A Human-Computer Interaction Method and Device Based on an AI Intelligent Pen Holder

Abstract

The present application provides a human-computer interaction method and device based on an AI intelligent pen holder, which is applicable to the field of data processing technology. The method includes: processing historical dialogue information and question voice information, and generating an initial historical dialogue keyword vector and an initial question voice keyword vector according to the generated historical dialogue text vector, question voice text vector, adjacent region radius, and adjacent region vector quantity information; generating a target historical dialogue keyword vector and a target question voice keyword vector according to the initial historical dialogue keyword vector, initial question voice keyword vector, and feature mapping vector, and combining the historical dialogue text vector to generate a reply text information and a reply voice information. The present application accurately extracts the keywords of the historical dialogue content and the current question content, clarifies the specific context and sub-topics through the keywords, generates reply information by introducing the reference of the historical dialogue content, and improves the context understanding ability of the AI intelligent pen holder through in-depth semantic analysis.

Description

Technical Field

[0001] This application belongs to the technical field of data processing, and particularly relates to a human-computer interaction method and device based on an AI intelligent pen holder. Background Art

[0002] With the rapid development of artificial intelligence technology and the growing demand for convenient interaction among people, the AI intelligent pen holder has emerged, which is used to provide users with a more intelligent and efficient information acquisition and interaction method, especially suitable for office and learning scenarios, helping people record, query, and process information more easily.

[0003] Existing AI pen holders usually integrate technologies such as speech recognition and natural language processing. The voice commands instantaneously input by users are converted into text through speech recognition, the semantics in the voice commands are understood by using natural language processing technology, and then accurate answers are provided according to the semantic parsing results.

[0004] However, in the existing technology, the AI dialogue process lacks in-depth mining of the dialogue content, it is difficult to capture the semantic associations between keywords, and the understanding of context and semantics is insufficient, resulting in low accuracy of keyword extraction and poor pertinence of the reply content. Summary of the Invention

[0005] In view of this, the embodiments of this application provide a human-computer interaction method and device based on an AI intelligent pen holder, aiming to solve the problems in the existing natural language processing technology that it is impossible to accurately extract keywords from the dialogue content provided by users, and there is a large error in understanding the context of the dialogue content, resulting in poor quality of the generated reply content.

[0006] The first aspect of the embodiments of this application provides a human-computer interaction method based on an AI intelligent pen holder, including:

[0007] Obtain historical dialogue information and the voice information of the question input by the user;

[0008] Perform conversion processing on the historical dialogue information and the question voice information to generate a plurality of historical dialogue text vectors and a plurality of question voice text vectors;

[0009] Generate an initial historical dialogue keyword vector and an initial question voice keyword vector according to a plurality of the historical dialogue text vectors, a plurality of the question voice text vectors, a preset adjacent region radius, and preset adjacent region vector quantity information;

[0010] Generate a target historical dialogue keyword vector and a target question voice keyword vector according to the initial historical dialogue keyword vector, the initial question voice keyword vector, and a plurality of preset feature mapping vectors;

[0011] Construct a dialogue text feature structure diagram based on the historical dialogue text vector and the target historical dialogue keyword vector;

[0012] Generate a reply text message and a reply voice message based on the dialogue text feature structure diagram and the target question voice text keyword vector.

[0013] The second aspect of the embodiments of the present application provides a human-computer interaction device based on an AI intelligent pen holder, including:

[0014] An information acquisition module, configured to acquire historical dialogue information and the question voice information input by the user;

[0015] A text vector generation module, configured to perform conversion processing on the historical dialogue information and the question voice information to generate a plurality of historical dialogue text vectors and a plurality of question voice text vectors;

[0016] An initial keyword vector generation module, configured to generate an initial historical dialogue keyword vector and an initial question voice keyword vector according to the plurality of historical dialogue text vectors, the plurality of question voice text vectors, a preset adjacent region radius, and preset adjacent region vector quantity information;

[0017] A target keyword vector generation module, configured to generate a target historical dialogue keyword vector and a target question voice keyword vector according to the initial historical dialogue keyword vector, the initial question voice keyword vector, and a plurality of preset feature mapping vectors;

[0018] A dialogue text feature structure diagram construction module, configured to construct a dialogue text feature structure diagram according to the historical dialogue text vector and the target historical dialogue keyword vector;

[0019] A reply information generation module, configured to generate a reply text message and a reply voice message according to the dialogue text feature structure diagram and the target question voice text keyword vector.

[0020] The beneficial effects of the embodiments of the present application compared with the prior art are as follows: The preliminary extraction of keyword information is performed based on the semantic feature density in the historical dialogue information and the question content information, and then the keyword information preliminarily extracted is strengthened through the preset feature mapping vectors, so as to filter out the redundant information mixed in during the preliminary extraction, so as to achieve the accurate extraction of keywords. By introducing historical dialogue information for semantic analysis, the semantic association features between each keyword in the historical dialogue information and the question content information are captured, so as to improve the context understanding ability of the AI intelligent pen holder for the dialogue content, and improve the accuracy and pertinence of the automatically generated reply information. Description of the Drawings

[0021] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for use in the embodiments or the description of the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0022] Figure 1 It is a schematic flowchart of the implementation of the human-computer interaction method based on an AI intelligent pen holder provided in the first embodiment of the present application;

[0023] Figure 2 It is a schematic flowchart of the implementation of the human-computer interaction method based on an AI intelligent pen holder provided in the second embodiment of the present application;

[0024] Figure 3 It is a schematic flowchart of the implementation of the human-computer interaction method based on an AI intelligent pen holder provided in the third embodiment of the present application;

[0025] Figure 4 It is a schematic flowchart of the implementation of the human-computer interaction method based on an AI intelligent pen holder provided in the fourth embodiment of the present application;

[0026] Figure 5 It is a schematic flowchart of the implementation of the human-computer interaction method based on an AI intelligent pen holder provided in the fifth embodiment of the present application;

[0027] Figure 6 It is a schematic flowchart of the implementation of the human-computer interaction method based on an AI intelligent pen holder provided in the sixth embodiment of the present application;

[0028] Figure 7 It is a schematic flowchart of the implementation of the human-computer interaction method based on an AI intelligent pen holder provided in the seventh embodiment of the present application;

[0029] Figure 8 It is a schematic flowchart of the implementation of the human-computer interaction method based on an AI intelligent pen holder provided in the seventh embodiment of the present application;

[0030] Figure 9 It is a schematic flowchart of the implementation of the human-computer interaction method based on an AI intelligent pen holder provided in the seventh embodiment of the present application;

[0031] Figure 10 It is a schematic structural diagram of the human-computer interaction device based on an AI intelligent pen holder provided in the embodiments of the present application. Detailed implementation manners

[0032] In the following description, specific details such as specific system architectures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0033] In order to illustrate the technical solutions described in the present application, the following will be described through specific embodiments.

[0034] Figure 1 The implementation flowchart of the human-computer interaction method based on the AI intelligent pen holder provided in the first embodiment of the present application is shown and described in detail as follows:

[0035] Step S101, obtain historical conversation information and the voice information of the question input by the user.

[0036] In this embodiment, the historical conversation information can be extracted from the memory of the AI intelligent pen holder, and can be the instruction information or question information input by the user during the past use of the AI intelligent pen holder, as well as the reply content automatically generated by the AI intelligent pen holder during past use. The voice information of the question can refer to the questions or instruction information raised by the user to the AI intelligent pen holder, and can be obtained through a component with a voice pickup function.

[0037] Step S102, perform conversion processing on the historical conversation information and the voice information of the question to generate a plurality of historical conversation text vectors and a plurality of voice text vectors of the question.

[0038] In this embodiment, the historical conversation information can be preprocessed first, which can include operations such as removing special characters, punctuation marks, and converting the case of letters to unify the text format, and then word segmentation processing, that is, splitting the sentence into multiple independent words or phrases, and then converting each word or phrase into a vector through a word embedding model to obtain historical conversation text vectors. The voice text vectors of the question can be first converted into text content through speech recognition technology, and then the text content is preprocessed, segmented, word-embedded, and vector aggregated to generate a plurality of voice text vectors of the question.

[0039] Step S103, generate an initial historical conversation keyword vector and an initial voice keyword vector of the question according to the plurality of historical conversation text vectors, the plurality of voice text vectors of the question, the preset adjacent area radius, and the preset adjacent area vector quantity information.

[0040] In this embodiment, the preset adjacent region radius and the preset number of adjacent region vectors information can be set manually. It can be understood that when the distribution of semantic feature information in the multi-dimensional space is relatively dense, that is, the similarity between semantic feature information is relatively high, the preset adjacent region radius can take a relatively small value. On the contrary, when the distribution of semantic feature information in the multi-dimensional space is relatively sparse, that is, the similarity between semantic feature information is relatively low, the preset adjacent region radius can take a relatively large value. Among them, the preset adjacent region radius can be from 0.1 to 0.5, or it can be from 1 to 5. The preset number of adjacent region vectors information can be related to the dimensions of the historical dialogue text vector and the question voice text vector, and the value can be 300.

