Reading assistant processing method and device, all-in-one machine and storage medium

Abstract

The invention provides a reading assistant processing method and device, an all-in-one machine and a storage medium, and relates to the technical field of intelligent reading, and the method does not depend on a preset script, and allows a user to initiate a conversation at any time in any reading stage, so that more natural and smooth user interaction experience is realized. Moreover, the general large language model adopted in the method does not need to depend on sample data in a specific field for pre-training, so that the collection cost of the sample data and the model training cost through the sample data can be reduced, and meanwhile, the limitation that a traditional model depends on the data in the specific field is broken through; and a more flexible and wide application scene is realized. Besides, the corresponding reading prompt information is determined by determining the demand information of the user, the prompt engineering technology is effectively used for stimulating and guiding the general large language model to generate more accurate and targeted reply content to be returned to the user, and the method has higher universality and adaptability.

Description

Technical Field

[0001] The present invention relates to the field of intelligent reading technology, and in particular to a reading assistant processing method, device, all-in-one machine and storage medium. Background Art

[0002] In the current education system, the new curriculum emphasizes the importance of reading ability, requiring students to not only be able to read, but also to develop thinking and comprehension abilities through reading. With the development of science and technology, modern technologies such as artificial intelligence (AI) can be used to assist students in the reading process, stimulate their interest in reading, and improve their reading comprehension abilities.

[0003] However, existing reading assistant product solutions usually adopt a semi-guided and semi-open dialogue mechanism based on preset scripts. User queries and feedback are often limited to preset processes and preset scripts, which cannot achieve true open dialogue and restore the user's offline learning scenario with high flexibility. Moreover, the traditional model used needs to rely on sample data in a specific field, which not only increases the cost of collecting sample data, but also increases the cost of model training through sample data. Summary of the invention

[0004] The present invention provides a reading assistant processing method, device, all-in-one machine and storage medium, which are used to solve the defects existing in the related technology.

[0005] The present invention provides a reading assistant processing method, comprising: Determine the reading content and demand information of users in the context of reading target reading materials; Determining reading prompt information corresponding to the demand information, wherein the reading prompt information includes an instruction for generating reply content; Generate a reading input text based on the reading prompt information and the reading content; Based on a general large language model, the application reads the input text and interacts with the user.

[0006] According to a reading assistant processing method provided by the present invention, the reading content is determined by the word selection operation of the user, and the demand information is determined by the demand selection operation of the user, or by the voice information or text information input by the user; Alternatively, the reading content is the full text content of the target reading material, and the demand information is determined by the preset function selection operation of the user.

[0007] A reading assistant processing method provided by the present invention also includes: receiving a reading role selected by the user for the target reading; Based on the universal large language model, the reader role is applied to interact with the user.

[0008] According to a reading assistant processing method provided by the present invention, the receiving of the reading role selected by the user for the target reading material further comprises: The preset background music is played in a loop, and the background of the character interaction interface that interacts with the user is replaced with the preset background superimposed with the reading character.

[0009] According to a reading assistant processing method provided by the present invention, the method of applying the reading role based on the universal large language model to interact with the user includes: receiving an interactive question from the user; Based on the universal large language model, the reader role is applied to answer the interactive question to obtain an interactive answer; Based on the text-to-speech technology, the interactive answer is played through the voice corresponding to the reading role.

[0010] A reading assistant processing method provided by the present invention also includes: Obtaining the user's question information for each reading material; The question information is classified based on a general large language model.

[0011] A reading assistant processing method provided by the present invention also includes: Receiving the reading companion question of the user, and the picture uploaded by the user or the name of the reading material input by the user; Based on a general large language model, the accompanying reading questions and the names of the pictures or readings are applied to interact with the user.

[0012] The present invention also provides a reading assistant processing device, comprising: The demand determination module is used to determine the reading content and demand information of the user in the scenario of reading the target reading material; A prompt determination module, used to determine reading prompt information corresponding to the demand information, wherein the reading prompt information includes an instruction for generating reply content; A text determination module, used for generating a reading input text based on the reading prompt information and the reading content; An interaction module is used for interacting with the user by applying the reading input text based on a general large language model.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the reading assistant processing method described in any one of the above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the reading assistant processing method described in any one of the above is implemented.

[0015] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the reading assistant processing method described in any one of the above is implemented.

