45 results about "Timed text" patented technology

The invention discloses an AI digital human conference proxy method and device under an offline local area network and a medium, and the method comprises the steps: collecting conference voice in real time, converting the conference voice into a real-time text, and pre-judging a subject set in combination with a localized industry knowledge base and a user historical conference track; if it is detected that the user or the to-be-decided item is mentioned, reply voice is generated in combination with historical corpora of the user; and if the conference enters the pre-judgment topic set, calling the pre-loaded user feature packet to generate reply voice, and controlling the digital person to generate a corresponding audio and video stream. The invention provides an AI digital human conference proxy method and device under an offline local area network and a medium, and aims to realize topic pre-judgment in combination with a local knowledge base and a historical conference track of a user and provide preparation for real-time reply; meanwhile, for different scenes, the user historical corpus and the user feature packet are called respectively to generate the reply, so that the problem that the real-time personalized reply generation of the AI digital person conference agency and the conference issue pre-judgment are difficult to collaboratively realize in an offline local area network environment can be solved.

This invention discloses a long text matching method combining noise filtering and a divide-and-conquer strategy. The method includes: constructing a long text matching model, which comprises a keyword extraction layer, an association extraction layer, and a filtering layer. The keyword extraction layer extracts keywords from the text; the filtering layer filters noise from the text based on sentence similarity to obtain a denoised text sequence; and the association extraction layer further removes keywords from the denoised text sequence to obtain the remaining associated text. The long text matching model is trained with the optimization objective of minimizing a set overall loss function, which reflects the global matching distribution and combines the keyword and association matching distributions. For the target text, real-time text matching is performed using the trained long text matching model. This invention improves the generalization ability and accuracy of text matching.

The application relates to the technical field of natural language processing, financial technology and medical technology, and discloses a standard time conversion method and device based on natural language, equipment and a medium, which comprises the following steps: collecting current session text information in real time; when target text representing a time attribute appears in the text, acquiring corresponding collection time and standardizing the collection time into a specified format reference time; extracting preset time period text containing the target text as context information; reasoning the target text and the context by combining the reference time with a preset natural language model to determine a target date, a target time word and a target number of days; and performing consistency verification on the target time word and the target number of days, and outputting the target date if the verification is passed. The application can be applied to a data processing platform in the fields of financial technology and medical health, improves the standardization conversion efficiency of time text strings, improves the accuracy of large models, and improves the convenience of the corresponding platform.

This invention belongs to the field of text printing technology and discloses a text printing method and system based on speech recognition. The method includes the following steps: constructing a text printing template database; constructing a mixed-language speech recognition model; acquiring speech audio data in real time and performing speech recognition; matching several corresponding text printing templates; selecting a text printing template; fusing the real-time speech text data with the selected text printing template; and printing the real-time text printing data. The system includes a database construction unit, a model construction unit, a storage unit, a speech audio acquisition unit, a speech recognition application unit, a template matching unit, a template selection unit, a data fusion unit, and a printer. This invention solves the problems of low intelligence, poor speech recognition effect, low recognition efficiency, and lack of organic integration in existing technologies.

The invention discloses a portable real-time text recognition and translation method and system based on a lightweight Transform, and belongs to the technical field of computer vision and natural language processing. The system comprises an image preprocessing module, a text recognition module, a text translation module, a result presentation module, a model optimization engine and a system control module. A lightweight Transform architecture is adopted, and the calculation complexity is reduced through a layered mixed attention mechanism, a shared weight design, dynamic depth adjustment and a compression feedforward network. The text recognition module adopts a CNN and Transform hybrid architecture to realize detection and recognition, and the text translation module shares partial encoder weight and supports incremental decoding. The system integrates compression technologies such as knowledge distillation, quantification and pruning, realizes end-side off-line real-time operation in combination with hardware perception optimization, supports multi-language image-text translation, and has the characteristics of high precision, low delay and low power consumption.