[0041] It can be to first calculate the similarity between each historical dialogue text vector and each question voice text vector, and then determine the text vector information included in the adjacent region according to the size relationship between the similarity and the preset adjacent region radius. Then, count the number of text vector information in each adjacent region, and judge the size relationship between the number of text vector information in the adjacent region and the preset number of adjacent region vectors information. Screen the adjacent regions where the number of text vector information in the adjacent region is greater than the preset number of adjacent region vectors information. It can be to use the text vector information in the screened adjacent region as the initial historical dialogue keyword vector and the initial question voice keyword vector; it can also be to calculate the similarity of the text vector information in the screened adjacent region, aggregate the text vector information with high similarity, and use the aggregated vector as the initial historical dialogue keyword vector and the initial question voice keyword vector. Among them, the similarity can be the cosine similarity. This embodiment is used for the preliminary extraction of historical dialogue keywords and question voice text keywords based on the density of semantic features in historical dialogue text information and question voice text information.

[0042] Step S104, generate a target historical dialogue keyword vector and a target question voice keyword vector according to the initial historical dialogue keyword vector, the initial question voice keyword vector, and multiple preset feature mapping vectors.

[0043] In this embodiment, multiple preset feature mapping vectors can be set manually. By performing dot product operations on the multiple preset feature mapping vectors with the initial historical dialogue keyword vector and the initial query voice keyword vector respectively, the distances between the values in each historical dialogue keyword vector and query voice keyword vector can be enlarged, so as to strengthen the initially extracted keyword vectors and avoid the mixing of redundant information. The strengthened historical dialogue keyword vector and query voice keyword vector are used as the target historical dialogue keyword vector and the target query voice keyword vector. It can be understood that in the existing natural language processing technologies, in keyword extraction, it is often difficult to accurately capture semantic associations and easy to ignore the context, resulting in one-sided extracted keywords. However, in this embodiment, by deeply extracting and analyzing semantic features, the accuracy of keyword extraction is improved.

[0044] Step S105: Construct a dialogue text feature structure diagram according to the historical dialogue text vector and the target historical dialogue keyword vector.

[0045] In this embodiment, the historical dialogue text vector and the target historical dialogue keyword vector can be used as the nodes of the dialogue text feature structure diagram, and the logical distances between the historical dialogue text vectors and the distances between the target historical dialogue keyword vectors are calculated as the edges between the nodes, so as to construct the dialogue text feature structure diagram through the nodes and edges.

[0046] Step S106: Generate a reply text message and a reply voice message according to the dialogue text feature structure diagram and the target query voice text keyword vector.

[0047] In this embodiment, the target question speech text keyword vector can be randomly inserted into the dialogue text feature structure diagram as a new node of the dialogue text feature structure diagram, and the original node of the dialogue text feature structure diagram is used as the old node, and then the logical distance between the new node and each old node is calculated. A logical distance threshold can be preset artificially. When the logical distance between the new node and each old node is less than the logical distance threshold, the old node is used as a reference node, and the logical distance between the reference node and the new node is used as a weight coefficient. The new node and the reference node are weighted and summed, and the results calculated for multiple new nodes are semantically spliced ​​as a reply text vector, and then the reply text vector is decoded by the language model to generate a reply text message, and then the reply text message is converted into a reply voice message by the speech synthesis technology. Among them, the reply text message can be visually output through the electronic screen of the AI ​​smart pen holder, and the reply voice message can be output through the speaker of the AI ​​smart pen holder. It is understandable that in the existing human-computer dialogue technology, usually only the question information currently provided by the user is analyzed and processed, which leads to large errors in the contextual understanding of the dialogue content, shallow analysis of complex semantics, and difficulty in accurately grasping the user's true intentions, resulting in poor quality of the generated reply content. This embodiment introduces historical dialogue content for reference and conducts deep semantic analysis to improve the AI ​​smart pen holder's ability to understand the context.

[0048] The human-computer interaction method based on the AI ​​smart pen holder provided in the embodiment of the present application performs preliminary extraction of keyword information based on the semantic feature density in the historical conversation information and the question content information, and then enhances the preliminary extracted keyword information through a preset feature mapping vector, thereby filtering out redundant information mixed in during the preliminary extraction to achieve accurate extraction of keywords, and by introducing historical conversation information for semantic analysis, the semantic association features between each keyword in the historical conversation information and the question content information are captured, thereby improving the AI ​​smart pen holder's ability to understand the context of the conversation content and improving the accuracy and pertinence of the automatically generated reply information.

[0049] Figure 2 The flowchart of the implementation of the human-computer interaction method based on the AI ​​smart pen holder provided in the second embodiment of the present application is shown. The difference between the second embodiment and the first embodiment is that the step S103 specifically includes:

[0050] Step S201, generating a human-computer interaction text vector set according to a plurality of the historical conversation text vectors and a plurality of the question voice text vectors; the human-computer interaction text vector set includes a plurality of human-computer interaction text vectors.

[0051] In this embodiment, it may be to splice a plurality of historical dialogue text vectors and a plurality of question voice text vectors to generate a set of human-computer interaction text vectors, which is used for subsequent preliminary semantic extraction of keyword features in the historical dialogue text and the question voice text. It can be understood that the elements in the set of human-computer interaction text vectors are used as human-computer interaction text vectors to determine the density of semantic feature information in a multi-dimensional space during subsequent calculations.

[0052] Step S202: Randomly extract elements from the set of human-computer interaction text vectors to obtain a human-computer interaction text center vector.

[0053] In this embodiment, it may be to randomly extract an element from the set of human-computer interaction text vectors, that is, randomly extract a human-computer interaction text vector as the human-computer interaction text center vector.

[0054] Step S203: Calculate the Euclidean distance between each of the human-computer interaction text vectors and the human-computer interaction text center vector to obtain an interaction text vector distance.

[0055] In this embodiment, it may be to first calculate the Euclidean distance between each human-computer interaction text vector and the human-computer interaction text center vector to obtain the interaction text vector distance of multiple human-computer interaction text vectors corresponding to the human-computer interaction text center vector.

[0056] Step S204: Determine whether the interaction text vector distance is less than or equal to a preset adjacent region radius; if so, use the human-computer interaction text vector corresponding to the interaction text vector distance as the human-computer interaction text adjacent region vector; if not, skip the human-computer interaction text vector corresponding to the interaction text vector distance.

[0057] In this embodiment, the preset adjacent region radius can be set manually. When the distance of the interactive text vector is less than or equal to the preset adjacent region radius, it indicates that the distance between the human-computer interaction text vector and the human-computer interaction text center vector in the vector space is small, that is, the semantic feature similarity is high. The human-computer interaction text vector can be classified into the adjacent region of the human-computer interaction text center vector, and the semantic feature representing the keyword is used to synthesize the keyword vector subsequently. When the distance of the interactive text vector is greater than the preset adjacent region radius, it indicates that the distance between the human-computer interaction text vector and the human-computer interaction text center vector in the vector space is large, that is, the semantic feature similarity is low. The human-computer interaction text vector cannot be classified into the adjacent region of the human-computer interaction text center vector. Then, the human-computer interaction text vector corresponding to the distance of the interactive text vector is skipped, and the current human-computer interaction text vector still exists in the human-computer interaction text vector set as the human-computer interaction text vector. Then, it is judged whether the distance between the next human-computer interaction text vector and the human-computer interaction text center vector can classify the next human-computer interaction text vector into the adjacent region of the human-computer interaction text center vector. Until all human-computer interaction text vectors are traversed, the adjacent region of the human-computer interaction text center vector is generated.

[0058] Step S205: Generate an initial historical dialogue keyword vector and an initial question voice keyword vector according to the human-computer interaction text adjacent region vector and the preset adjacent region vector quantity information.

[0059] In this embodiment, it can be to first count the quantity of the human-computer interaction text adjacent region vectors in the adjacent region of each human-computer interaction text center vector. The adjacent region where the quantity of the human-computer interaction text adjacent region vectors exceeds the preset adjacent region vector quantity information is used as the set for generating keyword vectors. Then, the vectors in the set are extracted according to their sources. The elements from the historical dialogue information are extracted to generate the initial historical dialogue keyword vector, and the elements from the question voice information are extracted to generate the initial question voice keyword vector.

[0060] The human-computer interaction method based on the AI intelligent pen holder provided by the embodiment of the present application automatically identifies complex semantic features with irregular logical shapes in the vector space. The automatic identification result is used as the adjacent region to achieve flexible category division of semantic features. At the same time, the semantic feature vectors that cannot be classified into the adjacent region are filtered out, so as to improve the accuracy of keyword extraction of the AI intelligent pen holder and improve the quality and efficiency of human-computer interaction.

[0061] Figure 3 The implementation flowchart of the human-computer interaction method based on the AI intelligent pen holder provided by the third embodiment of the present application is shown. The difference from the second embodiment above is that the step S205 specifically includes:

[0062] Step S301, count the number of vectors in the adjacent regions of the human-computer interaction text to obtain the information on the number of neighborhood vectors of the interaction text.

