Auxiliary reading method of target document, electronic device and computer readable storage medium

Abstract

The present disclosure relates to an auxiliary reading method of a target document, an electronic device and a computer readable storage medium. The method comprises: in response to an auxiliary reading request for the target document, determining a requested target auxiliary reading function, wherein the target auxiliary reading function comprises at least one of the following: an automatic summary function of document content, a reminder function of key information, a translation function of document content, a user question reply function; obtaining auxiliary reading information corresponding to the target auxiliary reading function; and outputting the auxiliary reading information.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and more particularly to a method for assisting in reading a target document, an electronic device, and a computer-readable storage medium. Background Technology

[0002] In the industrial and technical fields, documentation is an essential tool for conveying product information, operating instructions, or safety guidelines. For example, in the medical field, engineers or users of medical devices frequently need to read and understand a large amount of related documentation in order to operate, develop, or maintain the equipment.

[0003] In actual document reading, readers often face several challenges. For example, when faced with unfamiliar or information-heavy documents, it may be difficult to accurately grasp the key information. This problem is particularly pronounced when dealing with documents in a language other than one's native tongue. Furthermore, reading barriers can affect the accurate operation, development, and maintenance of equipment, impacting work efficiency and even safety. Summary of the Invention

[0004] This disclosure provides a technical solution for assisting in reading target documents.

[0005] According to one aspect of this disclosure, a method for assisting in reading a target document is provided, comprising:

[0006] In response to an assisted reading request for the target document, the requested target assisted reading function is determined, wherein the target assisted reading function includes at least one of the following: automatic summary function of document content, reminder function of key information, translation function of document content, and reply function to user questions;

[0007] Obtain the auxiliary reading information corresponding to the target auxiliary reading function;

[0008] Output the aforementioned auxiliary reading information.

[0009] In one possible implementation, the target assisted reading function includes the translation function, and obtaining the assisted reading information corresponding to the target assisted reading function includes:

[0010] Identify the content to be translated in the target document;

[0011] The content to be translated is preprocessed, wherein the preprocessing includes extracting text from the content to be translated;

[0012] The preprocessed content to be translated is translated to obtain the translated text of the content to be translated.

[0013] In one possible implementation, the preprocessing further includes at least one of the following:

[0014] Filter out content that does not contain text from the content to be translated;

[0015] Extract the metadata of the content to be translated;

[0016] Based on a pre-built user dictionary, the content contained in the user dictionary within the text of the content to be translated is masked.

[0017] In one possible implementation, if the content to be translated is audio or video, the metadata of the content to be translated includes a timestamp and a voiceprint.

[0018] In one possible implementation, translating the preprocessed content to be translated includes:

[0019] Classify the complexity of the preprocessed content to be translated;

[0020] Based on the complexity classification results, the preprocessed content to be translated is translated.

[0021] In one possible implementation, translating the preprocessed content to be translated based on the complexity classification result includes:

[0022] In response to the complexity classification result being classified as complex translation, the preprocessed content to be translated is translated using a translator based on a large language model.

[0023] or,

[0024] In response to the complexity classification result being classified as non-complex translation, the preprocessed content to be translated is translated using a Transformer-based translator.

[0025] In one possible implementation, the method further includes:

[0026] Post-processing is performed based on the translated text;

[0027] The post-processing includes at least one of the following:

[0028] Based on a pre-built user dictionary, the masked words in the translated text are demasked;

[0029] In response to the fact that the content to be translated is structured text, the translated text is backfilled into the original position in the target document;

[0030] In response to the fact that the content to be translated is non-text content, adaptive processing is performed according to the content type of the content to be translated, wherein the adaptive processing includes at least one of the following: image synthesis, text synthesis of audio.

[0031] In one possible implementation, the step of backfilling the translated text to its original position in the target document includes:

[0032] For any content block in the content to be translated, a first pointer and a second pointer are generated, wherein the first pointer points to the beginning position of the content block and the second pointer points to the end position of the content block.

[0033] Based on the first pointer and the second pointer, the translated content of the content block in the translated text is backfilled into the original position of the content block in the target document.

[0034] In one possible implementation, the structured text is in HTML or PDF format.

[0035] In one possible implementation, the method further includes:

[0036] On the display interface of the target document, auxiliary reading controls are displayed;

[0037] In response to the user triggering the auxiliary reading control, an auxiliary reading interface is displayed, wherein the auxiliary reading interface is used to provide the auxiliary reading information.

[0038] In one possible implementation, the assisted reading interface includes at least one of the following functional controls:

[0039] The summary function control is used to generate an auxiliary reading request corresponding to the automatic summary function of the document content;

[0040] A reminder function control is used to generate an auxiliary reading request corresponding to the reminder function for the key information;

[0041] A translation function control is used to generate an auxiliary reading request corresponding to the translation function of the document content;

[0042] Feedback control is used to obtain feedback information from user input regarding auxiliary reading information or target documents.

[0043] In one possible implementation, determining the requested target assistive reading function includes:

[0044] In response to the user triggering the auxiliary reading control to generate an auxiliary reading request, the requested target auxiliary reading function is determined to include the automatic summary function of the document content and the reminder function of the key information.

[0045] In one possible implementation, the auxiliary reading information corresponding to the key information reminder function includes at least one of the following:

[0046] Security notifications are used to remind users of security precautions in the target document.

[0047] Operating procedures are used to guide users to correctly install or operate the device according to the steps specified in the target document.

[0048] In one possible implementation, determining the requested target assistive reading function includes:

[0049] In response to a user sending a question in the assisted reading interface, it is determined that the requested target assisted reading function includes the user question reply function.

[0050] In one possible implementation, the target assisted reading function includes the user question response function, and obtaining the assisted reading information corresponding to the target assisted reading function includes:

[0051] Retrieve reference information from the target document that relates to the user question addressed by the assisted reading request;

[0052] Based on the reference information, auxiliary reading information is generated, wherein the auxiliary reading information includes a jump link, which is used to jump to the location of the reference information in the target document.

[0053] In one possible implementation, the target document includes a relevant technical manual for the medical device.

[0054] According to one aspect of this disclosure, an auxiliary reading device for a target document is provided, comprising:

[0055] The determining module is configured to determine the requested target auxiliary reading function in response to an auxiliary reading request for the target document, wherein the target auxiliary reading function includes at least one of the following: automatic summary function of document content, reminder function of key information, translation function of document content, and reply function to user questions;

[0056] The acquisition module is used to acquire the auxiliary reading information corresponding to the target auxiliary reading function;

[0057] The output module is used to output the auxiliary reading information.

[0058] In one possible implementation, the target-assisted reading function includes the translation function, and the obtaining module is used for:

[0059] Identify the content to be translated in the target document;

[0060] The content to be translated is preprocessed, wherein the preprocessing includes extracting text from the content to be translated;

[0061] The preprocessed content to be translated is translated to obtain the translated text of the content to be translated.