[0016] The reading assistant processing method, device, all-in-one machine and storage medium provided by the present invention do not rely on preset scripts, allowing users to initiate conversations at any time at any reading stage, thereby achieving a more natural and smooth user interaction experience. Moreover, the general large-scale language model used in the method does not need to rely on sample data in a specific field for pre-training, which not only reduces the cost of collecting sample data and the cost of model training through sample data, but also breaks through the limitation of traditional models relying on data in specific fields, and realizes more flexible and extensive application scenarios. In addition, by determining the corresponding reading prompt information by determining the user's demand information, prompt engineering technology is effectively used to stimulate and guide the general large-scale language model to generate more accurate and targeted reply content to return to the user, which has stronger versatility and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a flow chart of the reading assistant processing method provided by the present invention.

[0019] Figure 2 It is a structural schematic diagram of the reading assistant processing device provided by the present invention.

[0020] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Figure 1 FIG. 1 is a flow chart of a reading assistant processing method provided in an embodiment of the present invention. Figure 1 As shown, the method includes: S1, determine the reading content and demand information of the user in the reading target reading scenario; S2, determining reading prompt information corresponding to the demand information, wherein the reading prompt information includes an instruction for generating reply content; S3, generating a reading input text based on the reading prompt information and the reading content; S4, based on a general large language model, applying the reading input text to interact with the user.

[0023] Specifically, the reading assistant processing method provided in the embodiment of the present invention is executed by a reading assistant application (APP). The application can be installed in a smart terminal device. The smart terminal device can be a smart phone, a tablet computer, a computer, etc., or it can be a third-party device dedicated to running the application, that is, an all-in-one machine, which is not specifically limited here.

[0024] First, step S1 is performed to determine the reading content and demand information of the user in the reading target reading scene. Here, the user can be a student who needs a reading assistant application to assist in reading at each stage. The user can open the reading assistant application, and the homepage of the reading assistant application can display functions such as book list recommendation, book sea travel, reading report, and smart reading companion.

[0025] By clicking the book list recommendation function button, the user can enter the book list recommendation interface, which can include a must-read book list and an extended book list. By clicking the target book, the user can enter the scene of reading the target book.

[0026] In the scenario of reading a target reading material, the reading content may be content selected by the user through a mouse or touch screen, or may be the full text of the target reading material.

[0027] Furthermore, the demand information may be questions that users have about the reading content, and may include demands for appreciation, interpretation, full-text perception, full-text reading skills, and the like of the reading content.

[0028] Thereafter, step S2 is executed to determine the reading prompt information (prompt) corresponding to the demand information. Different demand information may correspond to different reading prompt information. For example, for appreciation, the reading prompt information may be "Please appreciate this paragraph", for full text perception, the reading prompt information may be "Please perceive this reading material as a whole", for full text reading skills, the reading prompt information may be "Teach me some reading skills", etc.

[0029] It is understandable that the reading prompt information includes instructions for instructing a general large language model (LLM) to generate reply content in the form of natural language text. The general large language model refers to a large language model for general domain pairs, not a large language model obtained by fine-tuning sample data in a specific domain.

[0030] Thereafter, step S3 is executed to obtain the reading input text by concatenating the reading prompt information and the reading content, or to obtain the reading input text by filling the reading prompt information and the reading content into a pre-set input text template, which is not specifically limited in the embodiment of the present invention.

[0031] Finally, step S4 is executed, using the universal large language model, applying the reading input text, and interacting with the user. Here, the reading input text can be input into the universal large language model, and the universal large language model obtains the reply content. If the demand information is appreciation, the reply content can be the detailed appreciation content of the reading content; if the demand information is interpretation, the reply content can be the meaning explanation of the reading content; if the demand information is the full-text perception demand, the reply content can be the writing background, author profile, main content summary, etc. of the target reading; if the demand information is the full-text reading skills, the reply content can be the reading strategy of the target reading.

[0032] By reading the prompt information and the reading content, the reading input text is generated as the input of the general large language model, thereby realizing the generation of reply content based on the general large language model, thereby realizing automatic reply content generation by calling the excellent semantic understanding and analysis capabilities of the general large language model. After that, the reply content is provided to the user to complete an interaction.

[0033] Based on the reply content, users can continue to ask questions and answer them through the general large language model, realizing multiple rounds of free interaction between the general large language model and users.