[0063] In this embodiment, count the number of vectors in the adjacent regions of all human-computer interaction texts, that is, count the number of vectors in the adjacent regions of the central vectors of each human-computer interaction text to obtain the information on the number of neighborhood vectors of the interaction text, which is used to determine whether the adjacent region is valid in the subsequent process.

[0064] Step S302, determine whether the information on the number of neighborhood vectors of the interaction text is greater than or equal to the preset information on the number of adjacent region vectors; if so, proceed to step S303; if not, proceed to step S304.

[0065] In this embodiment, when the information on the number of neighborhood vectors of the interaction text is greater than or equal to the preset information on the number of adjacent region vectors, it indicates that the density of semantic feature points in this adjacent region is relatively large, so the validity of this adjacent region is high and it can be used to generate keyword information; when the information on the number of neighborhood vectors of the interaction text is less than the preset information on the number of adjacent region vectors, it indicates that the density of semantic feature points in this adjacent region is relatively small, so the validity of this adjacent region is low, and whether it can be used to generate keyword information needs to be further determined subsequently.

[0066] Step S303, determine the central vector of the human-computer interaction text as the core vector of the human-computer interaction text.

[0067] In this embodiment, when the information on the number of neighborhood vectors of the interaction text is greater than or equal to the preset information on the number of adjacent region vectors, it indicates that the density of semantic feature points in this adjacent region is relatively large, so the validity of this adjacent region is high and it can be used to generate keyword information, and then all the vectors in this adjacent region can be determined as the core vectors of the human-computer interaction text.

[0068] Step S304, determine the central vector of the human-computer interaction text as the non-core vector of the human-computer interaction text.

[0069] In this embodiment, when the information on the number of neighborhood vectors of the interaction text is less than the preset information on the number of adjacent region vectors, it indicates that the density of semantic feature points in this adjacent region is relatively small, so the validity of this adjacent region is low, and whether it can be used to generate keyword information needs to be further determined subsequently, and then all the vectors in this adjacent region can be determined as the non-core vectors of the human-computer interaction text.

[0070] Step S305, count the number of core vectors and non-core vectors of the human-computer interaction text to obtain the number of core vectors of the interaction text and the number of non-core vectors of the interaction text.

[0071] In this embodiment, count the number of all core vectors of the human-computer interaction texts, and use the statistical result as the number of core vectors of the interaction texts; count the number of all non-core vectors of the human-computer interaction texts, and use the statistical result as the number of non-core vectors of the interaction texts.

[0072] Step S306, determine whether the sum of the number of core vectors of the interaction texts and the number of non-core vectors of the interaction texts is equal to the number of elements in the human-computer interaction text vector set; if so, proceed to step S307; if not, proceed to step S308.

[0073] In this embodiment, it is used to determine whether all the human-computer interaction text vectors in the human-computer interaction text vector set have been traversed; when the sum of the number of core vectors of the interaction texts and the number of non-core vectors of the interaction texts is equal to the number of elements in the human-computer interaction text vector set, it means that all the human-computer interaction text vectors in the human-computer interaction text vector set have been traversed, and all the human-computer interaction text vectors have been updated to core vectors of the human-computer interaction texts or non-core vectors of the human-computer interaction texts, and there is no need to perform traversal calculation anymore, then the synthesis process of the keyword vectors can be carried out through the core vectors of the human-computer interaction texts and the non-core vectors of the human-computer interaction texts; when the sum of the number of core vectors of the interaction texts and the number of non-core vectors of the interaction texts is less than the number of elements in the human-computer interaction text vector set, it means that not all the human-computer interaction text vectors in the human-computer interaction text vector set have been traversed, and then iterative calculation needs to be performed again to ensure that all the human-computer interaction text vectors are traversed and ensure that all the human-computer interaction text vectors are updated to core vectors of the human-computer interaction texts or non-core vectors of the human-computer interaction texts.

[0074] Step S307, generate an initial historical dialogue keyword vector and an initial question voice keyword vector according to the core vectors of the human-computer interaction texts and the non-core vectors of the human-computer interaction texts.

[0075] In this embodiment, when the sum of the number of core vectors of the interaction texts and the number of non-core vectors of the interaction texts is equal to the number of elements in the human-computer interaction text vector set, it means that all the human-computer interaction text vectors in the human-computer interaction text vector set have been traversed, and all the human-computer interaction text vectors have been updated to core vectors of the human-computer interaction texts or non-core vectors of the human-computer interaction texts, and there is no need to perform traversal calculation anymore, then the synthesis process of the keyword vectors can be carried out through the core vectors of the human-computer interaction texts and the non-core vectors of the human-computer interaction texts.

[0076] Step S308, obtain a remaining interaction text vector set according to the human-computer interaction text vector set, the core vectors of the human-computer interaction texts, and the non-core vectors of the human-computer interaction texts.

[0077] In this embodiment, when the sum of the number of interactive text core vectors and the number of interactive text non-core vectors is less than the number of elements in the human-computer interactive text vector set, it means that all human-computer interactive text vectors in the human-computer interactive text vector set have not been traversed, and it is necessary to re-iterate and calculate, and it is necessary to determine the human-computer interactive text vectors that have not been traversed. The elements in the human-computer interactive text vector set that have been updated to human-computer interactive text core vectors or human-computer interactive text non-core vectors can be eliminated to generate a remaining interactive text vector set for performing the traversal operation of the remaining human-computer interactive text vectors.

[0078] Step S309: taking the remaining interaction text vector set as the human-computer interaction text vector set, and returning to step S202.

[0079] In this embodiment, the remaining interactive text vector set is used as the human-computer interaction text vector set, and the elements are randomly re-extracted, the similarity is calculated, and the adjacent areas are divided, so as to realize recursive operation, which is used to ensure that all human-computer interaction text vectors are updated to human-computer interaction text core vectors or human-computer interaction text non-core vectors.

[0080] The human-computer interaction method based on the AI ​​smart pen holder provided in the embodiment of the present application avoids the omission of key semantic features in historical dialogue information and question content information by traversing and updating all human-computer interaction text vectors, and continuously expands the number of semantic feature vectors that can be used to generate keywords by continuously incorporating human-computer interaction text vectors into adjacent areas, thereby ensuring the accuracy and comprehensiveness of the generated keywords, thereby enabling the AI ​​smart pen holder to understand the dialogue topic more accurately, and then provide reply information that is more in line with the context, so as to effectively improve the human-computer interaction experience of the AI ​​smart pen holder.

[0081] Figure 4 The flowchart of the implementation of the human-computer interaction method based on the AI ​​smart pen holder provided in the fourth embodiment of the present application is shown. The difference between the fourth embodiment and the third embodiment is that the step S307 specifically includes:

[0082] Step S401, calculating the Euclidean distance between each non-core vector of the human-computer interaction text and each core vector of the human-computer interaction text to obtain the confirmation distance of the non-core point of the interaction text.

[0083] In this embodiment, the Euclidean distance between each non-core vector of the human-computer interaction text and each core vector of the human-computer interaction text is calculated, and the calculation result is used as the confirmation distance of the non-core point of the interaction text to further determine the validity of each non-core vector of the human-computer interaction text.

[0084] Step S402: Determine whether the confirmation distance of the non-core point of the interactive text is less than or equal to the preset adjacent area radius. If so, proceed to step S403; if not, proceed to step S404.

[0085] In this embodiment, when the confirmation distance of the non-core point of the interactive text is less than or equal to the preset adjacent area radius, it indicates that the semantic feature information represented by the non-core vector of the human-computer interaction text has a strong correlation with other semantic feature information, and the semantic validity of the non-core vector of the human-computer interaction text is relatively high, which can be used to generate keyword information. When the confirmation distance of the non-core point of the interactive text is greater than the preset adjacent area radius, it indicates that the semantic feature information represented by the non-core vector of the human-computer interaction text has a weak correlation with other semantic feature information, and the semantic validity of the non-core vector of the human-computer interaction text is relatively low, which can be used to generate keyword information.

[0086] Step S403: Use the non-core vector of the human-computer interaction text corresponding to the confirmation distance of the non-core point of the interactive text as the edge point vector of the human-computer interaction text.

[0087] In this embodiment, when the confirmation distance of the non-core point of the interactive text is less than or equal to the preset adjacent area radius, it indicates that the semantic feature information represented by the non-core vector of the human-computer interaction text has a strong correlation with other semantic feature information, and the semantic validity of the non-core vector of the human-computer interaction text is relatively high. The non-core vector of the human-computer interaction text can be used as the edge point vector of the human-computer interaction text to generate keyword information.

[0088] Step S404: Use the non-core vector of the human-computer interaction text corresponding to the confirmation distance of the non-core point of the interactive text as the redundant point vector of the human-computer interaction text.

