who have consumed the same theme promotion activities in the past two years but have not logged in in the past year". When determining the target user category of the target user, the UserID can be obtained, and based on the UserID, it can be queried whether the user data belonging to different user categories contains the UserID. If the UserID is included in the user data under a certain user category, the user category can be determined as the target user category.

[0057] Step S202, determining an emotion category recognition model corresponding to the target user according to the target user category;

[0058] Step S203: input the user behavior data into an emotion category recognition model to perform emotion category recognition processing to obtain an emotion category.

[0059] Specifically, in the embodiments of the present disclosure, in order to be able to specifically identify the emotion category of the target user towards the target object, in the process of emotion category recognition processing, a unified emotion category recognition model is not used to perform emotion category recognition processing on users belonging to different user categories. Considering that the emotional intentions expressed by users of different categories towards the same object are often different, the intelligent voice robot in the embodiments of the present disclosure may include emotion category recognition models corresponding to different user categories. When identifying the emotion category of the target user, the emotion category recognition model to be used can be determined based on the target user category, and emotion category recognition processing can be performed on the target user based on the determined emotion category recognition model to improve the accuracy of the obtained emotion category, and marketing is performed based on the emotion category, so that the intelligent voice robot can respond sensitively to user emotions, provide more considerate services, and enhance the emotional connection with users, thereby ensuring that the interactive content matches user needs, and improving the user experience by improving the accuracy of the interactive response.

[0060] Please see Figure 3 , which is a flow chart of generating target speech terms provided by an embodiment of the present disclosure. Figure 3 As shown, in some embodiments, generating target speech terms according to emotion categories and user portrait data may include the following steps S301-S303:

[0061] Step S301, determining a resource recommendation model corresponding to the target user according to the target user category, wherein the resource recommendation model is used to generate resource data corresponding to the user intention characteristics according to the user intention characteristics represented by the user behavior data;

[0062] Step S302, inputting user behavior data into a resource recommendation model for resource prediction processing to obtain target resource data;

[0063] Step S303, generating target speech terms according to the emotion category and the target resource data.

[0064] The resource recommendation model is used to obtain the user intention characteristics of the user for the marketing object based on the input user behavior data, and then predict the resource data corresponding to the user intention characteristics.

[0065] This resource data, namely "interest points", is a resource that can satisfy user intentions and promote user conversion rates. For example, in a product marketing scenario, for a certain beauty product, the resource data can be at least one of the following resources: coupons, free trial products or free product samples, and payment discounts.

[0066] Specifically, in the embodiments of the present disclosure, in order to improve user conversion rates while saving costs, resource data for different categories of users is often different. For example, in a product marketing scenario, for a certain beauty product, in addition to giving samples or trial packs and other physical items to all categories of users, more attractive discounts, affordable discount coupons, no-threshold coupons and other resources can be generated for users in the category of "expired membership renewal promotion"; and for users in the category of "first-order users", since this type of users often have a lower degree of attention to the target object, conversion rates can be promoted by providing membership gifts, blind box lotteries, and free points and other resources, so as to maximize user conversion rates while maintaining corporate costs.

[0067] In the disclosed embodiments, user behavior data is analyzed based on a resource recommendation model to extract user intention features and then generate corresponding resource data, so that the intelligent voice robot can accurately obtain resource data that meets the user's current needs, generate target speech terms based on the resource data, and interact with the target user based on the target speech terms. This can improve the target user's satisfaction and willingness to interact during the interaction process, so that the interaction content can better match user needs and improve the accuracy of the interaction response.

[0068] In some embodiments, in the above step S303, generating the target speech terms according to the emotion category and the target resource data may include: inputting the emotion category and the target resource data into the speech prediction model to perform speech prediction processing to obtain the target speech terms.

[0069] In the disclosed embodiment, the speech prediction model may be based on the target user's emotion category for the target object and the target resource data corresponding to the target user to perform speech prediction to generate a network model of speech terms that can meet the user's intentions.

[0070] It can be seen that in the embodiment of the present disclosure, by obtaining the emotion category corresponding to the target user, the target resource data that meets the target user's intention is obtained according to the emotion category, and then the emotion category and the target resource data are input into the speech prediction model for speech prediction, so that the generated target speech terms can adapt to the emotions of the target user and guide the user's emotion category to convert to a positive emotion category, thereby improving the sensitivity of the intelligent voice robot, so that in the process of interacting with the user, it can use the speech that meets the user's intention to interact with the user in a targeted manner, so as to improve the accuracy of the interactive response.