This invention relates to the field of artificial intelligence technology, and more particularly to a method, apparatus, computer device, and medium for generating incorrect sentences for model training. The method maps the number of terms in the text to be processed to an adjustment quantity, samples target terms from the text to be processed, adjusts the target terms to obtain augmented text, occludes the text to be processed, inputs it into a reconstruction model to obtain reconstructed text, trains a generative model based on the text to be processed, the augmented text, and the reconstructed text, inputs real-time text into the trained generative model to obtain generated incorrect sentence text, processes the text to be processed using multiple data augmentation methods to obtain text containing rich error types as labels, trains an end-to-end generative model, allowing real-time text to be directly input into the trained generative model to obtain generated incorrect sentence text without frequent switching of data augmentation methods, greatly improving the efficiency of incorrect sentence generation while ensuring high quality.

This specification provides a dialogue analysis system and method. The system includes: an acquisition module for acquiring a text sequence, wherein the text sequence is obtained by real-time text conversion of classroom dialogues; a matching module for searching, in a pre-established rule base, using a large language model, whether a target sequence matching the text sequence exists; wherein the rule base includes classification rule sequences for various dialogue types and / or reference text sequences corresponding to each dialogue type; and an analysis module for analyzing, based on the target sequence matched by the matching module, to obtain the target dialogue type corresponding to the classroom dialogue. This disclosure combines a rule base with a large language model to achieve usability and flexibility in classroom dialogue analysis, ensuring both deep integration of the dialogue analysis system with educational theory and the ability to process complex dialogue sequences through a large language model, achieving accurate dialogue classification and analysis.

The application discloses a kind of text image quality detection method, system, equipment and medium, the method includes: obtaining text image, text image is divided into training set and test set;Text image quality detection model is constructed, and training set is input into text image quality detection model for training;Test set is input into the text image quality detection model that has been trained, obtains detection result, and text image quality detection model is evaluated and optimized.The application detects the quality of text image using the text image quality detection model without reference image, does not need fixed reference image, is more suitable for practical application scene, is more flexible and practical, improves the detection precision of text image quality evaluation, and improves the accuracy and robustness of text image quality detection, realizes accurate and real-time text image quality evaluation.

A scalable, automated timed text workflow system designed to optimize the generation and refinement of time-synchronized textual content for video is disclosed. Integrated Machine Translation (MT) models and advanced AI-driven tools such as Computer Vision, Generative AI, and Traditional AI automatically generate and refine timed text, including subtitles, closed captions (CC), and SDH (Subtitles for the Deaf and Hard of Hearing). Human review is coordinated through a workflow orchestration system when needed, offering flexibility and scalability for handling high-volume media processing across various platforms and formats.

The application discloses a network information collection method and system based on real-time text analysis, comprising the following steps: acquiring real-time text data, preprocessing the real-time text data, acquiring first data and second data according to the preprocessed real-time text data, calculating the similarity of the first data and the similarity of the second data, weighting the similarity of the first data and the similarity of the second data to obtain a classification target, constructing a text classification model according to the classification target, inputting the real-time text data into the text classification model to obtain classification data, and outputting the classification data as collected network information. The method can not only improve the accuracy of network information collection, but also has good interpretability and can be directly applied to a network information collection system.

The present application relates to a memory-enhanced deep personalized input method system based on end-side large language model, and belongs to the technical field of artificial intelligence. The present application solves the problems of the prior art, such as the loose coupling of the model and the input core process, the shallow personalization mechanism, the delay and privacy risk caused by cloud deployment, and the insufficient efficiency of the end-side operation that cannot meet the real-time input requirements. The system includes a front-end real-time interaction module, a stylized large model post-training module, a hierarchical memory mechanism module, and an end-side reasoning optimization module. The present application can realize low-delay real-time text generation under the condition of highly limited mobile terminal resources, and at the same time, through continuous modeling of user input behavior, it can complete deep personalization adaptation and ensure the end-side retention and privacy security of user input data throughout the process.

This invention provides a method for determining the placement location and type of a centralizer based on multi-source data fusion, comprising: Step 1, acquiring casing record table data before the casing operation begins; Step 2, acquiring real-time text data and real-time video data at the start of the casing operation; Step 3, determining the target casing information currently being lowered and the type of centralizer to be placed therein based on the real-time text data; Step 4, determining the type of centralizer actually placed on the target casing based on the real-time video data; Step 5, determining the specific time of centralizer placement on the target casing based on the real-time video data; Step 6, determining whether the corresponding type of centralizer was correctly placed on the target casing during the actual placement process. This method for determining the placement location and type of a centralizer based on multi-source data fusion standardizes the casing lowering operation process, effectively replaces traditional manual monitoring methods, and improves on-site safety and supervision efficiency.