[0062] In one possible implementation, the preprocessing further includes at least one of the following:

[0063] Filter out content that does not contain text from the content to be translated;

[0064] Extract the metadata of the content to be translated;

[0065] Based on a pre-built user dictionary, the content contained in the user dictionary within the text of the content to be translated is masked.

[0066] In one possible implementation, if the content to be translated is audio or video, the metadata of the content to be translated includes a timestamp and a voiceprint.

[0067] In one possible implementation, the obtaining module is used for:

[0068] Classify the complexity of the preprocessed content to be translated;

[0069] Based on the complexity classification results, the preprocessed content to be translated is translated.

[0070] In one possible implementation, the obtaining module is used for:

[0071] In response to the complexity classification result being classified as complex translation, the preprocessed content to be translated is translated using a translator based on a large language model.

[0072] or,

[0073] In response to the complexity classification result being classified as non-complex translation, the preprocessed content to be translated is translated using a Transformer-based translator.

[0074] In one possible implementation, the device further includes:

[0075] The post-processing module is used to perform post-processing based on the translated text;

[0076] The post-processing includes at least one of the following:

[0077] Based on a pre-built user dictionary, the masked words in the translated text are demasked;

[0078] In response to the fact that the content to be translated is structured text, the translated text is backfilled into the original position in the target document;

[0079] In response to the fact that the content to be translated is non-text content, adaptive processing is performed according to the content type of the content to be translated, wherein the adaptive processing includes at least one of the following: image synthesis, text synthesis of audio.

[0080] In one possible implementation, the post-processing module is used for:

[0081] For any content block in the content to be translated, a first pointer and a second pointer are generated, wherein the first pointer points to the beginning position of the content block and the second pointer points to the end position of the content block.

[0082] Based on the first pointer and the second pointer, the translated content of the content block in the translated text is backfilled into the original position of the content block in the target document.

[0083] In one possible implementation, the structured text is in HTML or PDF format.

[0084] In one possible implementation, the device further includes:

[0085] The display module is used to display auxiliary reading controls on the display interface of the target document;

[0086] The display module is used to display an auxiliary reading interface in response to the user triggering the auxiliary reading control, wherein the auxiliary reading interface is used to provide the auxiliary reading information.

[0087] In one possible implementation, the assisted reading interface includes at least one of the following functional controls:

[0088] The summary function control is used to generate an auxiliary reading request corresponding to the automatic summary function of the document content;

[0089] A reminder function control is used to generate an auxiliary reading request corresponding to the reminder function for the key information;

[0090] A translation function control is used to generate an auxiliary reading request corresponding to the translation function of the document content;

[0091] Feedback control is used to obtain feedback information from user input regarding auxiliary reading information or target documents.

[0092] In one possible implementation, the determining module is used to:

[0093] In response to the user triggering the auxiliary reading control to generate an auxiliary reading request, the requested target auxiliary reading function is determined to include the automatic summary function of the document content and the reminder function of the key information.

[0094] In one possible implementation, the auxiliary reading information corresponding to the key information reminder function includes at least one of the following:

[0095] Security notifications are used to remind users of security precautions in the target document.

[0096] Operating procedures are used to guide users to correctly install or operate the device according to the steps specified in the target document.

[0097] In one possible implementation, the determining module is used to:

[0098] In response to a user sending a question in the assisted reading interface, it is determined that the requested target assisted reading function includes the user question reply function.

[0099] In one possible implementation, the target assisted reading function includes the user question response function, and the obtaining module is used for:

[0100] Retrieve reference information from the target document that relates to the user question addressed by the assisted reading request;

[0101] Based on the reference information, auxiliary reading information is generated, wherein the auxiliary reading information includes a jump link, which is used to jump to the location of the reference information in the target document.

[0102] In one possible implementation, the target document includes a relevant technical manual for the medical device.

[0103] According to one aspect of this disclosure, an electronic device is provided, comprising: one or more processors; a memory for storing executable instructions; wherein the one or more processors are configured to invoke the executable instructions stored in the memory to perform the method described above.

[0104] According to one aspect of this disclosure, a computer-readable storage medium is provided that stores computer program instructions thereon, which, when executed by a processor, implement the above-described method.

[0105] According to one aspect of this disclosure, a computer program product is provided, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in an electronic device, a processor in the electronic device performs the above-described method.

[0106] In this embodiment of the disclosure, in response to an assisted reading request for a target document, the requested target assisted reading function is determined. The target assisted reading function includes at least one of the following: automatic summary of document content, reminder of key information, translation of document content, and response to user questions. Assisted reading information corresponding to the target assisted reading function is obtained and output. This effectively assists the user in reading the target document, helps the user overcome obstacles in document reading, effectively and accurately conveys key information in the target document to the user, reduces the risk of misunderstanding or overlooking document content, and thus improves work efficiency.

[0107] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure.

[0108] Other features and aspects of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0109] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the specification, serve to illustrate the technical solutions of this disclosure.

[0110] Figure 1 A flowchart illustrating an auxiliary reading method for a target document provided in an embodiment of this disclosure is shown.

[0111] Figure 2 This diagram illustrates the method for assisting in reading a target document according to an embodiment of the present disclosure, in which an assisting reading control is displayed on the target document's display interface.

[0112] Figure 3 This diagram illustrates an auxiliary reading interface in the auxiliary reading method for a target document provided in an embodiment of the present disclosure.

[0113] Figure 4 Another schematic diagram shows the auxiliary reading interface in the auxiliary reading method for target documents provided in this embodiment of the present disclosure.

[0114] Figure 5 This diagram illustrates a block diagram of an aided reading device for a target document provided in an embodiment of the present disclosure.

[0115] Figure 6 A block diagram of an electronic device 1900 provided in an embodiment of this disclosure is shown. Detailed Implementation

[0116] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0117] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0118] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0119] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0120] In this embodiment of the disclosure, in response to an assisted reading request for a target document, the requested target assisted reading function is determined. The target assisted reading function includes at least one of the following: automatic summary of document content, reminder of key information, translation of document content, and response to user questions. Assisted reading information corresponding to the target assisted reading function is obtained and output. This effectively assists the user in reading the target document, helps the user overcome obstacles in document reading, effectively and accurately conveys key information in the target document to the user, reduces the risk of misunderstanding or overlooking document content, and thus improves work efficiency.

[0121] The method for assisting in reading the target document provided in the embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0122] Figure 1A flowchart illustrating an assisted reading method for a target document provided in an embodiment of this disclosure is shown. In one possible implementation, the entity executing the assisted reading method for the target document may be an assisted reading device for the target document. For example, the assisted reading method for the target document may be executed by a terminal device, a server, or other electronic devices. The terminal device may be a user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, in-vehicle device, or wearable device, etc. In some possible implementations, the assisted reading method for the target document may be implemented by a processor calling computer-readable instructions stored in memory. Figure 1 As shown, the auxiliary reading method for the target document includes steps S11 to S13.

