109 results about "Predictive text" patented technology

The invention discloses an emotion prediction and disease derivation method and system based on multi-modal fusion, and the system comprises a data collection and preprocessing module, an emotion fusion module, an abnormal condition detection and cloud uploading module, and a disease possibility derivation module. The data acquisition and preprocessing module comprises a video part, a text part and an audio part, and the video part comprises face emotion recognition and prediction and human motion recognition and prediction; the text part comprises text content emotion recognition and prediction; the audio part comprises voice-to-text and voice tone emotion recognition and prediction, the system comprehensively captures an emotion state by fusing multi-mode information such as video, text and voice, and the accuracy and prediction capability of emotion recognition are improved; and by predicting the future emotion trend, the abnormal condition is warned in advance, and the response timeliness is improved.

The invention belongs to the technical field of computer application, and discloses a video content understanding method and device based on structured grammar information, electronic equipment and a storage medium, and the method comprises the steps: inputting a training sample into a target model for content understanding processing, obtaining a prediction text, and constructing a syntactic tree corresponding to the prediction text; calculating a syntactic tree editing distance between the syntactic tree and the syntactic tree of the reference text; calculating language structure loss by using the syntactic tree editing distance, and updating model parameters of the target model by using the language structure loss; under the condition that the model parameters of the target model are trained, obtaining a target video; and inputting the target video into the target model for processing to obtain a content text of the target video. In the application, the target model is trained based on the language structure loss calculated based on the syntax tree, and the target video can be understood, so that the content text with accurate grammar and reasonable and natural sentence structure is obtained.

The invention discloses a data generation method based on a multi-modal large language model, and the method comprises the steps: obtaining inquiry text information used for representing a business task in a business application scene, and image information associated with the business task, the inquiry text information comprises at least one of task definition information used for representing a service task needing to be completed and task logic information used for representing service logic, reasoning the obtained inquiry text information and the image information based on the trained multi-mode large language model, and obtaining the inquiry text information and the image information. Generating prediction text information used for representing response information of the inquiry text information based on the image information, the prediction text information comprises at least one of entity information used for representing a target image in the image information and image position information where an entity is located, attribute information used for representing additional information of the entity, and rule information used for representing a rule disassembled based on task logic. According to the invention, the generated data has pyramid dimension information.

Method, software, and apparatus for improved Augmentative and Alternative Communication (AAC) solutions. In one aspect, a user interface is provided with a set of suggestions comprising text, phrases, etc., and navigation buttons that enable users to select words and phrases to add to be written and / or spoken in a manner that reduces the number of user inputs. The suggestions are displayed in alphabetical order in rows with navigation buttons adjacent to the rows, with activation of a navigation button resulting in generation of updated suggestions having alphabetical ranges that are bounded by suggestions in associated rows. This approach may be combined with predictive text means to enable users to easily formulate text and / or speech content.

Methods and systems for improving communication involving a personal with aphasia are disclosed. The methods and systems include: obtaining an audio input indicative of speech of the person with aphasia; providing the audio input to a personalized aphasia translation assistant, wherein the personalized aphasia translation assistant was trained to recognize and translate speech of the person with aphasia using a general dataset of aphasia-speech; determining a plurality of words from the audio input using an aphasia-specific recognition model; inputting the plurality of words into an aphasia generative model, the aphasia generative model comprising a natural language processing machine learning model trained using a general dataset of aphasia sentences and a corresponding dataset of translated sentences; generating one or more formulated and contextual sentences using the personalized aphasia generative assistant; and outputting the one or more formulated and contextual sentences. A further method provides the training of a personalized aphasia communication assistant, in particular the adaptation of a pre-trained speech recognition model and of a pre-trained generative speech model based on predicted text and a user confirmation feedback of the accuracy of the predicted text.

The present disclosure relates to systems, non-transitory computer-readable media, and methods for generating and implementing a vision-language model that identifies and understands text-rich content depicted in digital images. For example, the disclosed systems determine, from among a plurality of digital images with at least a threshold probability of depicting text-rich content, a subset of digital images corresponding to a set of text-rich image classifications. In some embodiments, the disclosed systems generate a ground truth text phrase utilizing an optical character recognition model to process a digital image from the subset of digital images. In certain embodiments, the disclosed systems also generate a predicted text phrase utilizing a vision-language model and compare the ground truth text phrase with the predicted text phrase. In some embodiments, the disclosed systems modify parameters of the vision-language model based on comparing the ground truth text phrase and the predicted text phrase.