The techniques disclosed herein provide enhanced controls for the display of real-time text (RTT) in calls and meetings. RTT is the ability for someone to send a text message on a character-by-character basis to everybody else in a call or meeting. The system disclosed herein integrates RTT, video, and live captions all in one central experience. This integrated experience allows users to participate equitably by making RTT accessible to users regardless of the operating mode they are in and still concurrently access other meeting content, including video streams, chat messages, live captions, transcripts, and artificial intelligence (AI) tools, such as Copilot. In one embodiment, during an online conference, in response to one of the attendees activating a RTT mode, when at least one user minimizes a meeting stage, such as for the purpose of multitasking while listening, the conference application maintains a display area for displaying RTT.

The invention discloses a real-time online text violation recognition method and system and a storage medium, and belongs to the technical field of text recognition. The method comprises the following steps: collecting user behavior data in real time, and generating a user-level weak tag; comparing and learning the user-level weak tags to obtain a single sentence code and a single sentence risk assessment model; automatically distributing a pseudo label and an attention weight for each sentence by utilizing a pseudo label distribution rule, and training to obtain a lightweight sentence-level model; for an input real-time text, calculating a violation probability through a lightweight sentence-level model to obtain a risk index; according to the risk indexes and a preset risk threshold value, performing grading processing, blocking users whose risk indexes are in a high risk interval, performing secondary reasoning and random sampling on the users whose risk indexes are in a middle and low risk area, and then performing manual auditing; and iteratively updating the user-level weak label according to a user disposal result and newly added report data, and realizing self-evolution of each model. A large number of user-level tags can be obtained at low cost, and the detection rate is improved.

The invention discloses a text processing method and device and a storage medium. Relates to the technical field of civil aviation information, and comprises the following steps: receiving a to-be-processed first text in a natural language form, and extracting a time text segment in the first text, the time text segment comprising time information; determining a time expression category of the time text segment, the time expression category being a category of an expression mode of the time text segment for the time information; processing the time text segment according to the time expression category to obtain a target time text; and replacing the time text segment in the first text with the target time text to obtain a second text. Through the method and the device, the problem of relatively low efficiency of processing the time text in related technologies is solved.

A method, apparatus, computer readable medium, and system for image processing include: 1105: obtaining an input image depicting a first element, a text description of the input image, and a modificat

The application provides an enterprise due diligence method, device, equipment and storage medium, the method comprises the following steps: obtaining the enterprise information of the client to be due diligence and converting into a label, matching the content of the speech from the historical speech database according to the label, and obtaining the personalized speech list; real-time speech recognition is performed on the due diligence recording collected in the due diligence field to obtain real-time text, the real-time text is subjected to keyword retrieval and comparison with the personalized speech list, respectively obtaining the violation items and the missing items, which are used for generating reminder information and sending to the due diligence personnel; after the due diligence process is completed, the due diligence recording is subjected to speech recognition to obtain the full-amount text, the key information is extracted from the full-amount text and filled into the report template to obtain the due diligence report; the position information of the due diligence field and the voiceprint characteristics of the due diligence recording are collected, the position information and the voiceprint characteristics are verified, and the due diligence report is labeled according to the verification result. The application solves the problem of weak due diligence authenticity verification capability in the prior art.

This invention belongs to the field of text information extraction technology and discloses a method and system for information extraction based on natural language text. The method includes the following steps: acquiring real-time natural language text and preprocessing it to obtain preprocessed real-time natural language text; using a text analysis model to analyze the preprocessed real-time natural language text to obtain corresponding real-time text analysis results; based on the real-time text analysis results, using a strategy generation model to generate strategies for the preprocessed real-time natural language text to obtain corresponding real-time information extraction strategies; and using an information extraction model to extract information from the preprocessed real-time natural language text according to the real-time information extraction strategies to obtain corresponding real-time key information. This invention solves the problems of slow processing speed, low accuracy, poor flexibility, and insufficient semantic understanding in existing technologies.