[0123] In step S11, in response to the assisted reading request for the target document, the requested target assisted reading function is determined, wherein the target assisted reading function includes at least one of the following: automatic summary function of document content, reminder function of key information, translation function of document content, and reply function to user questions.

[0124] In step S12, the auxiliary reading information corresponding to the target auxiliary reading function is obtained.

[0125] In step S13, the auxiliary reading information is output.

[0126] In this disclosure, the target document can be of various types, including Operation Manuals, User Manuals, Installation Manuals, Maintenance Manuals, Repair Manuals, Work Manuals, Service Manuals, Safety Manuals, Technical Reference Manuals, and Product Brochures, without limitation. Specifically, Operation Manuals provide daily operating methods and procedures for the equipment or system; User Manuals are geared towards end users, introducing product functions and usage guidelines; Installation Manuals guide users or technicians in installing and setting up the equipment; Maintenance Manuals include guidance on regular maintenance, inspection, and repair; Repair Manuals provide fault diagnosis and repair procedures, typically for professional maintenance personnel; Work Manuals provide detailed work instructions for performing specific tasks or processes; Work Manuals combine operation and maintenance guidelines to provide comprehensive technical support for service engineers; Safety Manuals emphasize safe operating procedures and emergency response measures; Technical Reference Manuals provide more in-depth technical details and system architecture information; and Product Brochures include product overviews, specifications, and features.

[0127] In one possible implementation, the target document includes a relevant technical manual for the medical device.

[0128] In this implementation, the target document can be an operation manual or installation manual for the medical device, and there are no restrictions on its use. The operation manual can describe in detail the startup, operation, monitoring, and shutdown processes of the medical device, and may include potential problems encountered during operation and their solutions. The installation manual can guide users or technicians on how to correctly install the medical device, including hardware configuration and software installation. Furthermore, the installation manual can provide precautions and safety guidelines during the installation process.

[0129] In this implementation, by responding to a request for assisted reading of a technical manual related to a medical device, the requested target assisted reading function is determined, the assisted reading information corresponding to the target assisted reading function is obtained, and the assisted reading information is output. This can effectively assist the user in reading the technical manual related to the medical device and effectively and accurately convey the key information in the technical manual to the user.

[0130] In this embodiment of the disclosure, the requested target assisted reading function can be determined in response to an assisted reading request for the target document. The assisted reading request can refer to a request initiated by a user for the target document, requesting the system to provide additional assistance to help the user better understand and read the target document.

[0131] In one possible implementation, an auxiliary reading control can be displayed on the target document's interface, and the user can initiate an auxiliary reading request by triggering (e.g., clicking) the auxiliary reading control.

[0132] In another possible implementation, in the assisted reading interface, users can initiate specific assisted reading requests by clicking different functional controls. For example, clicking the summary control can request an automatic summary of the document content, or clicking the translation control can request a translation of the document content.

[0133] In another possible implementation, users can enter questions in the input boxes of the assisted reading interface, and the system can initiate assisted reading requests by processing user questions, especially for the user question reply function.

[0134] In another possible implementation, user interactions while reading the target document, such as pausing, scrolling, or marking specific paragraphs, can be recognized by the system and used to automatically initiate assisted reading requests to provide more personalized help.

[0135] In another possible implementation, the system may automatically initiate assisted reading requests based on preset rules or conditions, such as automatically reminding the user when a security notification or operation procedure is detected in the target document.

[0136] In another possible implementation, if the system supports speech recognition, the user can initiate an assisted reading request via voice command.

[0137] In another possible implementation, if the system has image or video recognition capabilities, users can initiate translation or content recognition-assisted reading requests by pointing to or uploading images or videos containing text.

[0138] In this embodiment of the disclosure, the target-assisted reading function can refer to a specific function or service provided by the system in response to a user-initiated assisted reading request, used to help the user better understand and process the target document. In this embodiment of the disclosure, the target-assisted reading function includes at least one of the following: automatic summary of document content, reminder of key information, translation of document content, and response to user questions.

[0139] The automatic document summary function uses automatic summarization technology to generate a concise summary or outline of the target document's content. By analyzing the entire target document, key sentences and main points can be extracted to form a succinct summary. This allows users to quickly grasp the main content and core ideas of the target document without having to read the entire document, thus helping them quickly obtain information and save time.

[0140] The key information is designed to identify critical information within a document, such as security notices, operating procedures, and precautions, and to alert users as they read. In this embodiment, natural language processing technology can be used to scan the document content of the target document and mark key information. When a user reads a relevant section or under specific circumstances, the system can remind the user of this key information to ensure that the user does not miss important operating instructions or safety guidelines.

[0141] The document content translation function can provide automatic translation services for document content, translating the target document from the original language into a language familiar to the user. This is particularly useful for users in non-English speaking countries, as it helps them overcome language barriers and better understand and use the target document. In embodiments of this disclosure, the system can use machine translation techniques, such as statistical or neural network-based translation models, to provide accurate and fluent translations.

[0142] The user question and answer feature allows users to ask questions about document content and receive answers. The system can use large language models or question-and-answer systems to understand user questions and retrieve relevant information from the target document to generate answers. Furthermore, the system can provide jump links that directly guide users to the specific location within the target document where the answer is located, thus providing more intuitive and convenient assistance.

[0143] In one possible implementation, the method further includes: displaying an auxiliary reading control on the display interface of the target document; and displaying an auxiliary reading interface in response to a user triggering the auxiliary reading control, wherein the auxiliary reading interface is used to provide the auxiliary reading information.

[0144] In this implementation, one or more auxiliary reading controls can be displayed on the target document's interface. These auxiliary reading controls are part of the user interface, allowing users to directly interact with the target document and request auxiliary reading functions. Figure 2 This diagram illustrates the display of auxiliary reading controls on the target document's display interface in an auxiliary reading method provided by an embodiment of this disclosure. For example... Figure 2 As shown, an auxiliary reading control 210 can be displayed on the target document's display interface 200.

[0145] When users encounter questions or difficult-to-understand content while reading a target document, they can request assistance by triggering (e.g., clicking) an auxiliary reading control. The trigger control can be a mouse click, touch, or other interactive method.

[0146] In response to user-triggered actions, an auxiliary reading interface can be displayed. Figure 3 This diagram illustrates an auxiliary reading interface in the auxiliary reading method for a target document provided in this embodiment of the disclosure. (In conjunction with...) Figure 2 and Figure 3 It can respond to Figure 2 The auxiliary reading control 210 shown is triggered, displaying... Figure 3 The assisted reading interface 220 shown is specifically designed to provide assisted reading information. It can contain multiple functional controls and input boxes, through which users can access different assisted reading services.

[0147] This implementation provides users with an intuitive and convenient auxiliary reading interface, enabling them to easily access necessary supplementary reading information while reading the target document, thereby improving reading efficiency and comprehension depth. This design also allows users to receive immediate assistance when encountering difficulties, enhancing the user experience.