[0034] The reading assistant processing method provided by the embodiment of the present invention first determines the reading content and demand information of the user in the scene of reading the target reading material; then determines the reading prompt information corresponding to the demand information, and the reading prompt information includes an instruction for generating the reply content; thereafter, based on the reading prompt information and the reading content, generates the reading input text; finally, based on the general large language model, the reading input text is applied to interact with the user, thereby realizing the auxiliary reading of the user. The method can be applied to the introduction, accompanying reading, post-reading summary and other links of student reading, and can realize a real open dialogue by freely conversing with the user, and restore the learning scene with high offline flexibility of the user. The method does not rely on preset scripts, allowing the user to initiate a dialogue at any time at any reading stage, thereby realizing a more natural and smooth user interaction experience. Moreover, the general large language model used in the method does not need to rely on sample data in a specific field for pre-training, which can not only reduce the cost of collecting sample data and the cost of model training through sample data, but also break through the limitation of traditional models relying on data in specific fields, and realize more flexible and extensive application scenarios. In addition, by determining the user's demand information and the corresponding reading prompt information, prompt engineering technology is effectively used to stimulate and guide the general large-scale language model to generate more accurate and targeted reply content to the user, which has stronger versatility and adaptability.

[0035] Based on the above embodiment, the reading content is determined by the user's word selection operation, and the demand information is determined by the user's demand selection operation, or by the voice information or text information input by the user; Alternatively, the reading content is the full text content of the target reading material, and the demand information is determined by the preset function selection operation of the user.

[0036] Specifically, when determining the reading content and demand information of the user in the scenario of reading the target reading material, if the reading content is the content selected by the user, the reading content can be determined by the user through a word selection operation using a mouse or a touch screen.

[0037] On the display interface of the reading assistant application, after the user triggers the word selection operation, a word marking toolbar can be displayed. The word marking toolbar can be configured with function buttons. The user triggers the required selection operation by clicking the function button to be implemented, and then the reading assistant application can determine the user's demand information and enter the reading interaction interface.

[0038] For example, the function buttons that can be configured on the word-marking toolbar may include an appreciation function button, an interpretation function button, etc. In addition, the word-marking toolbar may also be configured with conventional function buttons such as a read aloud function button and a copy function button.

[0039] In addition, the user can directly trigger the reading interaction interface of the reading assistant application. On the reading interaction interface, an input area can be configured, and the user can enter text information in the input area. The text information is the demand information in text form. On the display interface of the reading assistant application, a voice receiving module can also be configured. The user can click on the voice receiving module and start speaking, and the voice receiving module can receive the user's voice information. The voice information is the demand information in voice form.

[0040] It is understandable that if the user inputs voice information, the voice information needs to be converted into text for subsequent processing.

[0041] On the reading interaction interface of the reading assistant application, preset function buttons can also be configured, such as a reading skills function button and a full-text perception function button. The user clicks on the preset function button that needs to be implemented to trigger the preset function selection operation, and the reading assistant application can then determine the user's demand information.

[0042] In the embodiment of the present invention, the user can input information through different ways, thus ensuring the convenience of using the reading assistant application.

[0043] Based on the above embodiment, it also includes: receiving a reading role selected by the user for the target reading; Based on the universal large language model, the reader role is applied to interact with the user.

[0044] Specifically, on the homepage of the reading assistant application, the user can click the book crossing function button to enter the book crossing interface. In the book crossing interface, the reading assistant application can play the role of the reading selected by the user through the universal large language model and interact with the user.

[0045] In the book sea crossing interface, multiple reading materials can be included, each of which corresponds to a text file and multiple preset characters, and each preset character is a character in the corresponding reading material. The number of preset characters can be set as needed, for example, important characters in the reading material can be selected as preset characters.

[0046] The user can click on the reading character corresponding to the target reading material and with whom he needs to communicate in the book-crossing interface, and the reading assistant application can receive the reading character and enter the character interaction interface.

[0047] On the role interaction interface, an input area can be configured, where the user can enter text questions, and a voice receiving module can be configured, where the user can click on the voice receiving module and start speaking, and the voice receiving module can receive the user's voice questions. After that, the general large language model is called to play the role of the reader, and the reply content is output in the tone of the reader role.

