Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

9891 results about "Speech model" patented technology

Serial Processing Models. Serial models of speech production present the process as a series of sequential stages or modules, with earlier stages comprising of the large units (i.e. sentences and phrases), and later stage comprising of their smaller unit constituents (i.e. distinct features like voicing, phonemes, morphemes, syllables).

Real-time contextually aware artificial intelligence (AI) assistant system and a method for providing a contextualized response to a user using ai

An artificial intelligence (AI) assistant system and a method for providing a contextualized response to a user using AI are disclosed. The system comprises an audio input device for receiving voice input, an audio output device for providing output, a processor, a wireless communication device, a contextual memory unit for storing conversational context data on a sliding window basis, and a non-volatile system memory unit. The processor executes instructions to receive voice input, determine user identification, update conversational context data with user identification and a tokenized representation of the voice input, process the voice input using a transformer-based language model to generate a response, update the conversational context data with a tokenized representation of the generated response, and output the response via the audio output device. The method comprises receiving voice input, determining user identification, updating conversational context data, processing voice input, and generating and outputting a conversational response.
Owner:VASYLYEV SERGIY

Automated content creation and content services for collaboration platforms

Embodiments described herein relate to systems and methods for automatically generating content, generating API requests and / or request bodies, structuring user-generated content, and / or generating structured content in collaboration platforms, such as documentation systems, issue tracking systems, project management platforms, and other platforms. The systems and methods described use a network architecture that includes a prompt generation service and a set of one or more purpose-configured large language model instances (LLMs) and / or other trained classifiers or natural language processors used to provide generative responses for content collaboration platforms.
Owner:ATLASSIAN PTY LTD +1

Method for constructing knowledge graph based on large language model and vector library

The invention discloses a method for constructing a knowledge graph based on a large language model and a vector library, and relates to the technical field of knowledge graphs, and the method comprises the following steps: carrying out word segmentation, part-of-speech tagging and semantic analysis processing on a collected natural language text by using a pre-trained RoBERTa model to obtain a semantic representation vector of the natural language text; through a Seq2Seq framework and an attention mechanism, obtaining association between the semantic representation vector and the corresponding named entity, and obtaining the named entity through the obtained association; constructing a knowledge graph; fusing the constructed knowledge graph with a relational database of a preset field to generate a fused knowledge graph; named entities and corresponding attributes are obtained through dependency syntax analysis, the entity relation of the named entities is obtained, and a structured triple fusing the named entities, the attributes and the entity relation of the knowledge graph is constructed. In order to solve the problem of high entity linking difficulty in the prior art, the knowledge graph is converted into low-dimensional dense vector representation, and an inverted index database and the like are constructed, so that the difficulty is reduced.
Owner:北京长河数智科技有限责任公司 +1

Method and System for Multi-Level Artificial Intelligence Supercomputer Design

A method for improving responses to large language model (LLM) prompts including receiving at an input broker a request input from a user including an LLM prompt, deriving a search query from the LLM prompt, searching a plurality of documents using the search query to identify a first subset of documents and a second subset of documents, generating a first answer using a first context-specific LLM, where the first context-specific LLM uses the first subset of documents as the context and the LLM prompt as the prompt, generating a second answer using a second context-specific LLM, where the second context-specific LLM uses the second subset of documents as the context and the LLM prompt as the prompt, providing each of the first answer and the second answer to an output broker, determining a primary result at the output broker, and transmitting the primary result to the user.
Owner:MADISETTI VIJAY DR

Dialogue system and a dialogue method

A dialogue system, comprising: an input configured to receive input data relating to speech or text provided by a user; an output configured to provide output data relating to speech or text to a user; and one or more processors, configured to: receive, by way of the input, input data relating to speech or text provided by a user; provide the input data to a subject safety module, the subject safety module configured to receive the input data and evaluate the input data before a system response is output, evaluating the input data comprising performing a first determination on the input data using a matching module and a second determination on the input data using a trained model, wherein the matching module performs the first determination to determine whether the input data matches one or more items from a pre-determined set of one or more items; generate a system response using a first process or using a second process, wherein the first process uses at least one trained language model to generate a dynamically determined system response and wherein the second process retrieves a pre-determined system response, wherein a selection between the first process and the second process is made based on the evaluation of the input data; output, by way of the output, the system response.
Owner:LIMBIC LTD