[0089] In this embodiment, when the confirmation distance of the non-core point of the interactive text is greater than the preset adjacent area radius, it indicates that the semantic feature information represented by the non-core vector of the human-computer interaction text has a weak correlation with other semantic feature information, and the semantic validity of the non-core vector of the human-computer interaction text is relatively low. The non-core vector of the human-computer interaction text is used as the redundant point vector of the human-computer interaction text, that is, it is removed as noise and not used to generate keyword information.

[0090] Step S405: Generate multiple interactive text keyword vectors based on the core vector of the human-computer interaction text and the edge point vector of the human-computer interaction text.

[0091] In this embodiment, the semantic content represented by the core vector of the human-computer interaction text can be used as the main reference information, and the semantic content represented by the edge point vector of the human-computer interaction text can be used as the auxiliary reference information. The interaction text keyword vector includes a main keyword vector and an auxiliary keyword vector. The core vector of the human-computer interaction text is aggregated to generate a main keyword vector; the edge point vector of the human-computer interaction text is aggregated to generate an auxiliary keyword vector. Among them, in subsequent calculations, a higher weight can be assigned to the main keyword vector, and a lower weight can be assigned to the auxiliary keyword vector.

[0092] Step S406: Perform numerical separation and extraction processing on the interaction text keyword vector to generate an initial historical dialogue keyword vector and an initial question voice keyword vector.

[0093] In this embodiment, the semantic features represented in the interaction text keyword vector can be separated according to different sources to obtain an initial historical dialogue keyword vector and an initial question voice keyword vector.

[0094] The human-computer interaction method based on the AI intelligent pen barrel provided by the embodiment of the present application determines the key semantic feature information in the dialogue content through the core vector of the human-computer interaction text, which is used to quickly lock the core content in the dialogue information, improve the accuracy and effectiveness of the AI intelligent pen barrel in grasping the main idea of the dialogue content, expand the semantic feature information in the dialogue content through the edge point vector of the human-computer interaction text, incorporate the dialogue content closely related to the core content, enrich the content of the keywords, and improve the comprehensiveness of keyword extraction, thereby helping the AI intelligent pen barrel to provide more detailed and user-demand-fitting reply information for users, effectively improving the quality and efficiency of human-computer interaction.

[0095] Figure 5 The flowchart of the implementation of the human-computer interaction method based on the AI intelligent pen barrel provided in the fifth embodiment of the present application is shown. The difference from the first embodiment above is that:

[0096] The multiple preset feature mapping vectors include a preset retrieval mapping vector, a preset index mapping vector, and a preset content mapping vector;

[0097] The step S104 specifically includes:

[0098] Step S501: Calculate an initial historical dialogue keyword retrieval vector, an initial historical dialogue keyword index vector, an initial historical dialogue keyword content vector, an initial question voice keyword retrieval vector, an initial question voice keyword index vector, and an initial question voice keyword content vector according to the initial historical dialogue keyword vector, the initial question voice keyword vector, the preset retrieval mapping vector, the preset index mapping vector, and the preset content mapping vector.

[0099] In this embodiment, the preset retrieval mapping vector, the preset index mapping vector, and the preset content mapping vector can all be set manually, and are used to further widen the distances of each semantic feature in the vector space. It can be to perform a dot product operation on the initial historical dialogue keyword vector and the preset retrieval mapping vector to obtain an initial historical dialogue keyword retrieval vector; it can be to perform a dot product operation on the initial historical dialogue keyword vector and the preset index mapping vector to obtain an initial historical dialogue keyword index vector; it can be to perform a dot product operation on the initial historical dialogue keyword vector and the preset content mapping vector to obtain an initial historical dialogue keyword content vector; it can be to perform a dot product operation on the initial query voice keyword vector and the preset retrieval mapping vector to obtain an initial query voice keyword retrieval vector; it can be to perform a dot product operation on the initial query voice keyword vector and the preset index mapping vector to obtain an initial query voice keyword index vector; it can be to perform a dot product operation on the initial query voice keyword vector and the preset content mapping vector to obtain an initial query voice keyword content vector.

[0100] Step S502: Calculate an initial historical dialogue keyword query variable according to the initial historical dialogue keyword retrieval vector and the initial historical dialogue keyword index vector.

[0101] In this embodiment, it can be to perform a dot product operation on the initial historical dialogue keyword retrieval vector and the transposed matrix of the initial historical dialogue keyword index vector, and the calculation result is used as the initial historical dialogue keyword query variable.

[0102] Step S503: Calculate an initial query voice keyword query variable according to the initial query voice keyword retrieval vector and the initial query voice keyword index vector.

[0103] In this embodiment, it can be to perform a dot product operation on the initial query voice keyword retrieval vector and the transposed matrix of the initial query voice keyword index vector, and the calculation result is used as the initial query voice keyword query variable.

[0104] Step S504: Perform a normalization process on the initial historical dialogue keyword query variable and the initial query voice keyword query variable to obtain an initial historical keyword query scaling variable and an initial query keyword query scaling variable.

[0105] In this embodiment, the normalization process can be implemented through a normalization function, through a Softmax function, or through a Sparsemax function. The initial historical dialogue keyword query variable and the initial query voice keyword variable can be used as the independent variables of the normalization function respectively, and the function values calculated are used as the initial historical keyword query scaling variable and the initial query keyword query scaling variable.

[0106] Step S505: Calculate the target historical dialogue keyword vector according to the initial historical keyword query scaling variable and the initial historical dialogue keyword content vector.

[0107] In this embodiment, the initial historical keyword query scaling variable and the initial historical dialogue keyword content vector can be multiplied, and the multiplication result is used as the target historical dialogue keyword vector.

[0108] Step S506: Calculate the target query voice keyword vector according to the initial query keyword query scaling variable and the initial query voice keyword content vector.

[0109] In this embodiment, the initial query keyword query scaling variable and the initial query voice keyword content vector can be multiplied, and the multiplication result is used as the target query voice keyword vector.

[0110] The human-computer interaction method based on the AI intelligent pen barrel provided by the embodiment of the present application captures semantic feature information in different logical subspaces through multiple preset feature mapping vectors, so that multiple fields and levels involved in the dialogue content can be fully analyzed, and thus keywords can be extracted for different fields or levels. The historical dialogue content and the current query content of the user in specific fields such as knowledge Q&A or emotional communication are accurately recognized, so as to provide more refined keywords and improve the pertinence and effectiveness of the generated reply information.

[0111] Figure 6 FIG. shows the implementation flowchart of the human-computer interaction method based on the AI intelligent pen barrel provided by the sixth embodiment of the present application, which is different from the first embodiment above: Step S105 specifically includes:

[0112] Step S601: Use the target historical dialogue keyword vector as the main node.

[0113] In this embodiment, the keyword content corresponding to the target historical dialogue keyword vector is used as the main reference information, and then the target historical dialogue keyword vector is used as the main node of the dialogue text feature structure diagram. It can be understood that the dialogue text feature structure diagram is constructed based on the vector space.

[0114] Step S602: Obtain a historical dialogue auxiliary text vector based on the historical dialogue text vector and the target historical dialogue keyword vector.

[0115] In this embodiment, the content in the historical dialogue information other than the content represented by the target historical dialogue keyword vector is used as auxiliary reference information. Then, the auxiliary reference information in the historical dialogue information is transformed into a vector through a word embedding technique and used as the historical dialogue auxiliary text vector.

[0116] Step S603: Use the historical dialogue auxiliary text vector as an auxiliary node.

[0117] In this embodiment, the content in the historical dialogue information other than the content represented by the target historical dialogue keyword vector is used as auxiliary reference information. Then, the auxiliary reference information in the historical dialogue information is transformed into a historical dialogue auxiliary text vector through a word embedding technique. Then, the historical dialogue auxiliary text vector is used as an auxiliary node of the dialogue text feature structure diagram.

[0118] Step S604: Calculate the similarity between each of the main nodes and the similarity between each of the auxiliary nodes.

[0119] In this embodiment, it may be to calculate the Euclidean distance between each of the main nodes and the Euclidean distance between each of the auxiliary nodes, and calculate the similarity between each of the main nodes and the similarity between each of the auxiliary nodes based on the Euclidean distance, so as to quantify the semantic relevance between each of the main nodes and between each of the auxiliary nodes.

[0120] Step S605: Calculate the similarity between each of the main nodes and each of the auxiliary nodes to obtain the similarity between the main and auxiliary nodes.

[0121] In this embodiment, it may be to calculate the Euclidean distance between each of the main nodes and each of the auxiliary nodes, and calculate the similarity between each of the main nodes and each of the auxiliary nodes based on the Euclidean distance, so as to quantify the semantic relevance between each of the main and auxiliary nodes.

[0122] Step S606: Determine whether the similarity between the main nodes is greater than a preset similarity threshold; if so, generate an edge between the main nodes; if not, do not generate an edge between the main nodes.