[0071] In some embodiments, in the above step S104, the interaction with the target user based on the target speech terms may include: performing voice outbound call processing on the target user; in the process of outbound voice interaction with the target user, inputting the acquired user real-time voice data into the strategy prediction model for interaction strategy prediction processing to obtain the target interaction strategy; based on the target interaction strategy and the target speech terms, generating corrected speech terms; and performing voice interaction with the target user based on the corrected speech terms.

[0072] Corrected speech terms are speech terms that are adapted to the real-time intentions of target users after correcting the target speech terms based on the target interaction strategy.

[0073] The correction processing of the target speech terms may be to change the target resource data according to the interaction strategy predicted by the strategy prediction model, adjust the voice interaction details, and so on.

[0074] For example, in a product marketing scenario, during voice interaction with the target user, if the user hears “That sounds good. Wait a minute. Let me think about it.”, it means that the target user is inclined to buy the product. In this case, the user's real-time voice data can be input into the strategy prediction model for interaction strategy prediction processing to generate an interaction strategy that can encourage the target user to quickly place an order. The target speech terms currently in use can be adjusted in a timely manner according to the interaction strategy. For example, a phrase like “Order now and enjoy xxx discount” can be added to generate corrected speech terms. The corrected speech terms can be used to interact with the target user based on the corrected speech terms to ensure that the interaction content matches user needs and improve the accuracy of the interaction response.

[0075] That is, considering that the target speech terms are the preliminary work done based on the user behavior data and user portrait data of the target user before the intelligent voice robot conducts voice interaction with the target user, that is, preprocessing work; however, in the actual call process, the intention of the target user may change with the interaction process. Therefore, if the real-time interaction with the target user is always based on the target speech terms, the real-time needs of the target user may not be met, resulting in a poor experience in the interaction process. In the embodiment of the present disclosure, by performing interaction strategy prediction processing on the user's real-time voice data based on the strategy prediction model, generating a real-time interaction strategy, and correcting the currently used speech based on the interaction strategy, the interaction process can be refined to improve the user experience.

[0076] In some embodiments, in the above step S104, the interaction with the target user based on the target speech terms may include: performing an outbound voice call processing on the target user; in the process of outbound voice interaction with the target user, determining the target timbre according to the gender of the target user; generating real-time interactive voice data according to the target timbre and target speech terms; and performing voice interaction with the target user based on the real-time interactive voice data.

[0077] The target voice can be a personalized voice package. For example, when the target user is a male user, the corresponding female voice with a gentle and sweet voice can be selected, and when the target user is a female user, the corresponding male voice with a natural and magnetic voice can be selected.

[0078] That is, in the process of voice interaction with the target user, in order to improve the user experience, the target tone corresponding to the gender of the target user can be selected to interact with the user, so as to make the interaction process more natural and avoid giving the user an overly mechanical interaction experience, thereby improving user retention rate.

[0079] In some embodiments, in the above step S104, the interaction with the target user based on the target speech terms may include: obtaining auxiliary channel information of the auxiliary interaction channel for auxiliary interaction with the target user, wherein the auxiliary channel information includes at least one of the text messages, emails and social platforms corresponding to the target user; generating interaction push information based on the target speech terms; based on the channel information, pushing the interaction push information to the auxiliary interaction channel; after pushing the interaction push information to the auxiliary interaction channel, performing voice interaction with the target user based on the target speech terms.

[0080] In the related art, when interacting with users, for example, in a product marketing scenario, when interacting with users to introduce a product, a single channel is often used to interact with users. For example, voice interaction is often used, while other interaction channels are ignored. Since users cannot see real-time product information during voice interaction, it is easy to bring users a poor interaction experience.

[0081] To this end, in the embodiments of the present disclosure, after generating the target speech terms, interactive push information can be generated based on the target speech terms, for example, marketing recommendation information such as "xxx product is on sale, exquisite gifts are given away for a limited time, click the link address: http: / / xxxxx" is generated, and the interactive push is pushed to the auxiliary interactive channel, that is, the interactive push information consistent with the target speech terms is pushed to the user in the form of SMS, email and social platform push. Afterwards, in the process of voice interaction with the target user, after broadcasting the corresponding information, additional content such as "the corresponding content has been pushed to your account through channels such as SMS, email and social platforms. You can click on the relevant link address to view detailed information after the call ends" can be broadcast, so as to improve the user experience through a multi-channel combination.

[0082] It can be seen that based on the embodiments of the present disclosure, interacting with users through a multi-channel joint approach can allow users to feel the company's attention and care through multiple channels, thereby improving user experience.