Systems and methods generate clinical documentation using large language models and artificial intelligence (AI). A template management module is provided to create customizable templates. A processing unit can receive input data from various sources and use AI to generate transcripts, summarize sessions, and produce clinical documentation such as clinical notes. The processing unit may also generate Current Procedural Terminology (CPT) and diagnosis codes, generate after-visit summaries, and generate referral letters. The AI may be trained on past clinical notes and can adapt to the clinician's style over time, with a feedback loop for continuous improvement. Additional features include cohort-based training, real-time language translation, predictive text, and analytics for documentation trends. The system supports customization of note length, style, and keywords, as well as integration with external medical databases and patient portals.

Various embodiments of the present disclosure provide machine learning architectures and data processing techniques for improving computer-based text comprehension. The techniques include generating, using a trained classifier model, target classification probabilities for labelled text-based objects from a testing portion of a labelled training dataset and identifying predictive text-based objects from the labelled text-based objects based on the target classification probabilities. The techniques include applying a staged prompting mechanism with a generative extraction model to identify a target set of explanatory text segments from the predictive text-based objects that may be clustered into semantic segment clusters. The techniques include generating explanatory summary segments respectively corresponding to the semantic segment clusters and generating a target classification signature based on a plurality of terms from the one or more explanatory summary segments.

The invention provides a speech translation method and device, and the method comprises the steps: translating source language speech data based on a speech translation model, and obtaining a target language text; a training target of the speech translation model comprises minimizing a difference between a first target language prediction text generated based on source language sample speech data and a translation label corresponding to the source language sample speech data, and minimizing a difference between speech features of the source language sample speech data and text features of a first source language sample text. And minimizing the difference between a second target language prediction text generated based on the pseudo source language speech features and a translation label corresponding to the second source language sample text. According to the method and the device, the problem of scarcity of annotated voice data is solved by efficiently utilizing relatively rich text data in a low-resource language scene, so that the performance of a voice translation model is improved.

The present application is suitable for the technical field of natural language processing, and provides a text generation method and apparatus, an electronic device, and a storage medium. The method comprises: acquiring a predicted token and a first hidden representation corresponding to text to be predicted, wherein the predicted token includes a new token obtained by performing inference on the text to be predicted, and the first hidden representation is an embedding representation obtained by performing, by means of a transformer layer, feature extraction on the text to be predicted; using the predicted token and the first hidden representation as an input of a trained acceleration network to obtain a candidate token outputted by the acceleration network, wherein the acceleration network is used for inferring the new token on the basis of the first hidden representation and the predicted token to obtain the candidate token; and on the basis of the predicted token and the candidate token, determining first target text corresponding to the text to be predicted. The present application can improve the quality of generated text.

The application relates to the field of artificial intelligence, and discloses an intelligent conversion method based on videos and texts, which comprises the following steps: acquiring a training video and corresponding video text of the training video, and extracting training pictures in the training video; using an encoder in a pre-constructed text-video conversion model to perform feature vector coding, vector masking and vector splicing on the training pictures and the video text to obtain picture-text splicing vectors; using a semantic analysis network in the pre-constructed text-video conversion model to identify predicted pictures and predicted texts of the picture-text splicing vectors, and then decoding to obtain predicted videos and predicted video texts; calculating model loss of the pre-constructed text-video conversion model according to the predicted videos and the predicted video texts, and the training videos and the video texts, so as to generate a trained text-video conversion model, realize scene conversion on to-be-converted scene data, and obtain a scene conversion result. The application can improve the scene conversion efficiency between videos and texts.

This application provides a text-guided method and system for detecting face forgery. The method includes obtaining a face image to be detected (whether it is genuine or fake), inputting the face image into a face detection model to obtain the face authenticity detection result. The face detection model includes: constructing a text prompt lexicon covering multiple granularities and generating multi-dimensional text prototypes; extracting visual features, optimizing the feature distribution of visual features to obtain global visual features; performing feature separation and enhancement on the global visual features; mapping the global visual features after feature separation and enhancement to predicted text features; applying similarity constraints to obtain cross-modal prototype matching results; applying discriminative constraints on different-dimensional text prototypes to obtain feature measurement learning results; and outputting the detection result based on the cross-modal prototype matching results and feature measurement learning results. This method solves the problem of poor generalization in face forgery detection and improves detection accuracy and generalization ability.

