13 results about "Readability" patented technology

This disclosure proposes a dialogue processing method, a training method for a question rewriting model, and a device thereof. The dialogue processing method includes: when information is missing in the user question, inputting the user question and historical dialogue content into the encoding layer of the question rewriting model to obtain the semantic representation vectors of each word in the historical dialogue content and the semantic representation vectors of each word in the user question, as well as the keyword scores of each word in the historical dialogue content; then, the decoding layer accurately selects keywords semantically related to the user question from the historical dialogue content based on the semantic representation vectors of each word in the user question, the semantic representation vectors of each word in the historical dialogue content, and the keyword scores, and explicitly generates a fully rewritten user question based on these keywords. This accurately supplements the missing information in the user question, achieving completeness and readability of the user question content and improving the user experience.

This invention relates to the field of artificial intelligence technology and discloses a method and system for optimizing and generating audio and video subtitles for less commonly spoken languages. The method includes the following steps: Step 1: Audio extraction and preprocessing; Step 2: Language recognition for the less commonly spoken language; Step 3: Language-aware punctuation restoration module; Step 4: Subtitle readability-driven segmentation module; Step 5: Subtitle format encapsulation and output. This application establishes a complete technical process encompassing audio preprocessing, language-aware speech recognition, structured text restoration, subtitle segmentation optimization, and time alignment. This method significantly improves the accuracy of speech-to-text conversion in less commonly spoken language videos and enhances the structural integrity and readability of subtitles. It is particularly suitable for practical application scenarios involving less commonly spoken languages ​​in regions such as Southeast Asia, where data is scarce and languages ​​are diverse. It has broad application value and practical significance in media dissemination, educational videos, and government services.

Apparatus and associated methods relate to automatically generating secured artificial intelligence input packages from sensitive data input. In an illustrative example, a sensitive questions answering system (SQAS) may receive a natural language input (NLI). For example, the SQAS may retrieve sensitive data related to the NLI. In some implementations, the SQAS may be configured to desensitize the sensitive data and / or generate a secured user input and media objects to be combined into a secured LLM (large language model) input array. For example, the secured LLM input array may be provided to an unsecured LLM to generate a human-readable response (HRR). In some implementations, a sensitized response may be generated by reintroducing some of the sensitive data to the HRR. For example, the sensitized response may include personal data. Various embodiments may advantageously ensure the secure handling of sensitive data while providing personal responses to user queries.

The application provides a machine simultaneous interpretation system, method, test method, device and related equipment. The system and method provide a multi-modal output function, which can include at least one of subtitle output of translated text, playing of synthesized audio stream, and video playing of audio stream and virtual person synthesis. The application considers multiple different factors affecting the quality of simultaneous interpretation, such as translation effect measured from BLEU score, fidelity, fluency, translation real-time between source language audio input and translated text output / synthesized audio output, readability represented by average erasure rate of translated text, speech synthesis effect, etc. Specific requirements are proposed for each influencing factor during the process of simultaneous interpretation, so that the overall quality of the machine simultaneous interpretation system is more balanced and comprehensive. The test method of the machine simultaneous interpretation system of the application can complete comprehensive testing of each technical index of the simultaneous interpretation system, and ensure the quality of the simultaneous interpretation system.

The application discloses a method for improving the reading and writing success rate of RFID printer labels, and belongs to the field of radio frequency identification technology. The method comprises a calibration stage and a reading and writing stage. The calibration stage comprises: moving an RFID sample label and performing step-by-step radio frequency scanning in the length direction of the RFID sample label, and recording the readability state of each discrete position point; identifying a set composed of adjacent readable points based on the state; determining a readable interval based on the length range covered by each set; and determining at least one reading and writing point in each readable interval, so that the total number of the cumulative determined reading and writing points is not less than two. The reading and writing stage comprises: sequentially driving target RFID labels of the same type to each reading and writing point and attempting to read and write, until a reading and writing attempt succeeds; and determining that the label reading and writing fails only when all attempts fail. The application determines multiple reading and writing points in the calibration stage, introduces a sequential fault tolerance mechanism in the reading and writing stage, improves the reading and writing success rate of the labels, and reduces the label loss.

A system for automatically assessing sentiment and readability of sentences in one or more documents is disclosed. The system applies a sentiment mismatch analysis and generates sentence rewrites based thereon. The system applies readability analysis and generates sentence rewrites based thereon. The system generates combined sentence rewrites based on the readability rewrites and the sentiment rewrites, and generates an updated output document based thereon. In some embodiments, the system intercepts and automatically halts publication of documents pending analysis for satisfaction of readability and sentiment criteria. In some embodiments, the readability and sentiment analyses are configured in accordance with user inputs and / or with information regarding third-party AI analyses to which the documents are expected to be subjected following publication.