The techniques disclosed herein provide enhanced controls for the display of real-time text (RTT) in calls and meetings. RTT is the ability for someone to send a text message on a character-by-character basis to everybody else in a call or meeting. The system disclosed herein integrates RTT, video, and live captions all in one central experience. This integrated experience allows users to participate equitably by making RTT accessible to users regardless of the operating mode they are in and still concurrently access other meeting content, including video streams, chat messages, live captions, transcripts, and artificial intelligence (AI) tools, such as Copilot. In one embodiment, during an online conference, in response to one of the attendees activating a RTT mode, when at least one user minimizes a meeting stage, such as for the purpose of multitasking while listening, the conference application maintains a display area for displaying RTT.

The application discloses a method and device for realizing intelligent question and answer service based on a large model, the method comprises the following steps: acquiring real-time text data; inputting the real-time text data into a trained intelligent question and answer network model to perform corresponding answer prediction, and outputting multiple candidate answers according to the answer prediction result; acquiring scores corresponding to the multiple candidate answers, and outputting an optimal answer in the multiple candidate answers according to the sorting result of the scores. The application adds a personalized capability to the large model, solves the problem that the large model is seriously illusory and difficult to achieve ideal effects of problems by relying on knowledge base retrieval, and can effectively and accurately predict and output answers to problems in multiple fields.

The invention discloses a real-time text-to-speech conversion method without Internet connection, equipment and a storage medium. The method comprises the following steps: deploying and configuring a preset speech synthesis engine on local computing equipment; starting a local service to receive a voice synthesis request from a front-end application, and analyzing the voice synthesis request to obtain text content to be synthesized; calling the speech synthesis engine to process the to-be-synthesized text content locally, and generating a corresponding audio data stream; and returning the generated audio data stream to the front-end application, and playing the audio data stream. According to the invention, the preset speech synthesis engine is deployed and configured on the local computing device, so that the whole text-to-speech process can be completed locally without depending on external cloud service, and text-to-speech conversion can still be carried out normally even in an environment without network connection.

The invention discloses a real-time discourse correction method, a related device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a current recognition text of a real-time audio stream, and obtaining a plurality of historical recognition texts subjected to text correction before the current recognition text and historical correction texts obtained through text correction; based on at least one of the current recognition text and the historical correction text, retrieving in the scene knowledge graph to obtain a reference knowledge text related to the execution of the text correction; wherein the scene knowledge graph is a reference knowledge graph related to an interaction scene where the real-time audio stream is located; and performing text correction on the current recognition text based on the historical correction text and the reference knowledge text to obtain a current correction text of the current recognition text. According to the scheme, the accuracy of real-time discourse correction can be improved, especially when the real-time discourse relates to terminologies or specific domain knowledge.

Methods are provided for improved flexibility for subscribers who wish to communicate in real time using different media inputs and / or with more than one subscriber. An indication to convert the voice call session between a first UE and a second UE to a hybrid session may be received by a network function (NF). Based on the indication, the hybrid session is established. The hybrid session comprises a rich communication service (RCS) session and a real-time text (RTT) session. The NF receives a first message from the first UE via the RCS session and communicates a second message to the second UE via the RTT session.

The disclosed techniques provide enhanced control of visual indicators providing attribution to participants communicating via real-time text (RTT) in an online meeting. The disclosed techniques also provide dynamic control of text-to-speech conversion of RTT during online meetings. The display of RTT and the text-to-speech conversion for RTT allows users to hear and visualize content shared through RTT. A person's name and / or image is highlighted if that person is designated as having a prerequisite while they are communicating text in RTT mode. The system also acknowledges such users as an “active speaker” so the content shared in RTT is also recorded equitably in meeting records and used in AI models for generating transcripts, summaries, tasks, etc.

The application relates to the technical field of knowledge base retrieval, and discloses a knowledge base automatic updating and retrieval method and system, which comprises the following steps: dividing a document set into semantic atoms, storing a benchmark vector into a database and initializing a differential log queue; generating a text differential patch based on a modification request and writing the patch into a queue, extracting change characteristic words to update an inverted index; performing two-way retrieval of the benchmark vector and the inverted index based on a query request to generate a candidate set; applying the patch to the candidate set to build a real-time text view; calculating the text matching degree of the query and the view, combining a benchmark vector similarity to calculate a drift penalty term, and reordering and outputting the atoms according to a comprehensive ranking score containing the drift penalty term. Through the differential log and two-way retrieval mechanism, the application realizes the decoupling of the static stability of the vector index and the dynamic timeliness of the text, avoids the high-frequency recalculation of the vector, and ensures the real-time accuracy of the retrieval result through semantic drift calibration.