A video time retrieval method based on bidirectional semantic enhancement comprises the following steps: firstly, respectively extracting video features and querying text features by using a pre-trained video encoder and a pre-trained text encoder; secondly, respectively inputting the extracted features into a feature alignment module to obtain aligned video features and text features; then, designing a TGVM module, dynamically enhancing video features according to global features of a text mode, and reducing irrelevant information in the video features; thirdly, designing a Feedback Decoder module and feeding back comparison loss, shortening the distance between decoded text features and original video features, and enhancing text feature representation through the original video features; then, using a cross attention mechanism to obtain joint features after interaction of the video and the text; finally, a corresponding time slice in the video is queried through a decoder prediction text, and a model is trained under the supervision of time retrieval joint loss; according to the method, the multi-modal reasoning capability of the multi-modal interaction part is enhanced, so that video and text information can be better aligned.

The application provides a machine customer service training system and method, a voice reply method and an electronic device. The machine customer service training system comprises a machine customer service model, a user model, a reward parameter configuration component and a termination component. The user model is used to generate a plurality of first predicted texts according to a first text output by the machine customer service model and a historical communication text of the first text. The machine customer service model is used to randomly determine a target predicted text from the plurality of first predicted texts, and generate a second predicted text according to the target predicted text and the historical communication text. The reward parameter configuration component is used to configure a first positive reward parameter for the machine customer service model when the current conversation between the user model and the machine customer service model ends successfully. The termination component is used to terminate the training of the machine customer service model when the number of training times of the machine customer service model is greater than a threshold value, so as to obtain a trained machine customer service model. The application can train a high-quality machine customer service model.

The invention discloses a voice representation model training method, device and equipment, a storage medium and a product, and the method comprises the steps: when a voice representation model is trained, carrying out the coding of obtained voice features, carrying out the token boundary prediction through employing the voice features, generating a token boundary, generating prediction text data according to the token boundary, carrying out the discretization of the prediction text data, and carrying out the training of a voice representation model. And discretized speech representation is obtained. Due to the fact that the token-level discretized speech features can be obtained, cross-modal expression with higher coupling degree of the speech mode and the text mode and closer information expression can be obtained, and semantic joint expression with more information is achieved, the coupling degree with a multi-modal language model is further improved, and the precision of a large model is prevented from being affected.

This application discloses a text input information processing method, apparatus, and storage medium. The method involves acquiring user text input information, including complete words already entered by the user and / or characters representing incomplete words currently being entered by the user; generating a first semantic feature based on the text input information to characterize the semantics of the text input information; wherein the first semantic feature further characterizes the positional information of complete words and / or input characters in the text input information, the positional information including word-level positional information corresponding to words and / or character-level positional information corresponding to characters; and generating output information corresponding to the text input information based on the first semantic feature, wherein the output information includes predicted words corresponding to the text input information. This method improves the accuracy of prediction results corresponding to the text input information and enhances compatibility with different types of prediction tasks, thereby improving applicability.

The invention relates to the technical field of speech recognition, and provides an end-to-end speech recognition method and system based on a keyword attention enhancement mechanism, and the method comprises the steps: extracting entity keywords through a keyword searcher; the voice cache manager receives continuous air traffic control audio clips, converts the continuous air traffic control audio clips into an audio Mel spectrogram and then converts the audio Mel spectrogram into an audio embedded sequence; the keyword encoder unit maps the lexical element embedding representation sequence into a keyword embedding sequence; the audio transliteration decoder unit performs keyword attention enhancement calculation and converts splicing vectors of the audio embedding sequence, the transliteration start mark and the lexical element embedding representation sequence into a prediction text sequence; and inputting the new lexical element embedded representation sequence into a text translator through autoregression until the audio transwriting decoder unit outputs a transwriting end mark, and outputting the predicted text sequence as a speech recognition text. According to the invention, real-time identification of the streaming input voice is realized, and the method has the advantages of high identification precision, low response delay, flexible deployment and the like.

A method for training an audio recognition model, an electronic device, and a computer-readable storage medium are provided The method includes: performing feature fusion on an audio feature and a related phrase feature, to obtain a first fused feature; performing phrase prediction based on the first fused feature and the audio feature, to obtain a first predicted phrase, and determining a first loss; performing text prediction based on the audio feature, the phrase feature and the first fused feature, to obtain a first predicted text, and determining a second loss; performing text prediction on an audio sample based on the audio feature and the first fused feature, to obtain a second predicted text, and determining a third loss based on the second predicted text and a text label; and training the audio recognition model based on the first loss, the second loss and the third loss.