[0123] In this embodiment, when the similarity between the main nodes is greater than a preset similarity threshold, it indicates that the semantic feature correlation represented by the two main nodes corresponding to this similarity is strong. Then, an edge can be generated between these two main nodes to represent that these two main nodes have a strong correlation. When the similarity between the main nodes is less than or equal to the preset similarity threshold, it indicates that the semantic feature correlation represented by the two main nodes corresponding to this similarity is weak. Then, an edge between these two main nodes may not be generated to represent that these two main nodes have a weak correlation or no correlation.

[0124] Step S607: Determine whether the similarity between the secondary nodes is greater than a preset similarity threshold. If so, generate an edge between the secondary nodes. If not, do not generate an edge between the secondary nodes.

[0125] In this embodiment, when the similarity between the secondary nodes is greater than a preset similarity threshold, it indicates that the semantic feature correlation represented by the two secondary nodes corresponding to this similarity is strong. Then, an edge can be generated between these two secondary nodes to represent that these two secondary nodes have a strong correlation. When the similarity between the secondary nodes is less than or equal to the preset similarity threshold, it indicates that the semantic feature correlation represented by the two secondary nodes corresponding to this similarity is weak. Then, an edge between these two secondary nodes may not be generated to represent that these two secondary nodes have a weak correlation or no correlation.

[0126] Step S608: Determine whether the similarity between the main and secondary nodes is greater than a preset similarity threshold. If so, generate an edge between the main and secondary nodes. If not, do not generate an edge between the main and secondary nodes.

[0127] In this embodiment, when the similarity between the main and secondary nodes is greater than a preset similarity threshold, it indicates that the semantic feature correlation represented by the corresponding main node and secondary node is strong. Then, an edge can be generated between these main and secondary nodes to represent that these two nodes have a strong correlation. When the similarity between the main and secondary nodes is less than or equal to the preset similarity threshold, it indicates that the semantic feature correlation represented by the corresponding main node and secondary node is weak. Then, an edge between these two nodes may not be generated to represent that these two nodes have a weak correlation or no correlation.

[0128] Step S609: Construct a dialogue text feature structure diagram based on the main nodes, secondary nodes, edges between the main nodes, edges between the secondary nodes, and edges between the main and secondary nodes.

[0129] In this embodiment, it can be by combining the main nodes, secondary nodes, edges between the main nodes, edges between the secondary nodes, and edges between the main and secondary nodes in the vector space to generate a dialogue text feature structure diagram.

[0130] The human-computer interaction method based on the AI intelligent pen holder provided by the embodiment of the present application realizes the introduction of historical conversation content for reference by modeling the complex relationship between the historical conversation content and the current question content through constructing a dialogue text feature structure diagram in a multi-dimensional vector space. It is used to capture the semantic relevance among the keyword, historical conversation content, and question content during the dialogue process in subsequent processing, mine the potential semantic information in the historical conversation content and the question content, and combine the mining results in the historical conversation content with the current question content to generate a reply message, enabling the AI intelligence to understand the coherence and pertinence in the dialogue process, enhancing the AI intelligent pen holder's in-depth understanding ability of the dialogue context content, and thus providing a more comprehensive reply for the user.

[0131] Figure 7 The flowchart of the implementation of the human-computer interaction method based on the AI intelligent pen holder provided in the seventh embodiment of the present application is shown. The difference from the sixth embodiment above is that the step S106 specifically includes:

[0132] Step S701: Use the keyword vector of the target question voice text as a new node.

[0133] In this embodiment, the keyword vector of the target question voice text is used as a new node of the dialogue text feature structure diagram.

[0134] Step S702: Randomly insert the new node into the dialogue text feature structure diagram to obtain an updated dialogue text feature structure diagram.

[0135] In this embodiment, the keyword vector of the target question voice text is used as a new node of the dialogue text feature structure diagram and randomly inserted into the dialogue text feature structure diagram to establish a connection between the question content and the historical conversation content, realize real-time information fusion, enable the AI intelligent pen holder to control the dialogue context of the current question content, and the dialogue text feature structure diagram after inserting the new node is used as the updated dialogue text feature structure diagram.

[0136] Step S703: Calculate the similarity between the new node and the main node to obtain the new main node similarity.

[0137] In this embodiment, it can be to calculate the Euclidean distance between the new node and the main node, and thus output the calculated Euclidean distance as the new main node similarity.

[0138] Step S704: Calculate the similarity between the new node and the auxiliary node to obtain the new auxiliary node similarity.

[0139] In this embodiment, it can be to calculate the Euclidean distance between the new node and the auxiliary node, and thus output the calculated Euclidean distance as the new auxiliary node similarity.

[0140] Step S705: Determine whether the similarity of the new main node is greater than a preset similarity threshold; if so, generate an edge between the new node and the main node; if not, do not generate an edge between the new node and the main node.

[0141] In this embodiment, when the similarity of the new main node is greater than the preset similarity threshold, it indicates that the semantic feature correlation between the main node corresponding to this similarity and the new node is strong, so an edge between the new node and the main node can be generated to represent that these two nodes have a strong correlation; when the similarity between the new main nodes is less than or equal to the preset similarity threshold, it indicates that the semantic feature correlation between the main node corresponding to this similarity and the new node is weak, so an edge between these two nodes can be not generated to represent that these two nodes have a weak correlation or no correlation.

[0142] Step S706: Determine whether the similarity of the new secondary node is greater than a preset similarity threshold; if so, generate an edge between the new node and the secondary node; if not, do not generate an edge between the new node and the secondary node.

[0143] In this embodiment, when the similarity of the new secondary node is greater than the preset similarity threshold, it indicates that the semantic feature correlation between the secondary node corresponding to this similarity and the new node is strong, so an edge between the new node and the secondary node can be generated to represent that these two nodes have a strong correlation; when the similarity between the new secondary nodes is less than or equal to the preset similarity threshold, it indicates that the semantic feature correlation between the secondary node corresponding to this similarity and the new node is weak, so an edge between these two nodes can be not generated to represent that these two nodes have a weak correlation or no correlation.

[0144] Step S707: Generate a reply text message and a reply voice message according to the new node, the main node, the secondary node, the edges between the main nodes, the edges between the secondary nodes, the edges between the main and secondary nodes, the edges between the new node and the main node, and the edges between the new node and the secondary node.

[0145] In this embodiment, different weight coefficients may be set for the new node, the main node, and the auxiliary node first. The connection paths between the new node, the main node, and the auxiliary node in the updated dialogue text feature structure diagram may be determined based on the edges between the main nodes, the edges between the auxiliary nodes, the edges between the main and auxiliary nodes, the edges between the new node and the main node, and the edges between the new node and the auxiliary node. The new nodes, the main nodes, and the auxiliary nodes on the multiple connection paths connected by the edges are weighted and summed, and the result obtained is used as the reply text vector. The reply text vector is then decoded by the language model to generate a reply text message, and then the reply text message is converted into a reply voice message by the speech synthesis technology. Among them, the reply text message can be visualized and output through the electronic screen of the AI ​​smart pen holder, and the reply voice message can be output through the speaker of the AI ​​smart pen holder.

[0146] The human-computer interaction method based on the AI ​​smart pen holder provided in the embodiment of the present application can build a complete semantic association network. Since each question asked by a user is new information, after the new information is added as a new node, it can be connected with the historical conversation content through the semantic association network to form a continuously expanding conversation text feature structure diagram, so that the AI ​​smart pen holder can understand the conversation context more comprehensively. When facing questions from different users and different types, new nodes can continuously enrich the semantic association network, so that the AI ​​smart pen holder can better adapt to diverse conversation needs and provide users with a high-quality interactive experience.

[0147] Figure 8 The flowchart of the implementation of the human-computer interaction method based on the AI ​​smart pen holder provided in the eighth embodiment of the present application is shown. The difference between the eighth embodiment and the seventh embodiment is that the step S707 specifically includes:

[0148] Step S801, counting the number of new nodes, the number of primary nodes, and the number of secondary nodes.

[0149] In this embodiment, the number of new nodes, the number of main nodes, and the number of auxiliary nodes in the updated dialogue text feature structure graph are counted for subsequent determination of the dimension of the adjacency variable matrix.

[0150] Step S802: Calculate the total number of nodes according to the number of new nodes, the number of primary nodes, and the number of secondary nodes.

[0151] In this embodiment, the number of new nodes, the number of primary nodes, and the number of secondary nodes are summed, and the sum result is used as the total number of nodes.

[0152] Step S803: Generate an adjacency variable matrix based on the total number of nodes, the edges between main nodes, the edges between auxiliary nodes, the edges between main and auxiliary nodes, the new nodes, the edges between new nodes and main nodes, and the edges between new nodes and auxiliary nodes.

[0153] In this embodiment, an empty matrix can be set first, with the total number of nodes as the dimension of the matrix. Then, traverse the entire updated dialogue text feature structure diagram. When traversing the edges between main nodes, the edges between auxiliary nodes, the edges between main and auxiliary nodes, the new nodes, the edges between new nodes and main nodes, and the edges between new nodes and auxiliary nodes, set the elements corresponding to the corresponding rows and columns in the empty matrix to 1; if not traversing these edges, set the elements corresponding to the corresponding rows and columns in the empty matrix to 0. The finally generated matrix is output as the adjacency variable matrix.

