203 results about "Text string" patented technology

The invention discloses a character recognition method and system based on a large model and an OCR technology, and relates to the technical field of character recognition, and the method comprises the steps: extracting picture information, carrying out the unified preprocessing, generating a text detection box of a character region through a DBNet lightweight text detection model, and obtaining a text position; quickly identifying characters in the textbox by using a lightweight OCR model to obtain text content, and generating an identification information group; carrying out average value calculation on the confidence coefficient of the lightweight OCR model result, and analyzing the overall confidence coefficient; and for the result with low confidence coefficient, inputting the corresponding identification information group into the multi-modal large model, and carrying out secondary identification. According to the method, the text position set and the text character string sequence are combined into the identification information group, so that the effect of structured storage of detection and identification results is achieved, subsequent information retrieval and multi-modal fusion analysis are facilitated, and the effect of optimizing the subsequent processing efficiency and precision is achieved through the combination of the confidence coefficient screening step.

Provided is a system and method for controlling a plurality of devices. The method includes generating a command script by processing a text string with at least one model, the text string including a natural language input by a user, modifying the command script based on contextual data, the command script including a configuration for at least one device, generating at least one command signal based on the command script, and controlling at least one device based on the at least one command signal.

A device may be configured to perform spatial extrapolation based on information included in a neural-network post-filter characteristics message. In one example, a neural-network post-filter characteristics message includes a syntax element indicating a purpose of the neural-network post-filter characteristics message is spatial extrapolation. In one example, a neural-network post-filter characteristics message includes a syntax element specifying a text string prompt used for generating contents of a spatial extrapolation image area.

In various embodiments, systems and methods for design-aware replacement font suggestions are provided. In some embodiments, a substitute font-suggestion algorithm holistically considers how the original string of text from the original layout appears when re-rendered in a same-sized text frame using a potential replacement font. In some embodiments, the substitute font-suggestion algorithm generates a first image of a text frame including the text string using the first font and generates a plurality of second images of the text string using candidate replacement fonts. A ranking of the candidate replacement fonts is generated based on computing a score for each of the individual second images that represents similarity between the first image and the individual second images. Based on the assessed similarities, a ranked listing of substitute font suggestions is displayed.

The invention discloses a hierarchical rhythm mapping-driven short video copywriting mood automatic adjustment method, and relates to the technical field of video processing, and the method comprises the steps: 1, receiving a text character string and a language type identifier, and building an occupation column for bearing a tone mark, an accent mark and a duration mark at each level; 2, dividing each sentence into phrase segments based on the hierarchical index table, freezing boundaries by taking the phrase segments as units, presetting sentence end termination styles according to punctuations, determining kernel phrases according to semantic anchor points, initializing trends of the kernel phrases, and performing time sequence elastic alignment and hierarchical backfilling to obtain a sentence end termination pattern; and finally outputting a triple sequence which covers all syllables and is composed of tone marks, accent marks and duration marks as a target rhythm control sequence. And step 3, performing audio generation based on the target rhythm control sequence to obtain new dubbing. According to the method, the tone accuracy and expressive force of short video dubbing are improved, and the time and cost of manual adjustment are remarkably reduced.

A method may include obtaining first audio data of a first communication session between a first and second device and during the first communication session, obtaining a first text string that is a transcription of the first audio data and training a model of an automatic speech recognition system using the first text string and the first audio data. The method may further include in response to completion of the training, deleting the first audio data and the first text string and after deleting the first audio data and the first text string, obtaining second audio data of a second communication session between a third and fourth device and during the second communication session obtaining a second text string that is a transcription of the second audio data and further training the model of the automatic speech recognition system using the second text string and the second audio data.

A method may include obtaining a text string that is a transcription of audio data and selecting a sequence of words from the text string as a first word sequence. The method may further include encrypting the first word sequence and comparing the encrypted first word sequence to multiple encrypted word sequences. Each of the multiple encrypted word sequences may be associated with a corresponding one of multiple counters. The method may also include in response to the encrypted first word sequence corresponding to one of the multiple encrypted word sequences based on the comparison, incrementing a counter of the multiple counters associated with the one of the multiple encrypted word sequences and adapting a language model of an automatic transcription system using the multiple encrypted word sequences and the multiple counters.

Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: performing natural language processing to process a text string of a user, wherein the text string specifies characteristics of a container image to be built; processing, with use of natural language processing, instances of text-based data that describe respective ones of a plurality of container images stored within a container image repository; selecting, in dependence on a result of the performing natural language processing, and the processing, a base image from the plurality of container images; and presenting prompting data to the user that prompts building of a new container image, wherein the prompting data references the base image.

Provided are a computer program product, system, and method for providing a repository of audio files having pronunciations for text strings to provide to a speech synthesizer. The repository has data structures for text strings in documents. A data structure for a text string indicates at least one attribute of a presentation of the text string in the document and at least one audio file providing at least one audio pronunciation of the text string. A search text string and a search attribute are received from the speech synthesizer. A determination is made of a data structure in the repository including a text string and an attribute matching the search text string and the search attribute, respectively. An audio file, indicated in the determined data structure, is returned to the speech synthesizer to output for the search text string in a document being processed by the speech synthesizer.

Systems, methods, and computer-readable storage media for identifying databases or resources which contain verifiable information used to create generative AI output. A system can receive a request for generation of a report, the request including a string of text. The system can then parse the string, resulting in a parsed request, and identify at least one verified data source for each piece of data within the parsed request. The system can then generate a query for each verified data source based on the parsed request, resulting in at least one query, and send those queries to the verified data sources. Those sources can respond with verified data, and the system can generate, using a generative Artificial Intelligence (AI) algorithm, the report using the verified data.

The invention relates to the field of data recognition, and particularly discloses an intelligent homework test paper reading and wrong question induction method and system and electronic equipment. The method comprises the following steps: acquiring image data of homework or test paper, preprocessing the acquired image data, detecting external polygon coordinates of characters in the preprocessed image through PANet, identifying a text line picture through CRNN, and outputting a text character string; for science categories, decomposing and scoring problem solving steps through a symbolic calculation engine, verifying a vector relationship through an image feature recognition engine, and for liberal arts categories, performing semantic scoring through an NLP deep analysis engine; and according to the reviewing result of each subject, respectively generating a wrong question set of the corresponding subject. According to the method, through task cutting, image processing and deep learning, on the premise that the marking precision is guaranteed, the full-process dependence of a large model is reduced, the system response speed is increased, and the hardware resource consumption is reduced.

In some aspects, the techniques described herein relate to a method including: receiving, at a query platform, a query text string; tokenizing, by the query platform, the query text string, wherein the tokenizing generates a query vector embedding; determining, by the query platform, a document vector embedding, wherein the document vector embedding is above a similarity threshold with respect to the query vector embedding; retrieving, by the query platform, textual document data related to the document vector embedding; sending, by the query platform, the query text string and the textual document data to a generative model engine; receiving, by the query platform and from the generative model engine, a natural language response to the query text string; and displaying, by the query platform, the natural language response via an interface.

Provided in the present application is a large-language-model-based long text generation method capable of realizing context compression. The method comprises: acquiring context text to be compressed and prompt text, and performing compression-based encoding processing, so as to obtain a corresponding compression vector and a prompt embedding vector; concatenating the compression vector and the prompt embedding vector, and performing autoregression-based decoding processing on a fused feature, which is obtained by means of concatenation, so as to obtain a plurality of corresponding token identifiers; and on the basis of a preset vocabulary, mapping the token identifiers into text character strings one by one, and combining the text character strings into compressed context text. By means of the present application, long context text processed by a large language model is compressed, thereby solving the technical problem in the prior art of a semantic model consuming a large amount of model calculation resources and data storage resources when processing long context text.

Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: examining foreground voice data of a multimedia stream that includes a video stream data and an audio stream; identifying in dependence on the examining an open time window that is absent of foreground voice data; processing, in dependence on the identifying, media stream data of multimedia stream; generating, in dependence on the processing, a text string for deployment in the open time window, wherein the text string describes content of the video stream; converting the text string into a synthesized voice segment; and adapting the audio stream data so that the synthesized voice segment is included in the audio stream and time bounded within the open time window.

The present disclosure relates to systems, non-transitory computer-readable media, and methods for generating stylized translated text using attention heads from a transformer neural network. In particular, in some embodiments, the disclosed systems obtain an input text string in a first language, the input text string comprising a style formatting element. Additionally, in some embodiments, the disclosed systems generate, using a transformer neural network to process the input text string, a translated text string in a second language different from the first language. Moreover, in some embodiments, the disclosed systems determine attention head values generated by the transformer neural network for words of the input text string as part of generating the translated text string in the second language. Furthermore, in some embodiments, the disclosed systems generate a translated style formatting element for the translated text string based on the attention head values for the words of the input text string.