A computer-implemented method comprises providing previous form data to a training and validation module, filtering the previous data to create a subset of training data meeting a validation threshold, for each of a plurality of web-based input forms training at least one artificial intelligence model relating to at least one field in at least one web-based input form using the subset of training data, deploying at least one model endpoint corresponding to a model based on the trained model and a template of the form, transmitting a web-based input form web page to a client device, wherein the web page is displayed on a user interface, wherein the web page includes an AI-based form field prediction text input portion used for the AI-based form field prediction and an interactive representation of the web-based input form separate and distinct from the free text input portion; receiving, from the client device, a first user input at the AI-based form field prediction text input portion of the web-based input form; requesting a plurality of suggested answers to the web-based input form based on the user input; selecting a respective model endpoint corresponding to the web-based input form; generating at least one predicted user input using the model endpoint and the first user input; modifying the interactive representation of the web-based input form to include the at least one predicted user input; transmitting the modified interactive representation of the web-based input form including the at least one predicted user input to the client device; receiving, from the client device, a second user input at certain fields of the modified interactive representation of the web-based input form relating to a user modification of the suggested answers; and storing the inputs at the fields of the completed web-based input form to the data platform.

Introduced here is a supervised spatial pooler for spatial pattern recognition using distance-based overlap measurement and distributed threshold-based winner selection. The supervised spatial pooler can incorporate batched learning and directed initialization. Moreover, a temporal memory system architecture is introduced, using synchronized spatial pooler components for spatial and temporal pattern learning. A language model extends the temporal memory system for predictive text generation, featuring an autoregressive encoder-decoder architecture with context-dependent token representations and end-of-sequence prediction. Additionally, an integrated pipeline architecture is described. The pipeline architecture uses the language model as its core learning algorithm and includes embedding and tokenization stages for neuromorphic computing applications.

The application discloses a semantic effect evaluation method and related device, the semantic effect evaluation method comprises: obtaining a to-be-evaluated dialogue; wherein the to-be-evaluated dialogue comprises predicted text related to a user intention; inputting the to-be-evaluated dialogue into a multi-turn dialogue test set, verifying the predicted text of different nodes in the to-be-evaluated dialogue by using the multi-turn dialogue test set; in response to at least one error node with an identification error existing in the to-be-evaluated dialogue, reconstructing content after the error node in the to-be-evaluated dialogue based on the multi-turn dialogue test set to obtain a first dialogue; and evaluating the first dialogue based on all nodes in the first dialogue. In this way, the semantic effect can be verified in an offline manner, and when a node with a semantic identification error is encountered in the test process, the remaining part after the error node is reconstructed in real time, so that the next sentence of dialogue can continue to flow into the next node for evaluation, and finally the evaluation of the semantic effect of all nodes is completely realized.

The invention provides a business expansion work order information extraction method based on deep learning. The method comprises the steps of obtaining a business expansion work order image; preprocessing the acquired images to obtain a unified image; inputting the unified image into a text recognition model, wherein the text recognition model is used for performing text recognition on the unified image to obtain a predicted text sequence; according to a predefined information extraction mode, an input text is generated based on the prediction text sequence, the input text is input into an information extraction model, and the information extraction model is used for carrying out information extraction on the input text to extract structured key information; correcting errors in the structured information based on business rules and dictionaries, and standardizing the structured key information into a uniform format to obtain an extraction result. By applying the method, the character-level sequence relationship can be optimized, the text recognition robustness in a complex scene is improved, the text recognition accuracy is enhanced, and character recognition in multiple arrangement modes is supported; automatic key information extraction can be realized, and the working efficiency is improved.

The invention belongs to the technical field of mining engineering, and particularly discloses a mine goaf instability risk grade classification prediction method and system and a medium. The prediction method comprises the following steps: acquiring goaf related data; performing preprocessing including text digitization on the data; carrying out hyper-parameter optimization on the CNN-LSTM by adopting an improved WOA; establishing a WOA '-CNN-LSTM model, substituting the preprocessed data into the model for training, and predicting risk level classification; the risk level classification prediction is evaluated by adopting accuracy, specificity, recall rate and F1-Score indexes; and carrying out textualization processing on the risk level classification prediction to obtain an intelligent classification result. The prediction system comprises a data acquisition module, a preprocessing module, a hyper-parameter optimization module, an intelligent classification module, a classification evaluation module and a result output module. The method has the characteristics of full-process automation, high reliability, and obvious improvement of accuracy and specificity.