[0154] Step S804: Generate reply text information and reply voice information based on the adjacency variable matrix, main nodes, auxiliary nodes, and new nodes.

[0155] In this embodiment, the main nodes, auxiliary nodes, and new nodes can be constructed into a node information matrix, and the elements of the node information matrix are the main nodes, auxiliary nodes, and new nodes. The node information matrix can be multiplied by the adjacency variable matrix to obtain a reply text vector. Then, the reply text vector is decoded through the language model to generate reply text information, and then the reply text information is converted into reply voice information through speech synthesis technology. Among them, the reply text information can be visually output through the electronic screen of the AI intelligent pen holder, and the reply voice information can be output through the speaker of the AI intelligent pen holder.

[0156] The human-computer interaction method based on the AI intelligent pen holder provided by the embodiments of the present application can clearly present the association structure between each information node in the dialogue by constructing an adjacency variable matrix. The main nodes represent key topics, the auxiliary nodes are relevant details, and the new nodes are the current questions. Through the adjacency variable matrix, these associations can be accurately expressed in a mathematical form, enabling the AI intelligent pen holder to understand the mutual relationship of information, thereby understanding the connection between the user's different questions and the topics in the historical dialogue information, and quickly locating the relevant information in the historical dialogue information, so that the reply information generated by the AI intelligent pen holder is closer to the context, improving the user experience.

[0157] Figure 9 The flowchart of the implementation of the human-computer interaction method based on the AI intelligent pen holder provided in Embodiment IX of the present application is shown. The difference from Embodiment VIII above is that in step S804, it specifically includes:

[0158] Step S901: Generate a master node feature matrix, a slave node feature matrix, and a new node feature matrix based on the master node, the slave node, and the new node.

[0159] In this embodiment, it may be to extract all the master nodes in the updated dialogue text feature structure diagram, determine the number of rows and columns in the matrix through the position information of each master node in the updated dialogue text feature structure diagram, and generate the elements in the matrix through the vector values of each master node, so as to generate the master node feature matrix. It may be to extract all the slave nodes in the updated dialogue text feature structure diagram, determine the number of rows and columns in the matrix through the position information of each slave node in the updated dialogue text feature structure diagram, and generate the elements in the matrix through the vector values of each slave node, so as to generate the slave node feature matrix. It may be to extract all the new nodes in the updated dialogue text feature structure diagram, determine the number of rows and columns in the matrix through the position information of each new node in the updated dialogue text feature structure diagram, and generate the elements in the matrix through the vector values of each new node, so as to generate the new node feature matrix.

[0160] Step S902: Concatenate the master node feature matrix, the slave node feature matrix, and the new node feature matrix to generate a node feature matrix.

[0161] In this embodiment, it may be to concatenate the master node feature matrix, the slave node feature matrix, and the new node feature matrix into a matrix, that is, the node feature matrix.

[0162] Step S903: Calculate the reply text vector according to the adjacency variable matrix and the node feature matrix.

[0163] In this embodiment, it may be to perform a dot product calculation on the adjacency variable matrix and the node feature matrix, and output the calculation result as the reply text vector.

[0164] Step S904: Generate reply text information and reply voice information according to the reply text vector.

[0165] In this embodiment, the reply text vector can be used through the decoding technology of the language model to generate the reply text information, and then the reply text information can be converted into the reply voice information through the speech synthesis technology. Among them, the reply text information can be visually output through the electronic screen of the AI intelligent pen holder, and the reply voice information can be output through the speaker of the AI intelligent pen holder.

[0166] The human-computer interaction method based on an AI intelligent pen holder provided by an embodiment of the present application quantifies key information in a conversation by generating a main node feature matrix, an auxiliary node feature matrix, and a new node feature matrix, highlights the key attributes of the core content through the main node feature matrix, provides content supplementary details through the auxiliary node feature matrix, and clarifies the latest concerns of the current user through the new node feature matrix, thereby improving the efficiency and accuracy of information processing for the conversation content, making the generated reply information more in line with user needs, and enhancing the fluency and efficiency of human-computer interaction.

[0167] Corresponding to the method in the above embodiment, Figure 10 The structural block diagram of the human-computer interaction device based on an AI intelligent pen holder provided by an embodiment of the present application is shown. For the sake of convenience of description, only the parts related to the embodiment of the present application are shown. Figure 10 The exemplary human-computer interaction device based on an AI intelligent pen holder may be the execution subject of the human-computer interaction method based on an AI intelligent pen holder provided in the foregoing Embodiment 1.

[0168] Referring to Figure 10 , the human-computer interaction device based on an AI intelligent pen holder includes:

[0169] An information acquisition module 1010, configured to acquire historical conversation information and the voice information of the question input by the user;

[0170] A text vector generation module 1020, configured to perform conversion processing on the historical conversation information and the voice information of the question to generate a plurality of historical conversation text vectors and a plurality of voice text vectors of the question;

[0171] An initial keyword vector generation module 1030, configured to generate an initial historical conversation keyword vector and an initial voice keyword vector of the question according to a plurality of the historical conversation text vectors, a plurality of the voice text vectors of the question, a preset adjacent area radius, and preset adjacent area vector quantity information;

[0172] A target keyword vector generation module 1040, configured to generate a target historical conversation keyword vector and a target voice keyword vector of the question according to the initial historical conversation keyword vector, the initial voice keyword vector of the question, and a plurality of preset feature mapping vectors;

[0173] A conversation text feature structure diagram construction module 1050, configured to construct a conversation text feature structure diagram according to the historical conversation text vectors and the target historical conversation keyword vector; and

[0174] A reply information generation module 1060, configured to generate a reply text information and a reply voice information according to the conversation text feature structure diagram and the target voice text keyword vector of the question.

[0175] For the process of each module in the human-computer interaction device based on the AI intelligent pen holder provided in the embodiments of the present application to implement its respective functions, reference may specifically be made to the description of Embodiment 1 as described above, and details are not repeated here. Figure 1 Shown in the above description of Embodiment 1, details are not repeated here.

[0176] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not imply the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0177] It should be understood that when used in the specification and the appended claims of the present application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations.

[0178] It should also be understood that the term "and / or" used in the specification and the appended claims of the present application refers to any combination and all possible combinations of one or more of the related listed items, and includes these combinations.

[0179] As used in the specification and the appended claims of the present application, the term "if" may be interpreted as "when", "once", "in response to determining", or "in response to detecting" depending on the context. Similarly, the phrase "if determined" or "if detecting [the described condition or event]" may be interpreted as meaning "once determined", "in response to determining", "once detecting [the described condition or event]", or "in response to detecting [the described condition or event]" depending on the context.

[0180] In addition, in the description of the specification and the appended claims of the present application, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and should not be construed as indicating or implying relative importance. It should also be understood that although the terms "first", "second", etc. are used in the text in some embodiments of the present application to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, the first table may be named the second table, and similarly, the second table may be named the first table, without departing from the scope of the various described embodiments. The first table and the second table are both tables, but they are not the same table.

[0181] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that specific features, structures, or characteristics described in connection with that embodiment are included in one or more embodiments of this application. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification do not necessarily all refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in another way. The terms "comprising", "including", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in another way.

[0182] The human-computer interaction method based on the AI intelligent pen holder provided in the embodiment of this application can be applied to the AI intelligent pen holder.

[0183] The AI intelligent pen holder of this embodiment includes: at least one processor and a memory, and a computer program that can run on the processor is stored in the memory. When the processor executes the computer program, the steps in each of the above-mentioned embodiments of the human-computer interaction method based on the AI intelligent pen holder are implemented, for example Figure 1 The steps S101 to S106 shown. Or, when the processor executes the computer program, the functions of each module / unit in each of the above-mentioned device embodiments are implemented, for example Figure 10 The functions of the modules 1010 to 1060 shown.

[0184] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0185] In some embodiments, the memory may be the internal storage unit of the AI intelligent pen holder, such as the hard disk or memory of the AI intelligent pen holder. The AI intelligent pen holder may also be an external storage device of the AI intelligent pen holder, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the AI intelligent pen holder. Further, the memory may also include both the internal storage unit of the AI intelligent pen holder and external storage devices. The memory is used to store an operating system, application programs, a BootLoader, data, and other programs, such as the program code of the computer program. The memory may also be used to temporarily store data that has been sent or will be sent.

[0186] In addition, in each embodiment of the present application, each functional unit may be integrated in a processing unit, may exist separately physically for each unit, or two or more units may be integrated in one unit. The above integrated unit may be implemented in the form of hardware or in the form of a software functional unit.

[0187] An embodiment of the present application also provides an AI intelligent pen holder, which includes at least one memory, at least one processor, and a computer program stored in the at least one memory and executable on the at least one processor. When the processor executes the computer program, the AI intelligent pen holder implements the steps in any of the above method embodiments.