[0148] As an example of this implementation, the assisted reading interface includes at least one of the following functional controls: a summary function control, used to generate an assisted reading request corresponding to the automatic summary function of the document content; a reminder function control, used to generate an assisted reading request corresponding to the reminder function of the key information; a translation function control, used to generate an assisted reading request corresponding to the translation function of the document content; and a feedback function control, used to obtain feedback information input by the user regarding the assisted reading information or the target document.

[0149] In this example, the summary function control can be used to generate an automatic summary request for document content (i.e., an assisted reading request corresponding to the automatic summary function). When the user triggers the summary function control, the system can process the document content of the target document and provide a summary or outline to help the user quickly understand the main content of the document.

[0150] The reminder function control can be used to generate critical information reminder requests (i.e., assisted reading requests corresponding to the critical information reminder function). The system can identify critical information in the target document, such as security notices or operating procedures, and remind the user of this information while reading.

[0151] The translation control allows users to request the translation of document content. It can be applied to the current conversation, the entire document, or a specific section of the document. When a user triggers the translation control, the system can translate the selected content from one language to another, helping the user overcome language barriers.

[0152] Feedback controls can be used to collect user feedback on supplementary reading information or target documents. Users can provide their opinions and suggestions on the system through feedback controls, helping to improve the system.

[0153] In one example, the assisted reading interface may also include an input box where the user can ask specific questions. The system can then process these questions and provide answers or relevant information, enhancing user interaction with the system.

[0154] Figure 4 This diagram illustrates another schematic of the assisted reading interface in the assisted reading method for target documents provided in this embodiment of the present disclosure. For example... Figure 4 As shown, the auxiliary reading interface 220 may include a summary function control 221, a reminder function control 222, a feedback function control 223, and a translation function control 224. Additionally, the auxiliary reading interface 220 may also include an input box 225, through which the user can input questions.

[0155] By adopting this example, users can intuitively find and use the functions they need, thus improving the usability of the assisted reading interface.

[0156] In one possible implementation, determining the requested target assisted reading function includes: in response to the user triggering the assisted reading control to generate an assisted reading request, determining that the requested target assisted reading function includes the automatic summary function of the document content and the reminder function of the key information.

[0157] As an example of this implementation, in response to a user triggering the assistive reading control, an assistive reading interface can be displayed, showing a summary of the target document's content and key information.

[0158] In this implementation, by responding to user-triggered assisted reading controls, a summary and key information of the target document's content are automatically provided. This allows users to find the information they need more quickly, reducing the time spent reading the entire document and thus improving work efficiency. Simultaneously, the reminders of key information help users avoid operational errors and enhance the accuracy of their operations.

[0159] In one possible implementation, in response to the requested target assisted reading function, which includes an automatic summary function of document content, a large language model corresponding to the summary function is invoked, and a summary of the target document is output through the large language model corresponding to the summary function. The large language model corresponding to the summary function may contain fixed prompt words. These fixed prompt words can represent predefined words or phrases. For example, the fixed prompt words in the large language model corresponding to the summary function could be "generate summary," etc.

[0160] In one possible implementation, in response to the requested target assisted reading function including a key information reminder function, the deep learning model corresponding to the reminder function can be invoked, and the key information of the target document can be output through the deep learning model corresponding to the reminder function.

[0161] The deep learning model corresponding to the reminder function can be a pre-trained multi-domain few-shot learning model. That is, the deep learning model corresponding to the reminder function can work effectively in different domains and only requires a small number of training samples. Multi-domain few-shot learning models are particularly suitable for handling situations where there may be a lack of a large amount of labeled data in a specific domain (such as a technical manual).

[0162] By using a small number of training samples from different domains, the deep learning model for the alert function can effectively learn how to classify key information (such as safety notifications and operation steps). As an example of this implementation, during the training process of the deep learning model for the alert function, the training samples can be annotated at the sentence level to identify which sentences contain safety notifications or operation steps. This detailed annotation helps the deep learning model for the alert function to better learn and understand the key information in the document.

[0163] Before inputting the sentence into the deep learning model corresponding to the reminder function, a word segmenter can be used to segment the sentence and calculate the word embedding for each word. This allows the deep learning model to understand the relationships and semantics between words. The sentence can then be encoded based on the word embeddings to form a holistic sentence representation, which serves as the input to the deep learning model for the reminder function.

[0164] As an example of this implementation, security notifications and operational steps can be displayed in two separate lists. This structured output allows users to quickly browse and understand key information.

[0165] In one possible implementation, the auxiliary reading information corresponding to the key information reminder function includes at least one of the following: a security notification, used to remind the user to pay attention to security operation matters in the target document; and operation steps, used to guide the user to correctly install or operate the device according to the steps specified in the target document.

[0166] In this implementation, safety notifications can be used to alert users to safety precautions during operation. These safety precautions may include, but are not limited to, correct operating procedures, potential risks, necessary protective measures, and emergency response strategies. The system automatically identifies safety-related statements in the target document and highlights or presents this information to the user in the form of notifications, ensuring that the user follows safety guidelines when performing tasks, thereby reducing the risk of accidents. This configuration effectively avoids the problem of overlooking safety precautions due to the reader's own negligence when dealing with documents with a large amount of information.

[0167] Operating procedures provide users with specific steps that must be followed when installing or operating the equipment. These procedures detail how to correctly set up, configure, use, and maintain the equipment. The system can extract operating procedure information from target documentation and present it to users in a clear and organized manner, helping them to correctly perform each operation according to the manufacturer's instructions and avoiding errors caused by misunderstanding or ignoring the instruction manual.

[0168] In this implementation, key information can be presented to the user in various ways, such as through highlighted text, pop-ups, sidebar annotations, or dedicated reminder areas. The presentation of this information is designed to attract the user's attention and provide intuitive guidance.

[0169] This implementation not only improves the efficiency of users reading target documents but also enhances operational security. Users can quickly identify and respond to critical security and operational information, thereby reducing operational errors and potential risks.

[0170] The inventors recognized that reading barriers posed by non-native language text can be a significant cause of low reading efficiency for readers. While translation tools exist to address this issue, conventional translation methods struggle to guarantee accuracy, especially for highly specialized engineering documents, where reduced accuracy leads to a serious decrease in credibility. Improvements have been made in one and more embodiments described below in this application.

[0171] In one possible implementation, the target assisted reading function includes the translation function, and obtaining the assisted reading information corresponding to the target assisted reading function includes: determining the content to be translated in the target document; preprocessing the content to be translated, wherein the preprocessing includes extracting text from the content to be translated; and translating the preprocessed content to be translated to obtain the translated text of the content to be translated.

[0172] In this implementation, the parts of the target document that need to be translated, i.e., the content to be translated, can be identified first. The content to be translated can include the entire target document, a specific chapter of the target document, or any text paragraph specified by the user.

[0173] After determining the content to be translated, preprocessing can be performed. Preprocessing may include extracting text from the content, that is, extracting plain text information from the content for the translation model to process. This can include extracting text from structured documents (such as HTML (HyperText Markup Language) documents, PDF (Portable Document Format) documents), or recognizing and extracting text from images (OCR (Optical Character Recognition) technology).