Large language model reasoning acceleration method and system based on dynamic video memory compression and memory isomerism

The invention discloses a big language model reasoning optimization method and system based on dynamic video memory compression and memory isomerism, and intelligent management of video memory resources is realized by integrating a dynamic compression strategy of KV Cache and a memory parallel architecture. The method comprises the following steps: 1) analyzing the spatial-temporal characteristics of the KV Cache in real time, adaptively selecting a quantization compression algorithm, a rarefaction algorithm or a low-rank decomposition algorithm, performing hierarchical storage based on attention head importance scores, keeping high precision of a core head, and implementing low-bit quantization on a secondary head; (2) the compressed inactive data are divided into a plurality of data blocks to be stored in a system memory, a parallel data channel group is established according to the number of physical channels, the compressed blocks are concurrently read through multiple channels during loading, and parallel decompression of a sparse matrix is accelerated through a GPU tensor core; and 3) constructing a KV Cache multiplexing mechanism and a parallel channel, and parallelizing a compression / decompression process and model calculation by adopting a hardware acceleration compression and asynchronous pipeline mechanism.
Owner:HANGZHOU AMTD YINGANG DIGITAL TECH CO LTD

Customer data processing and insight system based on large language model

The invention belongs to the technical field of artificial intelligence and big data, and discloses a customer data processing and insight system based on a big language model. The system is composed of a multi-source data access module, a data preprocessing and label fusion module, a large language model semantic understanding module, a knowledge enhancement and semantic linkage module, an insight generation and visualization module, an intelligent strategy output module and a feedback learning and self-optimization module. According to the method, multi-source heterogeneous data such as texts, voices and structured behaviors are integrated, and the deep semantic analysis capability of a large language model is combined, so that global modeling of customer behaviors and intentions is realized; a multi-modal synchronous acquisition and standardization mechanism eliminates data format barriers, and a dynamic label mechanism adapts to context changes, so that the system can capture deep semantic association in customer expression, and compared with a traditional keyword matching method, the semantic understanding accuracy is improved by more than 40%, and a more complete data base is provided for insight generation.
Owner:SICHUAN JUFUREN TECHNOLOGY CO LTD

Generating a unified metadata graph via a retrieval-augmented generation (RAG) framework systems and methods

Systems and methods for reducing data retrieval times when accessing siloed data across disparate locations by generating a unified metadata graph via a Retrieval-Augmented Generation (RAG) framework are disclosed. The system selects a first Large Language Model (LLM) prompt corresponding to a first metadata identifier of the set of metadata identifiers. The system then augments the first LLM prompt with the first metadata identifier to be provided to an LLM, where the LLM is configured to generate a first intermediate output. The system then augments the first LLM prompt with the second set of metadata identifiers corresponding to the first metadata identifier to be provided to the LLM, where the LLM generates a second intermediate output. The system then generates a domain-specific unified metadata graph. Responsive to determining that first performance metrics fail to satisfy a performance criteria, the system performs an update process on the domain-specific unified metadata graph.
Owner:CITIBANK N A

Systems and methods for graph-based dynamic information retrieval and synthesis

Disclosed herein are systems and methods for graph-based dynamic information retrieval and synthesis, particularly for enhancing the response generation of chatbot applications using large language models (LLMs). The methods involve receiving a user query, generating a query embedding, and accessing a knowledge graph comprising nodes representing documents with associated document embeddings. A set of documents is retrieved based on similarity scores, and a distance matrix is constructed to calculate shortest-path distances between nodes. Documents are clustered to identify a largest cluster from which content is retrieved to formulate a response. The disclosed embodiments improve upon retrieval augmented generation (RAG) by leveraging the hierarchical relationships of a knowledge graph and node interconnectedness to provide contextually coherent information and mitigate the inclusion of information from irrelevant or superficially related documents in LLM responses.
Owner:HARMAN CONNECTED SERVICES INC