[0188] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it can implement the steps in each of the above method embodiments.

[0189] An embodiment of the present application provides a computer program product. When the computer program product runs on an AI intelligent pen holder, it enables the AI intelligent pen holder to implement the steps in each of the above method embodiments when executed.

[0190] When the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of this application, it can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, as well as software distribution media, etc.

[0191] In the above embodiments, the descriptions of the various embodiments have their own focuses. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0192] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0193] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0194] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them; although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of this application, and should all be included in the protection scope of this application.

Claims

1. A human-computer interaction method based on AI smart pen holder, characterized in that: include: Obtain historical conversation information and voice information of questions input by users; Converting the historical conversation information and the question voice information to generate a plurality of historical conversation text vectors and a plurality of question voice text vectors; Generate an initial historical conversation keyword vector and an initial question voice keyword vector according to the plurality of historical conversation text vectors, the plurality of question voice text vectors, a preset adjacent area radius, and preset adjacent area vector quantity information; Generate a target historical conversation keyword vector and a target question voice keyword vector according to the initial historical conversation keyword vector, the initial question voice keyword vector and a plurality of preset feature mapping vectors; Constructing a conversation text feature structure diagram according to the historical conversation text vector and the target historical conversation keyword vector; Generate reply text information and reply voice information according to the dialogue text feature structure diagram and the target question voice text keyword vector; The step of generating an initial historical conversation keyword vector and an initial question voice keyword vector according to the plurality of historical conversation text vectors, the plurality of question voice text vectors, a preset adjacent area radius, and preset adjacent area vector quantity information specifically includes: Generate a human-computer interaction text vector set according to the plurality of historical conversation text vectors and the plurality of question voice text vectors; the human-computer interaction text vector set includes a plurality of human-computer interaction text vectors; Randomly extracting elements of the human-computer interaction text vector set to obtain a human-computer interaction text center vector; Calculating the Euclidean distance between each of the human-computer interaction text vectors and the human-computer interaction text center vector to obtain the interaction text vector distance; When the interactive text vector distance is less than or equal to a preset adjacent area radius, the human-computer interactive text vector corresponding to the interactive text vector distance is used as the human-computer interactive text adjacent area vector; Generate an initial historical dialogue keyword vector and an initial question voice keyword vector according to the human-computer interaction text adjacent region vector and preset adjacent region vector quantity information; The step of generating an initial historical dialogue keyword vector and an initial question voice keyword vector according to the human-computer interaction text adjacent region vector and preset adjacent region vector quantity information specifically includes: Counting the number of vectors in adjacent regions of the human-computer interaction text to obtain information on the number of vectors in the neighborhood of the interaction text; Determine whether the interactive text neighborhood vector quantity information is greater than or equal to the preset adjacent area vector quantity information; If yes, then determining the center vector of the human-computer interaction text as the core vector of the human-computer interaction text; If not, determining the central vector of the human-computer interaction text as a non-core vector of the human-computer interaction text; Counting the number of the human-computer interaction text core vectors and the number of the human-computer interaction text non-core vectors to obtain the number of interactive text core vectors and the number of interactive text non-core vectors; Determine whether the sum of the number of interactive text core vectors and the number of interactive text non-core vectors is equal to the number of elements in the human-computer interactive text vector set; If yes, generating an initial historical dialogue keyword vector and an initial question voice keyword vector according to the human-computer interaction text core vector and the human-computer interaction text non-core vector; If not, obtaining a remaining interactive text vector set according to the human-computer interaction text vector set, the human-computer interaction text core vector, and the human-computer interaction text non-core vector; The remaining interactive text vector set is used as a human-computer interaction text vector set, and the step of randomly extracting elements of the human-computer interaction text vector set to obtain a human-computer interaction text center vector is returned; The step of generating an initial historical dialogue keyword vector and an initial question voice keyword vector according to the human-computer interaction text core vector and the human-computer interaction text non-core vector specifically includes: Calculating the Euclidean distance between each non-core vector of the human-computer interaction text and each core vector of the human-computer interaction text to obtain a confirmation distance of a non-core point of the interaction text; Determine whether the confirmation distance of the non-core point of the interactive text is less than or equal to a preset adjacent area radius; If yes, the human-computer interaction text non-core vector corresponding to the interaction text non-core point confirmation distance is used as the human-computer interaction text edge point vector; If not, the non-core vector of the human-computer interaction text corresponding to the non-core point confirmation distance of the interaction text is used as a redundant point vector of the human-computer interaction text; Generate multiple interactive text keyword vectors according to the human-computer interactive text core vector and the human-computer interactive text edge point vector; Performing numerical separation and extraction processing on the interactive text keyword vector to generate an initial historical conversation keyword vector and an initial question voice keyword vector; The plurality of preset feature mapping vectors include a preset search mapping vector, a preset index mapping vector and a preset content mapping vector; The step of generating a target historical conversation keyword vector and a target question voice keyword vector according to the initial historical conversation keyword vector, the initial question voice keyword vector and a plurality of preset feature mapping vectors specifically includes: Calculate the initial historical conversation keyword retrieval vector, the initial historical conversation keyword index vector, the initial historical conversation keyword content vector, the initial question voice keyword retrieval vector, the initial question voice keyword index vector, and the initial question voice keyword content vector according to the initial historical conversation keyword vector, the initial question voice keyword vector, the preset retrieval mapping vector, the preset index mapping vector, and the preset content mapping vector; Calculating an initial historical conversation keyword query variable according to the initial historical conversation keyword retrieval vector and the initial historical conversation keyword index vector; Calculating an initial question voice keyword query variable according to the initial question voice keyword retrieval vector and the initial question voice keyword index vector; Standardizing the initial historical conversation keyword query variable and the initial question voice keyword query variable to obtain an initial historical keyword query scaling variable and an initial question keyword query scaling variable; Calculating a target historical conversation keyword vector according to the initial historical keyword query scaling variable and the initial historical conversation keyword content vector; Calculate a target question voice keyword vector according to the initial question keyword query scaling variable and the initial question voice keyword content vector; Performing a dot product operation on the initial historical conversation keyword vector and a preset search mapping vector to obtain an initial historical conversation keyword search vector; Performing a dot product operation on the initial historical conversation keyword vector and a preset index mapping vector to obtain an initial historical conversation keyword index vector; Performing a dot product operation on the initial historical conversation keyword vector and a preset content mapping vector to obtain an initial historical conversation keyword content vector; Performing a dot product operation on the initial question voice keyword vector and a preset search mapping vector to obtain an initial question voice keyword search vector; Performing a dot product operation on the initial question voice keyword vector and a preset index mapping vector to obtain an initial question voice keyword index vector; Performing a dot product operation on the initial question voice keyword vector and a preset content mapping vector to obtain an initial question voice keyword content vector; Performing a dot product operation on the initial historical conversation keyword search vector and the transposed matrix of the initial historical conversation keyword index vector, and using the calculation result as the initial historical conversation keyword query variable; Performing a dot product operation on the initial question voice keyword search vector and the transposed matrix of the initial question voice keyword index vector, and using the calculation result as the initial question voice keyword query variable; Multiplying the initial historical keyword query scaling variable and the initial historical conversation keyword content vector, and using the multiplication result as the target historical conversation keyword vector; The initial question keyword query scaling variable and the initial question voice keyword content vector are multiplied, and the multiplication result is used as the target question voice keyword vector.

2. The human-computer interaction method based on the AI ​​smart pen holder according to claim 1, characterized in that: The step of constructing a dialogue text feature structure diagram according to the historical dialogue text vector and the target historical dialogue keyword vector specifically includes: Taking the target historical conversation keyword vector as the main node; Obtaining a historical conversation auxiliary text vector according to the historical conversation text vector and the target historical conversation keyword vector; Using the historical conversation auxiliary text vector as an auxiliary node; Calculating the similarity between each of the primary nodes and the similarity between each of the secondary nodes; Calculate the similarity between each of the primary nodes and each of the secondary nodes to obtain the similarity between the primary and secondary nodes; When the similarity between the master nodes is greater than a preset similarity threshold, edges between the master nodes are generated; When the similarity between the auxiliary nodes is greater than a preset similarity threshold, edges between the auxiliary nodes are generated; When the similarity between the primary and secondary nodes is greater than a preset similarity threshold, an edge between the primary and secondary nodes is generated; A conversation text feature structure graph is constructed based on the main nodes, auxiliary nodes, edges between main nodes, edges between auxiliary nodes, and edges between main and auxiliary nodes.

3. The human-computer interaction method based on the AI ​​smart pen holder as claimed in claim 2, characterized in that: The step of generating reply text information and reply voice information according to the dialogue text feature structure diagram and the target question voice text keyword vector specifically includes: Taking the target question speech text keyword vector as a new node; Randomly inserting the new node into the dialogue text feature structure graph to obtain an updated dialogue text feature structure graph; Calculate the similarity between the new node and the main node to obtain the new main node similarity; Calculating the similarity between the new node and the auxiliary node to obtain the new auxiliary node similarity; When the similarity of the new master node is greater than a preset similarity threshold, an edge between the new node and the master node is generated; When the similarity of the new auxiliary node is greater than a preset similarity threshold, an edge between the new node and the auxiliary node is generated; Reply text information and reply voice information are generated according to the new node, the main node, the auxiliary node, the edge between the main nodes, the edge between the auxiliary nodes, the edge between the main and auxiliary nodes, the edge between the new node and the main node, and the edge between the new node and the auxiliary node.