[0174] After preprocessing, a translation model can be used to translate the content. This translation model can be a statistical machine translation model, a neural network model, or any other translation technique. The translation process converts the text from the source language into the target language, generating the translated text.

[0175] Automated translation processes can significantly reduce or eliminate the need for professional human translation services. Automatic translation makes target documents easier for non-native speakers to understand, improving readability and usability. More importantly, in the examples of this application, preprocessing the document before translation achieves higher efficiency and accuracy compared to direct mechanical translation, resulting in a translated document that is closer to language usage and ensures the reliability of the content. The implementation of preprocessing is illustrated below.

[0176] In one possible implementation, the preprocessing further includes at least one of the following: filtering out content that does not contain text from the content to be translated; extracting metadata from the content to be translated; and masking the content in the user dictionary contained in the text of the content to be translated, based on a pre-built user dictionary. As a technique in the field of computer algorithms, masking can be understood as a technique for marking or filtering specific regions. It can be figuratively understood as covering the original input text with a mask, thereby blocking or selecting certain elements. Its specific implementation can refer to any existing technology, and this disclosure is not limited thereto.

[0177] As an example of this implementation, preprocessing can include filtering out content that does not contain text from the content to be translated, i.e., content that does not need to be translated, such as non-text elements like images, charts, and professional drawings. Filtering out non-text content reduces unnecessary translation workload and avoids invalid translation attempts on non-text content. This improves efficiency while avoiding the decrease in accuracy caused by invalid translations.

[0178] As an example of this implementation, preprocessing may include extracting metadata from the content to be translated. This metadata may include file attributes (such as author, creation date, etc.), location information (such as the file's path in the system, the text's position in an image), timestamps (such as timestamps for video frames, audio frames, and the file's creation or modification time), and voiceprints. Timestamps indicate the exact point in time of a specific data segment within a media file. In audio or video translation, timestamps ensure that the translated text (such as subtitles) is synchronized with the dialogue or actions in the media file. This is crucial for ensuring that users correctly understand the sequence and context of the content. Voiceprints refer to an individual's unique vocal characteristics and can be used to identify the speaker. In audio synthesis, voiceprints can be used to preserve the vocal characteristics of the original audio, making the synthesized audio sound more natural and closer to human pronunciation. The metadata of the content to be translated helps preserve the document's context and structure during the translation process, ensuring that the translated document remains consistent with the original text, thereby further improving the reliability of the translation.

[0179] In one example, if the content to be translated is audio or video, the metadata of the content to be translated includes a timestamp and a voiceprint.

[0180] As an example of this implementation, preprocessing may include masking the content in the user dictionary contained within the text to be translated, based on a pre-built user dictionary. This user dictionary may contain technical terms, proper nouns, or other vocabulary that the user wishes to retain unchanged during translation. Furthermore, the user dictionary may record the mapping between user word numbers and user words. For example, words in the text to be translated that are already in the user dictionary can be replaced with special markers (such as __XX__), where XX is the user word number in the user dictionary. By using this example, the translation model can ignore words in the user dictionary during translation, thus reducing the impact on the overall fluency and accuracy of the translation. Especially for highly technical documents, masked technical terms can avoid the impact of inconsistencies before and after translation or differences from technical terminology.

[0181] In one example, semantic analysis can be used to determine the frequency of use and contextual meaning of words in a target document, in order to identify important terms and common words in the domain to which the target document belongs, thereby determining user words in the user dictionary.

[0182] In one example, technical terms and industry-specific vocabulary can be collected from target documents or related field materials (such as technical manuals, industry reports, academic papers, etc.) to determine user words in the user dictionary.

[0183] In one example, a list of abbreviations and terms could be created in the user dictionary, which could be provided by domain experts or extracted from the company’s internal documents.

[0184] In one example, the user dictionary can be adjusted and expanded based on user feedback and query habits. For instance, words that users frequently query during interactions with the system can be added to the user dictionary.

[0185] In one possible implementation, translating the preprocessed content to be translated includes: classifying the preprocessed content to be translated into complexity; and translating the preprocessed content to be translated based on the complexity classification result.

[0186] In this implementation, the preprocessed content to be translated can be categorized by complexity during the translation process. The purpose of the complexity categorization process is to determine the translation difficulty of the content, such as whether it contains complex language structures, technical terms, or expressions that require a deep understanding of the context.

[0187] As an example of this implementation, the complexity classification result can be classified as complex translation, non-complex translation, or no translation. For example, content to be translated that belongs to the complex translation category can be relatively long and complex, such as articles, paragraphs, or complete sentences. Content to be translated that belongs to the non-complex translation category can be short content fragments, such as phrases, titles, headings, and incomplete sentences.

[0188] In this implementation, classifying the complexity of the preprocessed content to be translated helps in selecting a suitable translation model.

[0189] In one possible implementation, translating the preprocessed content to be translated based on the complexity classification result includes: translating the preprocessed content to be translated using a translator based on a large language model in response to the complexity classification result being a complex translation class; or translating the preprocessed content to be translated using a translator based on a Transformer model in response to the complexity classification result being a non-complex translation class.

[0190] In this implementation, if the content to be translated is classified as complex, it may contain complex grammatical structures, specialized terminology, or require a deep understanding of the context. For content belonging to the complex translation category, a translator based on a large language model can be used. Large language models (such as BERT, GPT, etc.) are deep learning models trained on massive amounts of text data, capable of understanding and generating natural language text. Due to their sheer size and deep language understanding capabilities, large language models are well-suited for handling complex translation tasks.

[0191] If the content to be translated is classified as "non-complex translation," it is relatively simple and does not contain many technical terms or complex language structures. For content belonging to the non-complex translation category, a Transformer-based translator can be used. The Transformer model is an efficient neural network architecture, particularly suitable for handling sequence-to-sequence tasks, such as machine translation. Through its self-attention mechanism, it can capture the dependencies between any two positions in the input sequence, making it suitable for handling short text fragments and relatively simple translation tasks.

[0192] Additionally, content that is not subject to translation can be output directly without any translation processing. The translator's translation results and non-translated content can be combined to form the translated text, which serves as the output of the translation stage.

[0193] The above implementation method can balance the efficiency and accuracy of the processor performing the translation task. Especially when computing power is limited, it can rationally allocate resources, avoid wasting resources on content that does not need to be translated, and reasonably allocate the content to be translated according to the complexity of the translation.

[0194] In one possible implementation, the method further includes: post-processing based on the translated text; wherein the post-processing includes at least one of the following: unmasking masked words in the translated text according to a pre-built user dictionary; in response to the content to be translated being structured text, backfilling the translated text into its original position in the target document; in response to the content to be translated being non-text content, performing adaptive processing according to the content type of the content to be translated, wherein the adaptive processing includes at least one of the following: image synthesis, text-to-audio synthesis.

[0195] As an example of this implementation, post-processing is not required for plain text.