[0048] On the character interaction interface, the universal large language model can also start the conversation in the tone of the reading character. For example, if the reading character is Pavel Korchagin, the universal large language model can start the conversation with the following content: "Hello, classmate. I am Pavel Korchagin. Welcome to my world. Do you have anything to ask me?"

[0049] It can be understood that in the reading interaction interface under the scenario of reading the target reading material, the configured preset function buttons can also include a book-crossing function button. When the user clicks on the book-crossing function button, the reading assistant application displays multiple preset characters corresponding to the target reading material for the user to choose. When the user selects one of them as the reading material character, the reading assistant application can jump from the reading interaction interface to the character interaction interface, and then interact with the user in the tone of the reading material character through the universal large-scale language model.

[0050] In the embodiment of the present invention, by the user selecting a reading character, the reading character can interact with the user in the tone of the reading character, so that the user can have an immersive conversation with the reading character, thereby improving the user's reading experience.

[0051] Based on the above embodiment, the step of receiving the reading material role selected by the user for the target reading material further comprises: The preset background music is played in a loop, and the background of the character interaction interface that interacts with the user is replaced with the preset background superimposed with the reading character.

[0052] Specifically, after the user selects a reading role, the reading assistant application can call the preset background music corresponding to the reading role. The preset background music can be a predetermined background music corresponding to the reading role, and can be determined based on the character setting, personality characteristics and other attributes of the reading role in the target reading material. The preset background music can also be background music that matches the target reading material.

[0053] Simultaneously, the background of the character interaction interface interacting with the user can be replaced with a preset background superimposed with the reading character. The preset background can be a position where the reading character is often located in the target reading, or a random position involved in the target reading.

[0054] In the embodiment of the present invention, by playing preset background music and replacing the background of the role interaction interface with the preset background superimposed with the reading character, the user can have a more immersive conversation with the universal large language model playing the reading character in terms of vision and hearing, and can ask the universal large language model questions related to the reading character, so that the universal large language model can reply in the first person, providing the user with a more immersive conversation experience and creating a strong sense of immersion for the user, as if the user is in the world of the book and has a real interactive communication with the character. This unique interactive method is unprecedented in the prior art and can greatly improve the user's reading interest and learning effect.

[0055] On the basis of the above embodiment, the method of applying the reading role based on the universal large language model to interact with the user includes: receiving an interactive question from the user; Based on the universal large language model, the reader role is applied to answer the interactive question to obtain an interactive answer; Based on the text-to-speech technology, the interactive answer is played through the voice corresponding to the reading role.

[0056] Specifically, a general large language model is used to play the role of the reader. During the process of interacting with the user, the user's interactive questions can be received. For example, if the reader role is Sun Wukong, the interactive questions can be "Why did you make trouble in Heaven?", "Why did you knock down the ginseng fruit tree?", etc.

[0057] Afterwards, the interactive questions can be input into the general large language model, so that the general large language model can respond to the interactive questions in the tone of the reading character to obtain interactive answers.

[0058] After that, the interactive answers in text form can be converted into the voice corresponding to the reading character by using the Text-to-Speech (TTS) technology for playback. The tone and intonation of the voice corresponding to the reading character can be accurately imitated according to the characteristics of the reading character, making the reading character more vivid.

[0059] Based on the above embodiment, it also includes: Obtaining the user's question information for each reading material; The question information is classified based on a general large language model.

[0060] Specifically, the reading assistant application can also classify the user's question information for each reading material. The reading assistant application can obtain all the historical question information of the user for each reading material, and then input the question information into the general large-scale language model, and obtain the category of the question information through the powerful classification function of the general large-scale language model. The categories can include appreciation, language comprehension, exploration, and reading skills, etc., and can also include essential knowledge recognition, word and sentence comprehension, structural thinking analysis, technique appreciation, language strategy application, and others. The specific categories involved in the question information are not limited here.

[0061] Thereafter, the categories of the problem information may be statistically displayed, for example, the number or percentage of the problem information belonging to the same category may be determined and then displayed through a statistical chart. The statistical chart may include a bar chart, a pie chart, and the like.

[0062] On the homepage of the Reading Assistant application, users can click the Reading Report function button to enter the Reading Report interface. The Reading Report interface can display statistics on the categories of problem information.