A system for recognizing vertically oriented alphanumeric text in images, the system including a processor configured to receive one or more images comprising vertically oriented alphanumeric text and detect one or more regions-of-interest in each image via a trained text detector. The processor is configured to execute a cropping of the detected one or more regions-of-interest encompassing vertically oriented alphanumeric text from each image to obtain one or more text crop portions and rotate the one or more text crop portions to obtain one or more orthogonally rotated text crop portions. The processor is configured to execute a trained ensemble of two different text recognition models on each of the obtained one or more text crop portions and the one or more orthogonally rotated text crop portions and generate a set of candidate recognized text strings based on the executed trained ensemble and determine a final recognized text string.

The present disclosure relates to a text similarity measurement method and apparatus, device, storage medium, and program product. The method includes: obtaining a first text string and a second text string; constructing a joint probability distribution of the first text string and the second text string, and sampling the joint probability distribution to obtain a sampling string; calculating a distance from the first text string to the sampling string to obtain a first distance matrix, and calculating a distance from the second text string to the sampling string to obtain a second distance matrix; and determining a similarity between the first text string and the second text string based on the first distance matrix and the second distance matrix.

Methods and apparatus for supporting generation of a virtual 3D object Methods (100, 300) are disclosed for supporting generation of a virtual 3D object. The methods include, at a preprocessing node, obtaining heterogeneous input data for the virtual 3D object (110), aggregating the heterogeneous input data (120), and generating a contextual prompt from the aggregated heterogeneous data (130), the contextual prompt comprising a text string conveying semantic information with respect to the virtual 3D object, which semantic information is derived from the aggregated heterogeneous input data. On detection of a trigger event, the preprocessing node provides the contextual prompt to a processing node (140). The methods further include, at a processing node, inputting the contextual prompt to an ML model (220) operable to generate a prompt for a Text-to-XR function within the processing node, and contextual metadata for the virtual 3D object. The methods further include, at the processing node, using the generated prompt to generate a representation of the virtual 3D object (230), and providing the representation and the contextual metadata for the virtual 3D object to an XR client (240).

This invention provides a drawing processing method and system based on scene text recognition. It identifies all graphic and non-graphic elements within the global scope of a drawing. Based on the shape features of the graphic elements, it divides the drawing into several sets of non-graphic elements. Based on the connected components of all non-graphic elements within each set, it generates several text strings corresponding to those sets and performs fuzzy classification to identify the text results of the non-graphic element sets. Based on the spatial distribution information of the graphic and non-graphic elements on the drawing, it preprocesses the text results and overlays them onto the corresponding areas of the drawing, distinguishing between non-graphic and graphic elements. It also performs connectivity recognition on the non-graphic elements, initially labeling the text strings within them to obtain their corresponding text results. This achieves synchronous association recognition of graphics and text within the drawing, improving the efficiency and accuracy of drawing editing.

A computer-implemented method is presented for fingerprinting network services. The method includes: receiving a plurality of banners, where each banner contains text data describing a network service accessible at a port on a networked device in a computer network; generating a set of embeddings from the plurality of banners using a large language model, where each embedding in the set of embeddings represents one or more banners from the plurality of banners clustering the embeddings based on distance between the embeddings using a clustering method, thereby forming a group of clusters; and generating a fingerprint for a given cluster in the group of clusters, where the fingerprint is a text string may come from a given banner and identifies a network service or product accessible in the computer network.

An information handling system may include a memory and a processor communicatively coupled to the memory and configured to convert a string of text into a text-based scannable code comprising a string of characters including space characters, special characters, and linefeed characters, such that when the string of characters is displayed to a display device, the string of characters resembles an image-based scannable code.

A text segmentation and sensitive word detection method based on matrix multiplication, comprising the following steps: obtaining an original text string and a sensitive word library; constructing a deterministic finite state automaton tree diagram of sensitive words according to the sensitive word library; converting the original text string into a text character two-dimensional matrix and recording the length of the text character two-dimensional matrix; constructing a matching two-dimensional matrix according to a horizontal matching rule, a vertical matching rule, an oblique matching rule and an inverse oblique matching rule, the length of the matching two-dimensional matrix being the same as that of the text character two-dimensional matrix; performing dot multiplication processing on the text character two-dimensional matrix and the matching two-dimensional matrix to obtain a corresponding result matrix; generating a corresponding matching text string according to the result matrix, and matching with the deterministic finite state automaton tree diagram to determine whether there is a sensitive word. The application supports interval text character information detection, improves the sensitive word detection accuracy and detection efficiency.