[0196] As an example of this implementation, masked words in the translated text can be demasked based on the mapping relationship between user word numbers and user words in a pre-built user dictionary. For instance, after translation, the translated text can be traversed to find special markers for all masked words. For each found marker, the corresponding actual word can be found based on the mapping relationship in the user dictionary. Then, the special marker can be replaced with the found actual word, thus completing the demasking.

[0197] As an example of this implementation, in response to the content to be translated being structured text, the translated text can be backfilled into its original position in the target document. By using this example, the layout and style of the translated content can be kept consistent with the original text.

[0198] As an example of this implementation, adaptive processing can be performed based on the content type of the content to be translated, responding to whether the content to be translated is non-textual (such as static images, animated images (such as GIF images), audio, or video). This adaptive processing includes at least one of the following: image synthesis and text-to-audio synthesis. For example, for static images, image synthesis can be performed based on the translation text and location information in the metadata. For animated images, synthesis can be performed based on a series of static images. For audio, audio synthesis can be performed using a text-to-speech engine and voiceprints from the metadata. For video, post-processing techniques combining image synthesis and text-to-audio synthesis can be used.

[0199] In one possible implementation, the step of backfilling the translated text to its original position in the target document includes: generating a first pointer and a second pointer for any content block in the content to be translated, wherein the first pointer points to the beginning position of the content block and the second pointer points to the end position of the content block; and backfilling the translated content of the content block in the translated text to its original position in the target document according to the first pointer and the second pointer.

[0200] In this implementation, for any content block within the content to be translated, its exact location within the target document can be determined using two pointers. These two pointers can include a first pointer (start pointer) and a second pointer (end pointer). The first pointer points to the beginning of the content block, i.e., its starting point. The second pointer points to the end of the content block, i.e., its ending point. These two pointers define the scope of the content block within the target document.

[0201] After the content to be translated is translated into the target language, the first and second pointers can be used to determine where the translated text should be placed; that is, the translated text is backfilled. The backfilling process ensures that the translated text maintains the same format and layout as the original document. This is especially important for structured text, as it relies on specific formatting to maintain the presentation and organization of the content.

[0202] This implementation method improves the efficiency and accuracy of document translation because it reduces the need for manual adjustments to formatting and layout, and also reduces potential errors during the translation process. By employing this method, it ensures that the structured text remains consistent in its position before and after translation, thus preventing readers from becoming confused or unable to match the content.

[0203] In one possible implementation, the structured text is in HTML or PDF format.

[0204] HTML is a standard markup language used to create web pages and display information online. It consists of a series of elements (called tags) that tell the browser how to display content. HTML documents have a defined structure, including sections such as the head and body, as well as tags for formatting text, such as paragraph marks. ),title( <h1> arrive< / h1> <h6>) and list ( 、 wait).

[0205] PDF is a widely used file format for displaying documents in a way that is independent of application software and hardware platforms. PDF files can contain text, images, tables, charts, and other content, while retaining the original layout and formatting. PDF files can also contain hyperlinks, form fields, and interactive elements.

[0206] In this implementation, for HTML or PDF format text, the translated text can be backfilled into the corresponding position in the original document, maintaining the original structure and format of the target document, thereby helping to preserve the appearance and accuracy of the information in the target document.

[0207] Thus far, the implementation methods and technical effects of the translation function have been described in detail in the above examples of this application. In other examples, the auxiliary functions of this application may also include other implementation methods.

[0208] In one possible implementation, determining the requested target assisted reading function includes: in response to a user sending a user question in the assisted reading interface, determining that the requested target assisted reading function includes the user question reply function.

[0209] In this implementation, users can input questions into the assisted reading interface and send them to the system. The questions can be any inquiries about the content of the target document, such as operating procedures, safety guidelines, or terminology explanations. The system can activate the target assisted reading function based on the user's question and generate an answer by retrieving relevant information from the target document, using a built-in knowledge base, or applying natural language processing techniques.

[0210] In one possible implementation, the target assisted reading function includes the user question reply function, and obtaining the assisted reading information corresponding to the target assisted reading function includes: retrieving reference information related to the user question addressed by the assisted reading request from the target document; generating assisted reading information based on the reference information, wherein the assisted reading information includes a jump link, the jump link being used to jump to the location of the reference information in the target document.

[0211] In this implementation, for the target document, a unique location identifier can be created for each part of the document based on embedding and position encoding. This allows the system to retrieve and locate specific content within the target document based on the user's question, providing relevant reference information. In natural language processing, embedding is a technique that converts text into vector representations, while position encoding adds positional information to these vectors, ensuring that the position of each word or sentence within the document can be identified and retrieved.

[0212] In this implementation, when a user asks a question, the system can retrieve relevant embedding vectors from the target document. The system can then sort the extracted embedding vectors according to their confidence level; higher confidence indicates a greater relevance between the content and the user's question. As an example, the system can perform a weighted sum of high-confidence embedding vectors at the page level and provide the user with a link to the page with the highest final value.

[0213] As an example of this implementation, after the user clicks the jump link to jump to the location of the reference information, they can also make suggestions for modification or correction requests to the target document through the feedback function control.

[0214] This implementation effectively understands and responds to user questions, providing precise in-document links to help users quickly locate and obtain relevant information. This not only improves the efficiency of information retrieval but also enhances the readability and usability of the target document.

[0215] In one possible implementation, the large language model generates supplementary reading information based on a retrieval-augmented generation (RAG) strategy. RAG is a hybrid approach combining retrieval and generation. It first uses a retrieval step to search for and identify information relevant to the user's question, and then uses this information to assist the generation step in producing a more accurate and relevant response.

[0216] In one possible implementation, three types of supplementary reading information can be provided in response to user questions: response information generated by a large language model based on a retrieval-enhanced generation strategy; jump links related to the user question; and key information in the target document (such as security notices and operating procedures).

[0217] In one example, users can manually type their questions or problems about a target document into input boxes on the assisted reading interface. The system can use technologies such as natural language processing to identify the key elements and intent of the user's question. Based on its understanding of the question, the system can retrieve relevant information from the target document or knowledge base to find possible answers. This may include searching specific chapters, paragraphs, or sentences within the target document. If the system finds information relevant to the question, it can provide an answer and recommend relevant pages within the target document. For example, it can provide the user with a direct text answer and / or a link to a specific location within the target document. If the system does not find a relevant answer during the search, it can suggest that the user use the feedback function to provide further feedback. This feedback function can be a built-in feedback system that allows users to report unresolved questions or provide additional context so that the system can improve its answer or human intervention can be initiated. User feedback can be used to improve system performance, such as enhancing the knowledge base, optimizing the retrieval algorithm, or adjusting the natural language processing model.