The application provides a shape robustness text detection method based on improved CharNet and a system thereof. Step 1, inputting an image to be detected into a CharNet network; step 2, performing backbone network feature extraction on the image to be detected to form a feature map; step 3, inputting the feature map into two parallel branches of the CharNet network respectively, wherein the two parallel branches include a character branch for single character detection and recognition and a text instance detection branch for predicting a text instance bounding box; step 4, obtaining character recognition information based on the character branch for single character detection and recognition; step 5, obtaining character position information based on the text instance detection branch for predicting a text instance bounding box; and step 6, integrating the character recognition information and the character position information to obtain a text recognition result. The method is used to solve the problems of extremely long text and arbitrary shape text that are difficult to identify in text recognition.

Systems and methods for using metadata for improved transcription search are disclosed. In an example method, a computing system receives an audio stream from a client device. The method further involves predicting text for the audio stream using a speech-to-text model, including determining multiple segments, each segment including one or more terms and a confidence value for each term. The method further involves, for each segment, ranking the terms according to the confidence values. The method further involves generating a transcription including a highest-ranked prediction for each segment and metadata including the remaining lower-ranked predicted text for each segment. The method further involves providing a graphical user interface to the client device including the transcription and the metadata. The method further involves receiving, from the client device, revisions to the transcription and updating the transcription. The method further involves updating the speech-to-text model using the user revisions.

The invention provides a model training method and device, a task processing method and device, equipment, a medium and a product, and the model training method comprises the steps: processing training data according to a text generation model of a current training round to obtain a prediction text set and a sample label set corresponding to the training data, determining reward values corresponding to the prediction text set in a plurality of preset optimization dimensions; according to the reward value corresponding to each preset optimization dimension, determining an advantage value of each preset optimization dimension; according to the preset dimension weight of each preset optimization dimension and the advantage value of each preset optimization dimension, determining a dimension loss value of each preset optimization dimension; and training a text generation model according to the dimension loss values of the plurality of preset optimization dimensions to obtain a text generation model of the next training round. According to the embodiment of the invention, the training precision of the text generation model can be improved.

The application relates to the technical field of speech recognition, and provides an end-to-end speech recognition method and system based on a keyword-enhanced attention mechanism, which comprises the following steps: a keyword retriever extracts entity keywords; a speech buffer manager receives continuous air traffic control audio segments, converts the audio segments into audio mel spectrum graphs, and then converts the audio mel spectrum graphs into audio embedding sequences; a keyword encoder unit maps a word embedding representation sequence into a keyword embedding sequence; an audio transcription decoder unit performs keyword-enhanced attention calculation, converts an audio embedding sequence, a transcription start flag and a splicing vector of a word embedding representation sequence into a predicted text sequence; a new word embedding representation sequence is input into a text transcriptioner through self-recurrence until a transcription end flag is output by the audio transcription decoder unit; and the predicted text sequence is output as a speech recognition text. The application realizes real-time recognition of streaming input speech, and has the advantages of high recognition accuracy, low response delay and flexible deployment.

The application provides a retrieval enhancement generation method based on sequence generation, which comprises the following steps: retrieving a query text to obtain a candidate relevant text; generating a current predicted text block vector based on an initial context sequence and a sequence generation model, and matching a most relevant text sequence from the candidate relevant text based on the current predicted text block vector; updating the initial context sequence based on the current most relevant text sequence, generating a next predicted text block vector based on the updated initial context sequence and the sequence generation model, and obtaining a next round of most relevant text sequence until the obtained most relevant text sequence meets a preset condition; and determining a target prompt word based on the obtained most relevant text sequence. The method provided by the application optimizes the candidate relevant text obtained through retrieval by using a sequence generation model, realizes the generation of a coherent and high-quality prompt word, and further improves the performance of a downstream task of fine-tuning or prompt learning based on the prompt word.

The invention discloses a model optimization method for a large language model and related equipment. According to the embodiment of the invention, a plurality of training texts and a candidate information set obtained through retrieval can be obtained; constructing a training sample set based on the plurality of training texts and the candidate information set; disturbing the arrangement sequence of multiple pieces of candidate information in the candidate information set to obtain a disturbed candidate information set; calling the to-be-optimized large language model to perform text reply on the training text based on the perturbed candidate information set, and generating a predicted text reply result; performing result consistency evaluation on the predicted text reply result and the training text reply result, and constructing a preference data pair corresponding to the training text based on an evaluation result; and based on the preference data pair corresponding to the training text, performing model optimization on the to-be-optimized large language model to obtain an optimized large language model. According to the method, the retrieval enhancement generation capability of the large language model can be improved.