[0083] In some embodiments, after interacting with the target user based on the target speech terms and the interaction is completed, the method also includes: obtaining current interaction data, wherein the current interaction data includes user voice data in the process of interacting with the target user based on the target speech terms and response voice data corresponding to the user voice data; generating training data based on the current interaction data; and optimizing and training at least one of the emotion category recognition model, the resource recommendation model, and the speech prediction model based on the training data.

[0084] The emotion category recognition model, resource recommendation model, and speech prediction model are machine learning models that can be built into the intelligent voice robot.

[0085] Specifically, after the interaction with the user is completed, the user's voice data and response voice data during the interaction process can be obtained with the user's authorization, and training data can be constructed based on the interaction data containing the two to continuously optimize the models used by the intelligent voice robot, so that these models can continuously learn and optimize from user feedback, thereby improving the intelligence level of the intelligent voice robot, so that the intelligent voice robot can meet the ever-changing user needs through continuous autonomous learning without excessive human intervention, ensure the match between the provided interactive content and user needs, and improve the accuracy of the interactive response to achieve the effect of improving the user experience.

[0086] In addition, it should be noted that in the actual implementation process, after the interaction with the target user is completed based on the target speech terms, the method further includes: obtaining an interaction result, which is used to indicate the target user's interaction satisfaction; if the interaction result indicates that the target user's satisfaction is low, a new target speech term can be generated after a preset time interval, and the target user can be interacted with based on the new target speech term. The preset time interval can be, for example, 3 days, or can be set as needed.

[0087] For example, in a product marketing scenario, if the user does not purchase the product after the current interaction with the user is completed based on the target speech terms, it means that the user is less satisfied with the interaction. In this case, after a preset time interval, the intelligent voice robot can generate new target speech terms based on the above steps S101-S103 again, and interact with the target user again based on the new target speech terms to improve the user experience.

[0088] It should be noted that, in actual implementation, one or more of the above embodiments may be combined for use, and no special limitation is made here.

[0089] For ease of understanding, the following example uses this method in a product marketing scenario, with the target object being the marketing object. Figure 4 The user interaction method provided by the embodiment of the present disclosure is schematically described, see Figure 4 , the method may include the following steps S401-S409:

[0090] Step S401, obtaining user behavior data and user portrait data of the target user;

[0091] It is understandable that step S401 needs to be performed with the user's authorization, and the specific form of obtaining the user's authorization is not particularly limited here.

[0092] Step S402, obtaining the target user's emotion category for the marketing object based on the user behavior data and the user portrait data;

[0093] Step S403, generating target speech terms according to the emotion category and user portrait data;

[0094] Step S404, obtaining auxiliary channel information of an auxiliary interaction channel for auxiliary interaction with a target user, generating interaction push information according to target speech terms, and pushing the interaction push information to the auxiliary interaction channel based on the channel information;

[0095] Step S405, performing an outbound call processing on the target user, in the process of outbound voice interaction with the target user, using a target timbre and target speech terms that match the gender of the target user, generating real-time voice data, and performing voice interaction with the target user based on the real-time interactive voice data;

[0096] Step S406, during the process of outbound voice interaction with the target user, the acquired real-time voice data of the user is input into the strategy prediction model for marketing strategy prediction processing to obtain a target marketing strategy, a correction speech term is generated based on the target marketing strategy and the target speech term, and voice interaction is performed with the target user based on the correction speech term;

[0097] After the interaction is completed, step S407 may be performed to obtain current interaction data, wherein the current interaction data may include marketing results;

[0098] Step S408, generating training data based on the current interaction data, and using the training data to optimize the training of multiple models; and, step S409 can also be executed to obtain marketing results. When the marketing results indicate that the target user has not purchased the marketing object, a new target speech terminology is generated after a preset time interval, and the target user is interacted with again based on the new target speech terminology.

[0099] It can be seen that based on the method provided in the embodiment of the present disclosure, on the premise of obtaining the user's authorization to use their behavior data and portrait data, user interaction is performed by combining one or more of user portrait processing, emotion recognition processing, personalized resource data recommendation processing, multi-channel assisted interaction processing, real-time interaction strategy correction processing, personalized timbre interaction processing, and model autonomous continuous optimization learning processing. In the process of interaction with users, intelligent interaction with thousands of people can be realized in a refined manner, the accuracy of interactive responses can be improved, and the effect of improving user experience can be achieved.

[0100] It can be understood that the above-mentioned various method embodiments mentioned in the present disclosure can be combined with each other to form a combined embodiment without violating the principle logic. Due to space limitations, the present disclosure will not repeat them. It can be understood by those skilled in the art that in the above-mentioned method of the specific implementation method, the specific execution order of each step should be determined according to its function and possible internal logic.