[0218] The following describes the assisted reading method for a target document provided in this disclosure through a specific application scenario. In this scenario, assisted reading controls can be displayed on the target document's display interface. The assisted reading interface can be displayed in response to a user triggering the assisted reading controls. Furthermore, in response to a user triggering the assisted reading controls, assisted reading requests corresponding to the automatic summary function of the document content and the key information reminder function can be generated, and the summary and key information of the target document can be displayed through the assisted reading interface. The key information of the target document may include security notifications and / or operation steps.

[0219] The assisted reading interface may include summary control, reminder control, translation control, and feedback control. The summary control generates assisted reading requests for automatically summarizing document content; the reminder control generates assisted reading requests for reminding users of key information; the translation control generates assisted reading requests for translating document content; and the feedback control receives user-inputted feedback information regarding the assisted reading information or the target document.

[0220] In addition, the auxiliary reading interface can also include an input box, through which users can enter questions.

[0221] It is understood that the various method embodiments mentioned above in this disclosure can be combined with each other to form combined embodiments without violating the principle and logic. Due to space limitations, this disclosure will not elaborate further. Those skilled in the art will understand that in the above methods of specific implementation, the specific execution order of each step should be determined by its function and possible internal logic.

[0222] In addition, this disclosure also provides an auxiliary reading device, electronic device, computer-readable storage medium, and computer program product for target documents. All of the above can be used to implement any of the auxiliary reading methods for target documents provided in this disclosure. The corresponding technical solutions and technical effects can be found in the relevant descriptions in the method section, and will not be repeated here.

[0223] Figure 5 This diagram illustrates a block diagram of an aided reading device for a target document provided in an embodiment of this disclosure. For example... Figure 5 As shown, the auxiliary reading device for the target document includes:

[0224] The determining module 51 is configured to determine the requested target auxiliary reading function in response to the auxiliary reading request for the target document, wherein the target auxiliary reading function includes at least one of the following: automatic summary function of document content, reminder function of key information, translation function of document content, and reply function to user questions;

[0225] The obtaining module 52 is used to obtain the auxiliary reading information corresponding to the target auxiliary reading function;

[0226] Output module 53 is used to output the auxiliary reading information.

[0227] In one possible implementation, the target-assisted reading function includes the translation function, and the obtaining module 52 is used for:

[0228] Identify the content to be translated in the target document;

[0229] The content to be translated is preprocessed, wherein the preprocessing includes extracting text from the content to be translated;

[0230] The preprocessed content to be translated is translated to obtain the translated text of the content to be translated.

[0231] In one possible implementation, the preprocessing further includes at least one of the following:

[0232] Filter out content that does not contain text from the content to be translated;

[0233] Extract the metadata of the content to be translated;

[0234] Based on a pre-built user dictionary, the content contained in the user dictionary within the text of the content to be translated is masked.

[0235] In one possible implementation, if the content to be translated is audio or video, the metadata of the content to be translated includes a timestamp and a voiceprint.

[0236] In one possible implementation, the obtaining module 52 is used for:

[0237] Classify the complexity of the preprocessed content to be translated;

[0238] Based on the complexity classification results, the preprocessed content to be translated is translated.

[0239] In one possible implementation, the obtaining module 52 is used for:

[0240] In response to the complexity classification result being classified as complex translation, the preprocessed content to be translated is translated using a translator based on a large language model.

[0241] or,

[0242] In response to the complexity classification result being classified as non-complex translation, the preprocessed content to be translated is translated using a Transformer-based translator.

[0243] In one possible implementation, the device further includes:

[0244] The post-processing module is used to perform post-processing based on the translated text;

[0245] The post-processing includes at least one of the following:

[0246] Based on a pre-built user dictionary, the masked words in the translated text are demasked;

[0247] In response to the fact that the content to be translated is structured text, the translated text is backfilled into the original position in the target document;

[0248] In response to the fact that the content to be translated is non-text content, adaptive processing is performed according to the content type of the content to be translated, wherein the adaptive processing includes at least one of the following: image synthesis, text synthesis of audio.

[0249] In one possible implementation, the post-processing module is used for:

[0250] For any content block in the content to be translated, a first pointer and a second pointer are generated, wherein the first pointer points to the beginning position of the content block and the second pointer points to the end position of the content block.

[0251] Based on the first pointer and the second pointer, the translated content of the content block in the translated text is backfilled into the original position of the content block in the target document.

[0252] In one possible implementation, the structured text is in HTML or PDF format.

[0253] In one possible implementation, the device further includes:

[0254] The display module is used to display auxiliary reading controls on the display interface of the target document;

[0255] The display module is used to display an auxiliary reading interface in response to the user triggering the auxiliary reading control, wherein the auxiliary reading interface is used to provide the auxiliary reading information.

[0256] In one possible implementation, the assisted reading interface includes at least one of the following functional controls:

[0257] The summary function control is used to generate an auxiliary reading request corresponding to the automatic summary function of the document content;

[0258] A reminder function control is used to generate an auxiliary reading request corresponding to the reminder function for the key information;

[0259] A translation function control is used to generate an auxiliary reading request corresponding to the translation function of the document content;

[0260] Feedback control is used to obtain feedback information from user input regarding auxiliary reading information or target documents.

[0261] In one possible implementation, the determining module 51 is used to:

[0262] In response to the user triggering the auxiliary reading control to generate an auxiliary reading request, the requested target auxiliary reading function is determined to include the automatic summary function of the document content and the reminder function of the key information.

[0263] In one possible implementation, the auxiliary reading information corresponding to the key information reminder function includes at least one of the following:

[0264] Security notifications are used to remind users of security precautions in the target document.

[0265] Operating procedures are used to guide users to correctly install or operate the device according to the steps specified in the target document.

[0266] In one possible implementation, the determining module 51 is used to:

[0267] In response to a user sending a question in the assisted reading interface, it is determined that the requested target assisted reading function includes the user question reply function.

[0268] In one possible implementation, the target assisted reading function includes the user question reply function, and the obtaining module 52 is used for:

[0269] Retrieve reference information from the target document that relates to the user question addressed by the assisted reading request;

[0270] Based on the reference information, auxiliary reading information is generated, wherein the auxiliary reading information includes a jump link, which is used to jump to the location of the reference information in the target document.

[0271] In one possible implementation, the target document includes a relevant technical manual for the medical device.

[0272] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the methods described in the above method embodiments. The specific implementation and technical effects can be referred to the description of the above method embodiments. For the sake of brevity, they will not be repeated here.

[0273] This disclosure also provides a computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, implement the above-described method. The computer-readable storage medium may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium.

[0274] This disclosure also proposes a computer program including computer-readable code, wherein when the computer-readable code is run in an electronic device, a processor in the electronic device executes the above-described method.

[0275] This disclosure also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in an electronic device, the processor in the electronic device executes the above-described method.

[0276] This disclosure also provides an electronic device, including: one or more processors; a memory for storing executable instructions; wherein the one or more processors are configured to invoke the executable instructions stored in the memory to perform the above-described method.

[0277] Electronic devices can be provided as terminals, servers, or other forms of devices.