[0063] In addition, the reading assistant application can also count the number of completed readings, cumulative reading time, and cumulative number of questions asked by the user, and can also count the distribution of the user's reading time within a preset time period, for example, it can count the reading time and reading time details of each day within the preset time period, and can also switch to view the reading time and reading time details of each day within the preset time period before and after. Here, the preset time period can be one week, 15 days, or one month, etc.

[0064] In the embodiment of the present invention, the user's question information for each reading material is classified through a general large-scale language model, which can not only accurately classify the students' questions into multiple dimensions, but also based on these classification statistical results, the general large-scale language model can be used to intelligently analyze the students' needs and confusions, and tailor solutions and reading suggestions for each student.

[0065] Based on the above embodiment, it also includes: Receiving the reading companion question of the user, and the picture uploaded by the user or the name of the reading material input by the user; Based on a general large language model, the accompanying reading questions and the names of the pictures or readings are applied to interact with the user.

[0066] Specifically, by clicking the smart reading companion function button, the user can enter the smart reading companion interface and start a free conversation. The free conversation can be started by a general large language model. For example, the greeting sentence to start the conversation can be "Hello, thank you for choosing me as your reading partner. Let's start reading together! You can take a photo to tell me what you are reading, or you can take an article!"

[0067] At this point, the user can upload a picture or enter the name of the reading material, which can be a picture of the reading content. If there is a picture, the general large language model can give a default response, such as "What do you want to do with these contents?", and can provide users with follow-up options such as summary / appreciation / interpretation as accompanying questions. The copy can be expressed as "Help me summarize", "Help me appreciate this text", "Help me interpret this text", etc. The subsequent multiple rounds of dialogue between the general large language model and the user will be based on the picture for knowledge enhancement.

[0068] Users can also enter personalized reading questions, which can be in text or voice form. If the general large language model recognizes that the reading questions are not related to topics such as reading and learning, it can prompt the user, such as "You can ask me questions related to reading and learning, or upload other pictures and texts. Users can upload pictures again."

[0069] After receiving a new picture, the general large language model can identify the content of the new picture and start a new round of free conversation based on the content of the new picture.

[0070] After entering the smart reading companion interface, the general large language model can also start a normal style conversation with the user.

[0071] In the embodiment of the present invention, a universal large-scale language model can be used to provide users with a companion reading function for other reading materials, thereby broadening the user's usage scenarios and increasing the user's usage frequency.

[0072] In summary, the reading assistant processing method provided in the embodiment of the present invention can guide users to understand and appreciate famous works and articles in an heuristic way, improve students' reading interest and reading-related literacy, and empower existing student-end tablets with a general large-scale language model, responding to the current policy requirements for intelligent learning terminals. The method provides a purely open dialogue system that does not rely on preset scripts, and guides students to understand and appreciate famous works and articles in an interesting and immersive reading method, which will improve students' reading interest and reading-related literacy; at the same time, with the help of a general large-scale language model, it can guide students to appreciate, interpret and summarize texts, which can significantly improve students' reading skills and overall perception. In addition, the questions asked by students during the reading process are counted, and solutions and reading suggestions are provided to students based on the statistical results. The implementation of this method will promote students' after-school autonomous learning ability, effectively enhance students' reading interest and motivation, and improve reading effectiveness.

[0073] like Figure 2 As shown, based on the above embodiment, an embodiment of the present invention provides a reading assistant processing device, including: The demand determination module 21 is used to determine the reading content and demand information of the user in the scenario of reading the target reading material; A prompt determination module 22, used to determine reading prompt information corresponding to the demand information, wherein the reading prompt information includes instructions for generating reply content; A text determination module 23, used to generate a reading input text based on the reading prompt information and the reading content; The interaction module 24 is used to interact with the user by applying the reading input text based on a universal large language model.

[0074] Based on the above embodiment, the reading content is determined by the user's word selection operation, and the demand information is determined by the user's demand selection operation, or by the voice information or text information input by the user; Alternatively, the reading content is the full text content of the target reading material, and the demand information is determined by the preset function selection operation of the user.

[0075] Based on the above embodiment, it also includes: A reading material role receiving module, used for receiving the reading material role selected by the user for the target reading material; The interaction module is also used for: Based on the universal large language model, the reader role is applied to interact with the user.

[0076] Based on the above embodiment, it also includes: The display update module is used to play the preset background music in a loop and replace the background of the role interaction interface that interacts with the user with the preset background superimposed with the reading character.