The invention relates to the field of vehicles, and provides a vehicle-mounted man-machine interaction method, a vehicle-mounted man-machine interaction device and electronic equipment aiming at the problem of how to enable a vehicle-mounted terminal to accurately respond to the intention of a user and realize an application function expected by the user. The method comprises the steps that any one or a combination of more of vehicle data, driver data and environment and context data serves as a data source, a semantic text sequence is obtained according to the data source, and the semantic text sequence comprises one or more feature texts; the feature text is used for describing any one or a combination of more of a vehicle feature, a driver feature, an environment feature and a context feature, and the feature text is a text character string described by referring to a natural language; taking the semantic text sequence as the input of an intelligent model, and obtaining a prediction result output by the intelligent model; and calling the corresponding application service according to the prediction result. According to the method provided by the embodiment of the invention, the prediction accuracy of the intelligent model can be improved, and the user experience of man-machine interaction is improved.

A device may be configured to perform spatial extrapolation based on information included in a neural-network post-filter characteristics message. In one example, a neural-network post-filter characteristics message includes a syntax element specifying a text string prompt used for generating contents of a spatial extrapolation image area and a syntax element having a value specifying auxiliary input data includes character values derived from the text string prompt The device may be configured to determine a prompt string position and derive the text string prompt based on the prompt string position.

This invention discloses a method for handling rail transit anomalies and training a rail transit anomaly handling model. The method includes: acquiring a knowledge base of current train anomalies and train schedule adjustments; determining an anomaly text string based on the current train anomaly; performing vector transformation on each phrase in the anomaly text string to obtain at least one first phrase vector; performing vector transformation on each phrase of each entry in the train schedule adjustment knowledge base to obtain at least one second phrase vector; calculating the similarity between each first phrase vector and each second phrase vector; filtering entries in the train schedule adjustment knowledge base based on the similarity between each first phrase vector and each second phrase vector to obtain the optimal schedule adjustment entry; and processing the current train anomaly based on the optimal schedule adjustment entry. This invention can improve the efficiency of rail transit train operation schedule adjustments.

This application provides a form highlighting method, apparatus, computer device, and storage medium based on a large language model. The method involves inputting a screenshot of the form to be recognized into an image recognition model to obtain the image recognition result. Then, a unique identifier is added, a first mapping relationship is determined, and a text string and an identifier string are further constructed. The large language model is then instructed to generate the result information required for the highlighting function based on this information. Finally, based on the result information, the method can quickly locate and highlight the form in response to the user's highlighting request. This solution realizes a complete process from form image to structured information extraction and visualization, effectively relying on a large language model to improve the efficiency, accuracy, and convenience of form highlighting processing.

A method for generating spectroscopic data includes inputting, by a computing device, a text string comprising a structural representation of a material to an encoder of a natural language processing (NLP) model implemented with a deep neural network. The method includes generating, using the encoder of the NLP model, an encoded representation of the text string. The text string may include latent chemical bond information of the material. The method includes mapping, by the computing device, the encoded representation including the latent chemical bond information to a spectrum array, the spectrum array including predicted spectroscopic data of the material. The method also includes outputting, by the computing device, the spectrum array.

The arrangements described herein relate to a computing system configured to receive a file comprising text strings, determine a validity of information in the text strings using at least one ML model, in response to determining the validity of the information in the text strings, tokenize the file by generating a digital asset corresponding to the file, the token includes a pointer to the file, and record the digital asset on a distributed ledger database / blockchain.

A method for displaying a search result to a user in a client device is provided. The method includes receiving, in a server, a search query from a user, the search query including a string of text characters. The method also includes converting the search query into a sequence of numbers based on a semantic content of the string of text characters, the sequence of numbers defining a vicinity in a multidimensional space, identifying a text string associated with a point within the vicinity in the multidimensional space, ranking the text string according to a similarity value with the search query, and providing a link to a media file associated with the text string to the user as a search result. A system, a memory storing instructions which, when executed by a processor cause the system to perform the above method, and the processor are also provided.