[0278] Figure 6 A block diagram of an electronic device 1900 provided according to an embodiment of this disclosure is shown. For example, the electronic device 1900 may be provided as a server or a terminal. (Refer to...) Figure 6 The electronic device 1900 includes a processing component 1922, which further includes one or more processors, and memory resources represented by memory 1932 for storing instructions, such as application programs, that can be executed by the processing component 1922. The application programs stored in memory 1932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1922 is configured to execute instructions to perform the methods described above.

[0279] Electronic device 1900 may also include a power supply component 1926 configured to perform power management of electronic device 1900, a wired or wireless network interface 1950 configured to connect electronic device 1900 to a network, and an input / output interface 1958 (I / O interface). Electronic device 1900 can operate on an operating system stored in memory 1932, such as Microsoft Server operating system (Windows Server). TM Apple's graphical user interface-based operating system (MacOS X) TM ), a multi-user, multi-process computer operating system (Unix) TM Linux is a free and open-source Unix-like operating system. TM ), the open-source Unix-like operating system (FreeBSD) TM (or similar.)

[0280] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions that can be executed by a processing component 1922 of an electronic device 1900 to perform the above-described method.

[0281] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0282] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0283] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0284] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as "C" or similar languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions to implement various aspects of this disclosure.

[0285] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0286] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0287] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0288] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0289] The computer program product can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0290] The description of the various embodiments above tends to emphasize the differences between the various embodiments. The similarities or similarities between them can be referred to, and for the sake of brevity, they will not be repeated here.

[0291] If the technical solution of this disclosure involves personal information, the product applying the technical solution of this disclosure has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If the technical solution of this disclosure involves sensitive personal information, the product applying the technical solution of this disclosure has obtained the user's separate consent before processing the sensitive personal information, and also meets the requirement of "express consent". For example, at personal information collection devices such as cameras, clear and prominent signs are set up to indicate that the user has entered the scope of personal information collection and that personal information will be collected. If the user voluntarily enters the collection scope, it is deemed to have consented to the collection of their personal information; or on the personal information processing device, with clear signs / information informing the user of the personal information processing rules, authorization is obtained from the user through pop-up information or by asking the user to upload their personal information; wherein, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the types of personal information processed.

[0292] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein. < / h6>

Claims

1. A method for assisting in reading a target document, characterized in that, include: In response to an assisted reading request for the target document, the requested target assisted reading function is determined, wherein the target assisted reading function includes at least one of the following: automatic summary function of document content, reminder function of key information, translation function of document content, and reply function to user questions; Obtain the auxiliary reading information corresponding to the target auxiliary reading function; Output the aforementioned auxiliary reading information.

2. The method according to claim 1, characterized in that, The target assisted reading function includes the translation function, and obtaining the assisted reading information corresponding to the target assisted reading function includes: Identify the content to be translated in the target document; The content to be translated is preprocessed, wherein the preprocessing includes extracting text from the content to be translated; The preprocessed content to be translated is translated to obtain the translated text of the content to be translated.

3. The method according to claim 2, characterized in that, The preprocessing also includes at least one of the following: Filter out content that does not contain text from the content to be translated; Extract the metadata of the content to be translated; Based on a pre-built user dictionary, the content contained in the user dictionary within the text of the content to be translated is masked.

4. The method according to claim 3, characterized in that, When the content to be translated is audio or video, the metadata of the content to be translated includes a timestamp and a voiceprint.

5. The method according to claim 2, characterized in that, The translation of the preprocessed content to be translated includes: Classify the complexity of the preprocessed content to be translated; Based on the complexity classification results, the preprocessed content to be translated is translated.

6. The method according to claim 5, characterized in that, The process of translating the preprocessed content based on the complexity classification results includes: In response to the complexity classification result being classified as complex translation, the preprocessed content to be translated is translated using a translator based on a large language model. or, In response to the complexity classification result being classified as non-complex translation, the preprocessed content to be translated is translated using a Transformer-based translator.

7. The method according to claim 2, characterized in that, The method further includes: Post-processing is performed based on the translated text; The post-processing includes at least one of the following: Based on a pre-built user dictionary, the masked words in the translated text are demasked; In response to the fact that the content to be translated is structured text, the translated text is backfilled into the original position in the target document; In response to the fact that the content to be translated is non-text content, adaptive processing is performed according to the content type of the content to be translated, wherein the adaptive processing includes at least one of the following: image synthesis, text synthesis of audio.

8. The method according to claim 7, characterized in that, The step of filling the translated text back into its original position in the target document includes: For any content block in the content to be translated, a first pointer and a second pointer are generated, wherein the first pointer points to the beginning position of the content block and the second pointer points to the end position of the content block. Based on the first pointer and the second pointer, the translated content of the content block in the translated text is backfilled into the original position of the content block in the target document.

9. The method according to claim 7, characterized in that, The structured text is in HTML or PDF format.

10. The method according to any one of claims 1 to 9, characterized in that, The method further includes: On the display interface of the target document, auxiliary reading controls are displayed; In response to the user triggering the auxiliary reading control, an auxiliary reading interface is displayed, wherein the auxiliary reading interface is used to provide the auxiliary reading information.

11. The method according to claim 10, characterized in that, The auxiliary reading interface includes at least one of the following functional controls: The summary function control is used to generate an auxiliary reading request corresponding to the automatic summary function of the document content; A reminder function control is used to generate an auxiliary reading request corresponding to the reminder function for the key information; A translation function control is used to generate an auxiliary reading request corresponding to the translation function of the document content; Feedback control is used to obtain feedback information from user input regarding auxiliary reading information or target documents.

12. The method according to any one of claims 1 to 9, characterized in that, The determination of the requested target assisted reading function includes: In response to the user triggering the auxiliary reading control to generate an auxiliary reading request, the requested target auxiliary reading function is determined to include the automatic summary function of the document content and the reminder function of the key information.

13. The method according to any one of claims 1 to 9, characterized in that, The supplementary reading information corresponding to the key information reminder function includes at least one of the following: Security notifications are used to remind users of security precautions in the target document. Operating procedures are used to guide users to correctly install or operate the device according to the steps specified in the target document.

14. The method according to claim 1, characterized in that, The determination of the requested target assisted reading function includes: In response to a user sending a question in the assisted reading interface, it is determined that the requested target assisted reading function includes the user question reply function.

15. The method according to claim 1 or 14, characterized in that, The target assisted reading function includes the user question reply function, and obtaining the assisted reading information corresponding to the target assisted reading function includes: Retrieve reference information from the target document that relates to the user question addressed by the assisted reading request; Based on the reference information, auxiliary reading information is generated, wherein the auxiliary reading information includes a jump link, which is used to jump to the location of the reference information in the target document.

16. The method according to any one of claims 1 to 9, characterized in that, The target documents include relevant technical manuals for medical devices.

17. An electronic device, characterized in that, include: One or more processors; Memory used to store executable instructions; The one or more processors are configured to invoke executable instructions stored in the memory to perform the method according to any one of claims 1 to 16.

18. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 16.

Images