The invention provides a click rate estimation method and device, equipment and a medium, and relates to the technical field of click rate prediction. The method comprises the following steps: acquiring a request timestamp and an online timestamp to generate a current time slice feature, and calculating to generate a timeliness correction coefficient; and according to the brand keyword bound with the timeliness correction coefficient, obtaining a real-time text stream and performing sentiment analysis to generate a brand reputation health index. And comparing the brand reputation health index with a preset public opinion fusing threshold value, and generating a reputation dynamic intervention factor according to a comparison result. And performing operation according to the reputation dynamic intervention factor and the timeliness correction coefficient to generate a comprehensive marketing probability weight. And obtaining a basic click conversion rate value corresponding to the target advertisement position, and correcting the basic click conversion rate value by using the comprehensive marketing probability weight to generate a final click rate estimated value. And according to the final click rate estimated value, matching the advertisement bidding step table to generate a real-time bidding strategy amount.

This invention discloses a method, system, device, medium, and program product for real-time text-to-speech streaming conversion. The method includes: first, receiving a user input request, and then generating a text content block based on the user request; identifying the type of the text content block by monitoring the thinking state, and writing it into the corresponding inference queue and response queue. After the inference content in the inference queue meets a first preset condition, the inference content is synthesized using a first preset speech synthesis parameter, and the inference content and inference audio block are streamed using a unified data structure. When the response queue is first written with response content, the system switches to the response stage, and the response content is synthesized using a second preset speech synthesis parameter; after all the remaining inference content in the inference queue is output and cleared, the response content and response audio block are streamed using a unified data structure. This invention significantly reduces the initial packet latency between the response text and audio, and effectively improves the smoothness of switching between inference and response content.

This invention proposes an online handwritten Chinese text line recognition method based on real-time updating of local feature maps, comprising the following steps: Step S1, initializing the real-time recognition model; Step S2, when starting stroke input, taking the touch of the pen tip on the handwriting surface of the handwriting device as the start marker; Step S3, acquiring the trajectory of the online handwritten Chinese text line, recording the coordinate point sequence of the strokes, and taking the lifting of the pen as the end marker of a stroke; Step S4, extracting the local image corresponding to the newly input stroke and performing preprocessing; Step S5, calculating the CNN features of the local image corresponding to the newly input stroke, replacing the new features on the previous feature map, and realizing real-time updating of the local features of the new stroke; Step S6, using a language model to decode and update the recognition result, realizing real-time text line recognition. This invention significantly improves speed, has low image size dependence, strong scalability and robustness, and can be used in various resource-constrained embedded devices, possessing high application value.

The application provides a test method, device and equipment based on machine vision and real-time text recognition, which comprises the following steps: extracting words from a first image of sample text to obtain a sample dictionary through a first camera; obtaining a test video stream through the first camera, determining the start and end positions of a scanned word dictionary pen according to the test video stream; determining a sample character set from the sample dictionary according to the start and end positions, wherein the sample character set comprises at least one sample character; obtaining a scanned character set scanned by the second camera, wherein the scanned character set comprises at least one scanned character; and determining a target test result by comparing the scanned character set and the sample character set. According to the technical scheme of the embodiment of the application, sample information can be obtained through machine vision as a content reference for testing, and the test result can be automatically determined by comparing the characters scanned by the scanned word dictionary pen, so that manual checking is not needed, and the test efficiency and accuracy are improved.

The application discloses a kind of food material recommendation methods based on knowledge graph, comprising the following steps: obtaining real-time audio and video data, real-time text data and historical review text data;It is preprocessed, and obtain effective real-time voice, real-time text, real-time video and historical review text data;Transcription effective real-time voice and video data are voice text data and image text data;According to the voice text data, real-time text data and image text data, obtain real-time vector matrix;The historical vector matrix corresponding to the historical review text data is obtained;Based on the real-time vector matrix and historical vector matrix, real-time semantic similarity and historical semantic similarity are fused to generate food material recommendation list and output.The method effectively improves the accuracy of recommended food material content, and improves the effect of user.