[0101] In addition, the present disclosure also provides a user interaction device, an electronic device, and a computer-readable storage medium, all of which can be used to implement any user interaction method provided by the present disclosure. The corresponding technical solutions and descriptions are referred to in the corresponding records of the method part and will not be repeated here.

[0102] Figure 5 A block diagram of a user interaction device provided in an embodiment of the present disclosure.

[0103] Reference Figure 5 The embodiment of the present disclosure provides a user interaction device, which includes: an acquisition unit 501, an emotion category recognition unit 502, a speech generation unit 503 and an interaction unit 504.

[0104] The acquisition unit 501 is used to acquire user behavior data and user portrait data of the target user.

[0105] The emotion category identification unit 502 is used to obtain the emotion category of the target user for the target object according to the user behavior data and the user portrait data.

[0106] The speech generating unit 503 is used to generate target speech terms according to the emotion category and the user portrait data, wherein the target speech terms correspond to the emotion category and are used to guide the user's emotion category to convert to a positive emotion category.

[0107] The interaction unit 504 is used to interact with the target user based on the target speech terms.

[0108] In some embodiments, when the emotion category recognition unit 502 obtains the emotion category of the target user for the target object based on the user behavior data and the user portrait data, it can be used to: determine the target user category of the target user based on the user portrait data; determine the emotion category recognition model corresponding to the target user based on the target user category; input the user behavior data into the emotion category recognition model to perform emotion category recognition processing to obtain the emotion category.

[0109] In some embodiments, when the speech generation unit 503 generates target speech terms based on emotion categories and user portrait data, it can be used to: determine a resource recommendation model corresponding to the target user based on the target user category, wherein the resource recommendation model is used to generate resource data corresponding to the user intention characteristics based on the user intention characteristics represented by the user behavior data; input the user behavior data into the resource recommendation model for resource prediction processing to obtain target resource data; generate target speech terms based on the emotion category and the target resource data.

[0110] In some embodiments, when the speech generation unit 503 generates target speech terms based on the emotion category and the target resource data, it can be used to: input the emotion category and the target resource data into the speech prediction model to perform speech prediction processing to obtain the target speech terms.

[0111] In some embodiments, when the interaction unit 504 interacts with the target user based on the target speech terms, it can be used to: perform voice outbound call processing for the target user; in the process of outbound voice interaction with the target user, input the acquired user real-time voice data into the strategy prediction model for interaction strategy prediction processing to obtain the target interaction strategy; based on the target interaction strategy and the target speech terms, generate corrected speech terms; and perform voice interaction with the target user based on the corrected speech terms.

[0112] In some embodiments, when the interaction unit 504 interacts with the target user based on the target speech terms, it can be used to: perform outbound voice call processing for the target user; determine the target timbre according to the gender of the target user during the outbound voice interaction with the target user; generate real-time interactive voice data according to the target timbre and target speech terms; and perform voice interaction with the target user based on the real-time interactive voice data.

[0113] In some embodiments, when the interaction unit 504 interacts with the target user based on the target speech terms, it can be used to: obtain auxiliary channel information of the auxiliary interaction channel for auxiliary interaction with the target user, wherein the auxiliary channel information includes at least one of the text messages, emails and social platforms corresponding to the target user; generate interaction push information based on the target speech terms; push the interaction push information to the auxiliary interaction channel based on the channel information; after pushing the interaction push information to the auxiliary interaction channel, perform voice interaction with the target user based on the target speech terms.

[0114] It can be seen that according to the user interaction device provided by the embodiment of the present disclosure, the user behavior data and user portrait data of the target user are acquired based on the acquisition unit, and the emotion category of the target user for the target object is obtained based on the emotion category recognition unit according to the user behavior data and the user portrait data; thereafter, the target speech terms are generated according to the emotion category and the user portrait data based on the speech term generation unit, and the interaction unit interacts with the target user based on the target speech terms. Since the target speech terms correspond to the emotion category of the target user for the target object and can guide the user's emotion to convert to a positive emotion category, the intelligent voice robot can interact with the user based on the embodiment of the present disclosure so as to respond sensitively to the user's emotions, provide more considerate services, and enhance the emotional connection with the user, thereby achieving the effect of improving the user experience.

[0115] Each module in the above user interaction device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0116] Figure 6 A block diagram of an electronic device provided in an embodiment of the present disclosure.

[0117] Reference Figure 6 An embodiment of the present disclosure provides an electronic device, which includes: at least one processor 601; at least one memory 602, and one or more I / O interfaces 603, connected between the processor 601 and the memory 602; wherein the memory 602 stores one or more computer programs that can be executed by the at least one processor 601, and the one or more computer programs are executed by the at least one processor 601 so that the at least one processor 601 can execute the above-mentioned user interaction method.