This application provides a watermarking method, an electronic device, and a computer-readable storage medium, relating to the field of digital watermarking technology. The method includes: acquiring a file to be processed and a first watermark key; parsing the file to be processed to extract target text lines containing English characters; adjusting the text layout parameters of the target text lines according to the first watermark key to obtain adjusted layout parameters, the adjusted layout parameters including the target width of each English character in the target text line; and generating a target file with watermark information based on the adjusted layout parameters of the target text line, the watermark information being represented by the target width of each English character in the target text line. This application achieves invisible watermark embedding by modifying the width of English characters in the file to be processed, subtly encoding the watermark information into the character width, thus not affecting the readability and format compatibility of the file, while achieving an efficient and covert copyright protection mechanism.

Computer-based processes are disclosed for analyzing and improving document readability. Document readability is improved by using rules and associated logic to automatically detect various types of writing problems and to make and / or suggest edits for eliminating such problems. Many of the rules seek to generate more concise formulations of the analyzed sentences, such as by eliminating unnecessary words, rearranging words and phrases, and making various other types of edits. Proposed edits can be conveyed, e.g., through a word processing platform, by changing the visual appearance of text to indicate how the text would appear with (or with and without) the edit.

This invention discloses an adaptive text watermarking method based on dynamic red-green lists, belonging to the fields of natural language processing and artificial intelligence security. Specifically, the method works as follows: First, the user inputs prompts into a large language model, generating text A. Then, the top-k candidate words and red-green lists for each position in the text are obtained. Next, each candidate word is checked to see if it overlaps with the green list. If so, the word with the highest probability of overlapping with the green list is selected from the top-k candidate words as the output. Otherwise, if all top-k candidate words are in the red list, an exemption rule is activated, and the top-1 word is directly selected as the equivalent green list output. The watermarking process ends. Finally, the proportion of equivalent green list words is statistically analyzed to obtain a statistic, verifying whether the text contains a watermark. This invention maintains a high watermark detection rate while ensuring the fluency and readability of the generated text.

This invention provides a text generation method, apparatus, electronic device, and storage medium, relating to the field of artificial intelligence technology. The text generation method includes: acquiring target information input by a user; acquiring auxiliary information based on the target information; generating initial text based on the target information and the auxiliary information; performing predetermined analysis processing on the initial text to obtain modification suggestions for the initial text; the predetermined analysis processing is used to determine the similarity between the content information of the target type represented by the initial text and the target information; analyzing the modification suggestions for the initial text based on the determined similarity; the modification suggestions are used to make the content information of the target type represented by the initial text conform to the target information; and modifying the content of the initial text based on the obtained modification suggestions to obtain target text generated for the user. Therefore, this solution can balance the readability and richness of the generated text.

This invention provides a text summarization extraction method, apparatus, device, and storage medium, comprising: acquiring text information to be processed; inputting the text information to be processed into a summarization generation model to obtain a summarization generation result output by the summarization generation model; if the summarization generation result does not meet preset quality assessment conditions, performing syntactic analysis and extraction on the text information to be processed to obtain a summarization extraction result, and using the summarization extraction result as the final summarization result. This invention generates a summarization result corresponding to the text information to be processed through a summarization generation model. If the summarization generation result does not meet preset quality assessment conditions, syntactic analysis and extraction are performed on the text information to be processed, improving the correlation between sentences in the summary, thereby solving the technical problem of low accuracy and readability of text summarization, and improving the accuracy and readability of summary extraction.

The present invention relates to a computing system for classifying readability of Chinese content into a plurality of difficulty levels based on Hong Kong local linguistics, comprising: an input module for retrieving one or more corpora comprising Traditional Chinese and Cantonese vocabulary; a pre-processing unit for generating four levels of linguistic features for each document in the one or more corpora, wherein the four levels of linguistic features comprise: character (C), word (W), syntax (S) and discourse (D) levels; and a local processing unit having a local artificial intelligence (AI) engine for training two machine learning algorithms to classify readability based on the four levels of linguistic features, wherein the two machine learning algorithms comprise a random forest (RF) algorithm and a support vector machine (SVM); wherein the difficulty levels have 24 difficulty levels based on 12 grades, and each grade comprises 2 form levels according to the Hong Kong education system.

The application discloses a processing method and device of insurance exclusion clauses and electronic equipment. The method comprises the following steps: pre-processing an insurance contract text to obtain a candidate text segment including an exclusion clause; identifying the exclusion clause in the candidate text segment, and generating a preliminary merged text according to the exclusion clause; performing compliance verification and readability optimization on the preliminary merged text, and generating a spoken explanation script and a synchronous subtitle text; and pushing the spoken explanation script and the synchronous subtitle text to a double-recording device to realize real-time synchronization of the spoken explanation script, the synchronous subtitle text and the double-recording function. The application can avoid problems such as missing key exclusion elements and inaccurate expression, is convenient for policyholders to understand, reduces compliance risks, does not depend on the professional ability of sales personnel, and improves the standardization degree.