4. The human-computer interaction method based on the AI ​​smart pen holder as claimed in claim 3, characterized in that: The step of generating a reply text message and a reply voice message according to the new node, the main node, the auxiliary node, the edge between the main node, the edge between the auxiliary nodes, the edge between the main and auxiliary nodes, the edge between the new node and the main node, and the edge between the new node and the auxiliary node specifically includes: Counting the number of new nodes, the number of primary nodes, and the number of secondary nodes; Calculate the total number of nodes according to the number of new nodes, the number of primary nodes, and the number of secondary nodes; Generate an adjacency variable matrix according to the total number of nodes, edges between main nodes, edges between auxiliary nodes, edges between main and auxiliary nodes, edges between new nodes and main nodes, and edges between new nodes and auxiliary nodes; Reply text information and reply voice information are generated according to the adjacency variable matrix, the main node, the auxiliary node and the new node.

5. The human-computer interaction method based on the AI ​​smart pen holder as claimed in claim 4, characterized in that: The step of generating a reply text message and a reply voice message according to the adjacency variable matrix, the main node, the auxiliary node and the new node specifically includes: Generate a main node feature matrix, an auxiliary node feature matrix and a new node feature matrix according to the main node, the auxiliary node and the new node; The main node feature matrix, the auxiliary node feature matrix and the new node feature matrix are concatenated to generate a node feature matrix; Calculate the reply text vector according to the adjacency variable matrix and the node feature matrix; Generate reply text information and reply voice information according to the reply text vector.

6. A human-computer interaction device based on AI smart pen holder, characterized in that: include: The information acquisition module is used to obtain historical conversation information and voice information of questions input by users; A text vector generation module, used to convert the historical conversation information and question voice information to generate a plurality of historical conversation text vectors and a plurality of question voice text vectors; An initial keyword vector generation module, used to generate an initial historical conversation keyword vector and an initial question voice keyword vector according to the plurality of historical conversation text vectors, the plurality of question voice text vectors, a preset adjacent area radius, and preset adjacent area vector quantity information; A target keyword vector generation module, used to generate a target historical conversation keyword vector and a target question voice keyword vector according to the initial historical conversation keyword vector, the initial question voice keyword vector and a plurality of preset feature mapping vectors; A dialogue text feature structure graph construction module, used to construct a dialogue text feature structure graph according to the historical dialogue text vector and the target historical dialogue keyword vector; as well as A reply information generation module, used to generate reply text information and reply voice information according to the dialogue text feature structure diagram and the target question voice text keyword vector; The step of generating an initial historical conversation keyword vector and an initial question voice keyword vector according to the plurality of historical conversation text vectors, the plurality of question voice text vectors, a preset adjacent area radius, and preset adjacent area vector quantity information specifically includes: Generate a human-computer interaction text vector set according to the plurality of historical conversation text vectors and the plurality of question voice text vectors; the human-computer interaction text vector set includes a plurality of human-computer interaction text vectors; Randomly extracting elements of the human-computer interaction text vector set to obtain a human-computer interaction text center vector; Calculating the Euclidean distance between each of the human-computer interaction text vectors and the human-computer interaction text center vector to obtain the interaction text vector distance; When the interactive text vector distance is less than or equal to a preset adjacent area radius, the human-computer interactive text vector corresponding to the interactive text vector distance is used as the human-computer interactive text adjacent area vector; Generate an initial historical dialogue keyword vector and an initial question voice keyword vector according to the human-computer interaction text adjacent region vector and preset adjacent region vector quantity information; The step of generating an initial historical dialogue keyword vector and an initial question voice keyword vector according to the human-computer interaction text adjacent region vector and preset adjacent region vector quantity information specifically includes: Counting the number of vectors in adjacent regions of the human-computer interaction text to obtain information on the number of vectors in the neighborhood of the interaction text; Determine whether the interactive text neighborhood vector quantity information is greater than or equal to the preset adjacent area vector quantity information; If yes, then determining the center vector of the human-computer interaction text as the core vector of the human-computer interaction text; If not, determining the central vector of the human-computer interaction text as a non-core vector of the human-computer interaction text; Counting the number of the human-computer interaction text core vectors and the number of the human-computer interaction text non-core vectors to obtain the number of interactive text core vectors and the number of interactive text non-core vectors; Determine whether the sum of the number of interactive text core vectors and the number of interactive text non-core vectors is equal to the number of elements in the human-computer interactive text vector set; If yes, generating an initial historical dialogue keyword vector and an initial question voice keyword vector according to the human-computer interaction text core vector and the human-computer interaction text non-core vector; If not, obtaining a remaining interactive text vector set according to the human-computer interaction text vector set, the human-computer interaction text core vector, and the human-computer interaction text non-core vector; The remaining interactive text vector set is used as a human-computer interaction text vector set, and the step of randomly extracting elements of the human-computer interaction text vector set to obtain a human-computer interaction text center vector is returned; The step of generating an initial historical dialogue keyword vector and an initial question voice keyword vector according to the human-computer interaction text core vector and the human-computer interaction text non-core vector specifically includes: Calculating the Euclidean distance between each non-core vector of the human-computer interaction text and each core vector of the human-computer interaction text to obtain a confirmation distance of a non-core point of the interaction text; Determine whether the confirmation distance of the non-core point of the interactive text is less than or equal to a preset adjacent area radius; If yes, the human-computer interaction text non-core vector corresponding to the interaction text non-core point confirmation distance is used as the human-computer interaction text edge point vector; If not, the non-core vector of the human-computer interaction text corresponding to the non-core point confirmation distance of the interaction text is used as a redundant point vector of the human-computer interaction text; Generate multiple interactive text keyword vectors according to the human-computer interactive text core vector and the human-computer interactive text edge point vector; Performing numerical separation and extraction processing on the interactive text keyword vector to generate an initial historical conversation keyword vector and an initial question voice keyword vector; The plurality of preset feature mapping vectors include a preset search mapping vector, a preset index mapping vector and a preset content mapping vector; The step of generating a target historical conversation keyword vector and a target question voice keyword vector according to the initial historical conversation keyword vector, the initial question voice keyword vector and a plurality of preset feature mapping vectors specifically includes: Calculate the initial historical conversation keyword retrieval vector, the initial historical conversation keyword index vector, the initial historical conversation keyword content vector, the initial question voice keyword retrieval vector, the initial question voice keyword index vector, and the initial question voice keyword content vector according to the initial historical conversation keyword vector, the initial question voice keyword vector, the preset retrieval mapping vector, the preset index mapping vector, and the preset content mapping vector; Calculating an initial historical conversation keyword query variable according to the initial historical conversation keyword retrieval vector and the initial historical conversation keyword index vector; Calculating an initial question voice keyword query variable according to the initial question voice keyword retrieval vector and the initial question voice keyword index vector; Standardizing the initial historical conversation keyword query variable and the initial question voice keyword query variable to obtain an initial historical keyword query scaling variable and an initial question keyword query scaling variable; Calculating a target historical conversation keyword vector according to the initial historical keyword query scaling variable and the initial historical conversation keyword content vector; Calculate a target question voice keyword vector according to the initial question keyword query scaling variable and the initial question voice keyword content vector; Performing a dot product operation on the initial historical conversation keyword vector and a preset search mapping vector to obtain an initial historical conversation keyword search vector; Performing a dot product operation on the initial historical conversation keyword vector and a preset index mapping vector to obtain an initial historical conversation keyword index vector; Performing a dot product operation on the initial historical conversation keyword vector and a preset content mapping vector to obtain an initial historical conversation keyword content vector; Performing a dot product operation on the initial question voice keyword vector and a preset search mapping vector to obtain an initial question voice keyword search vector; Performing a dot product operation on the initial question voice keyword vector and a preset index mapping vector to obtain an initial question voice keyword index vector; Performing a dot product operation on the initial question voice keyword vector and a preset content mapping vector to obtain an initial question voice keyword content vector; Performing a dot product operation on the initial historical conversation keyword search vector and the transposed matrix of the initial historical conversation keyword index vector, and using the calculation result as the initial historical conversation keyword query variable; Performing a dot product operation on the initial question voice keyword search vector and the transposed matrix of the initial question voice keyword index vector, and using the calculation result as the initial question voice keyword query variable; Multiplying the initial historical keyword query scaling variable and the initial historical conversation keyword content vector, and using the multiplication result as the target historical conversation keyword vector; The initial question keyword query scaling variable and the initial question voice keyword content vector are multiplied, and the multiplication result is used as the target question voice keyword vector.

Images