[0118] Each module in the above electronic device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0119] The present disclosure also provides a computer-readable storage medium on which a computer program is stored, wherein the computer program implements the above-mentioned user interaction method when executed by a processor. The computer-readable storage medium may be a volatile or non-volatile computer-readable storage medium.

[0120] The embodiment of the present disclosure also provides a computer program product, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above-mentioned user interaction method.

[0121] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable storage medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium).

[0122] As is known to those of ordinary skill in the art, the term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable program instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technology, portable compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. In addition, it is known to those of ordinary skill in the art that communication media typically contain computer-readable program instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and may include any information delivery medium.

[0123] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0124] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0125] The computer program product described herein may be implemented in hardware, software, or a combination thereof. In one optional embodiment, the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (SDK), etc.

[0126] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0127] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0128] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0129] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0130] Example embodiments have been disclosed herein, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for limiting purposes. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly noted, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, those skilled in the art will appreciate that various changes in form and detail may be made without departing from the scope of the present disclosure as set forth in the appended claims.

Claims

1. A user interaction method, characterized in that: include: Obtain user behavior data and user portrait data of target users; Obtaining the emotion category of the target user toward the target object according to the user behavior data and the user portrait data; Generate target speech terms according to the emotion category and the user portrait data, wherein the target speech terms correspond to the emotion category and are used to guide the user's emotion category to convert to a positive emotion category; Interact with the target user based on the target speech terms.

2. The method according to claim 1, characterized in that The step of obtaining the emotion category of the target user toward the target object according to the user behavior data and the user portrait data includes: Determining a target user category of the target user according to the user portrait data; Determining an emotion category recognition model corresponding to the target user according to the target user category; The user behavior data is input into the emotion category recognition model to perform emotion category recognition processing to obtain the emotion category.

3. The method according to claim 2, characterized in that Generating target speech terms according to the emotion category and the user portrait data includes: Determine a resource recommendation model corresponding to the target user according to the target user category, wherein the resource recommendation model is used to generate resource data corresponding to the user intention feature according to the user intention feature represented by the user behavior data; Inputting the user behavior data into the resource recommendation model to perform resource prediction processing to obtain target resource data; The target speech term is generated according to the emotion category and the target resource data.

4. The method according to claim 3, characterized in that: The step of generating the target speech term according to the emotion category and the target resource data includes: The emotion category and the target resource data are input into a speech prediction model to perform speech prediction processing to obtain the target speech terms.

5. The method according to claim 1, characterized in that The interacting with the target user based on the target speech terminology includes: Performing voice outbound call processing on the target user; In the process of outbound voice interaction with the target user, the acquired real-time voice data of the user is input into the strategy prediction model for interaction strategy prediction processing to obtain the target interaction strategy; Based on the target interaction strategy and the target speech terms, generating correction speech terms; Perform voice interaction with the target user based on the corrected speech terms.

6. The method according to claim 1, characterized in that The interacting with the target user based on the target speech terminology includes: Performing voice outbound call processing on the target user; During the outbound voice interaction with the target user, determining a target timbre according to the gender of the target user; Generating real-time interactive voice data according to the target timbre and the target speech terms; Perform voice interaction with the target user according to the real-time interactive voice data.

7. The method according to claim 1, characterized in that The interacting with the target user based on the target speech terminology includes: Acquire auxiliary channel information of an auxiliary interaction channel for auxiliary interaction with the target user, wherein the auxiliary channel information includes at least one of a text message, an email, and a social platform corresponding to the target user; Generate interactive push information based on the target speech terms; Based on the channel information, the interaction push information is pushed to the auxiliary interaction channel; After the interaction push information is pushed to the auxiliary interaction channel, voice interaction is performed with the target user based on the target speech terms.

8. A user interaction device, characterized in that: include: An acquisition unit, used to acquire user behavior data and user portrait data of target users; An emotion category recognition unit, used to obtain the emotion category of the target user towards the target object according to the user behavior data and the user portrait data; A speech generating unit, used to generate target speech terms according to the emotion category and the user portrait data, wherein the target speech terms correspond to the emotion category and are used to guide the user's emotion category to convert to a positive emotion category; An interaction unit is used to interact with the target user based on the target speech terms.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores one or more computer programs executable by the at least one processor, and the one or more computer programs are executed by the at least one processor to enable the at least one processor to perform the user interaction method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the computer program implements the user interaction method according to any one of claims 1 to 7.

Images