[0077] Based on the above embodiment, the interaction module is further used for: receiving an interactive question from the user; Based on the universal large language model, the reader role is applied to answer the interactive question to obtain an interactive answer; Based on the text-to-speech technology, the interactive answer is played through the voice corresponding to the reading role.

[0078] Based on the above embodiment, it also includes: A question information acquisition module, used to acquire the question information of the user for each reading material; The question classification module is used to classify the question information based on a general large language model.

[0079] Based on the above embodiment, it also includes: An information input module, used to receive the reading companion questions of the user, and the pictures uploaded by the user or the names of the reading materials input by the user; The interaction module is further used for: Based on a general large language model, the accompanying reading questions and the names of the pictures or readings are applied to interact with the user.

[0080] Specifically, the functions of each module in the reading assistant processing device provided in the embodiment of the present invention correspond one-to-one to the operation flow of each step in the above-mentioned method embodiment, and the effects achieved are also consistent. Please refer to the above-mentioned embodiment for details, and no further description will be given in the embodiment of the present invention.

[0081] Figure 3 The following is a schematic diagram of the physical structure of a reading assistant all-in-one machine. Figure 3 As shown, the reading assistant integrated machine may include: a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other through the communication bus 340. The memory 330 may store a general large language model, and the processor 310 may store and run a computer program.

[0082] When executing a computer program, the processor 310 may call the general large language model in the memory 330 to execute the reading assistant processing method provided in the above embodiments.

[0083] It is understandable that the reading assistant all-in-one machine can be a smart device such as a student tablet, a learning terminal, etc.

[0084] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the relevant technology or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0085] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the reading assistant processing method provided in the above embodiments.

[0086] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to execute the reading assistant processing method provided in the above embodiments.

[0087] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0088] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A reading assistant processing method, characterized in that: include: Determine the reading content and demand information of users in the context of reading target reading materials; Determining reading prompt information corresponding to the demand information, wherein the reading prompt information includes an instruction for generating reply content; Generate a reading input text based on the reading prompt information and the reading content; Based on a general large language model, the application reads the input text and interacts with the user.

2. The reading assistant processing method according to claim 1, characterized in that: The reading content is determined by the word selection operation of the user, and the demand information is determined by the demand selection operation of the user, or by the voice information or text information input by the user; Alternatively, the reading content is the full text content of the target reading material, and the demand information is determined by the preset function selection operation of the user.

3. The reading assistant processing method according to claim 1, characterized in that: Also includes: receiving a reading role selected by the user for the target reading; Based on the universal large language model, the reader role is applied to interact with the user.

4. The reading assistant processing method according to claim 3, characterized in that: The step of receiving the reading material role selected by the user for the target reading material comprises: The preset background music is played in a loop, and the background of the character interaction interface that interacts with the user is replaced with the preset background superimposed with the reading character.

5. The reading assistant processing method according to claim 3, characterized in that: The method of applying the reading role based on the universal large language model to interact with the user includes: receiving an interactive question from the user; Based on the universal large language model, the reader role is applied to answer the interactive question to obtain an interactive answer; Based on the text-to-speech technology, the interactive answer is played through the voice corresponding to the reading role.

6. The reading assistant processing method according to any one of claims 1 to 5, characterized in that: Also includes: Obtaining the user's question information for each reading material; The question information is classified based on a general large language model.

7. The reading assistant processing method according to any one of claims 1 to 5, characterized in that: Also includes: Receiving the reading companion question of the user, and the picture uploaded by the user or the name of the reading material input by the user; Based on a general large language model, the accompanying reading questions and the names of the pictures or readings are applied to interact with the user.

8. A reading assistant processing device, characterized in that: include: The demand determination module is used to determine the reading content and demand information of the user in the scenario of reading the target reading material; A prompt determination module, used to determine reading prompt information corresponding to the demand information, wherein the reading prompt information includes an instruction for generating reply content; A text determination module, used for generating a reading input text based on the reading prompt information and the reading content; An interaction module is used for interacting with the user by applying the reading input text based on a general large language model.

9. A reading assistant integrated machine, characterized in that: The invention comprises a memory, a processor, a general large language model stored in the memory, and a computer program stored in and executable on the processor; When the processor executes the computer program, the universal large language model is called to implement the reading assistant processing method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the reading assistant processing method according to any one of claims 1 to 7 is implemented.

Images