System that conducts dialogs using artificial intelligence

An artificial intelligence system that conducts dialogs with one or more participants to achieve an educational outcome, such as teaching material with a Socratic question-and-answer method, reviewing material to reinforce ideas or to prepare a participant for future tests or events, or evaluating a participant's knowledge or communication skills. The system may use an artificial intelligence engine that can understand and generate natural language, such as a large-language model like ChatGPT. Text-to-speech and speech-to-text converters may be coupled to the AI engine to enable the system to communicate directly with participants using spoken language. The system may provide an interface for entry of contextual information for a dialog, such as background information on topics for discussion. It may provide an interface for moderation of a dialog, for example by a teacher, lawyer, thesis advisor, etc. System may evaluate the performance of the participants at the end of a dialog.
Owner:SOCRATIC METRIC LLC

Incorporating large language model prompts in graph query language

Systems and methods are directed to incorporating large language model (LLM) prompts into graph queries using graph query language and executing these graph queries. The system receives a graph query that includes a LLM prompt in the graph query language. The LLM prompt comprises an input context and a query to be answered. Based on the graph query, the system traverses a knowledge graph to obtain context data associated with the input context. The context data may be constrained by access control policies of an enterprise providing the knowledge graph. A context table may be generated using the context data based on the input context indicated in the LLM prompt. The system then invokes an LLM to determine a result by applying the query of the LLM prompt to the context data obtained from the traversal of the knowledge graph. The result is then presented on a user interface of a client device.
Owner:MICROSOFT TECHNOLOGY LICENSING LLC

Vision generation method and device based on semantic association modeling, equipment and medium

The invention relates to the technical field of voice semantics, can be applied to business scenes of financial science and technology, medical health, poster design and the like, and discloses a visual sense generation method and device based on semantic association modeling, equipment and a medium. Generating a demand text containing theme and style parameters; semantic features in the demand text are extracted, semantic association weights are constructed, and element layout coordinates are optimized in combination with spatial distribution constraints; and encoding the layout information into a control matrix, fusing the control matrix with the initial noise, adjusting a noise reduction process through an encoding and decoding network, and generating target visual content highly matched with the semantic meaning of the user instruction. According to the method, the layout optimization function is constructed, the diffusion model is guided to focus the semantic salient region in space, language model output and the visual generation process are closely combined, structured response and space mapping of user semantic requirements are achieved, and the expression consistency and personalized adaptation capacity of visual content generation are improved.
Owner:SHENZHEN PINGAN COMM TECH CO LTD

Large model illusion relieving method and device, equipment and storage medium

The invention discloses a large model illusion relieving method and device, equipment and a storage medium, and the method comprises the steps: generating an initial answer and a corresponding reasoning process according to initial input information, and enabling the reasoning process to be obtained through employing a thinking chain prompt word to extract a generation process of the initial answer; performing illusion judgment on an initial question and answer pair to obtain an illusion judgment result, the initial question and answer pair including the initial input information and the initial answer; and when the illusion judgment result is that illusion possibly exists, performing illusion correction on the initial answer through an evidence set in combination with the reasoning process, and outputting correction content. According to the method, the illusion is reduced by optimizing the RAG technology, so that the reliability and the accuracy of large language model generated content are improved.
Owner:SHENZHEN FARBEN INFORMATION TECH CO LTD

Bidding document generation method based on retrieval enhancement generation and large language model

The invention relates to the technical field of artificial intelligence, and discloses a bidding document generation method based on retrieval enhancement generation and a large language model, and the method comprises the steps: analyzing technical parameters, legal terms and score weights in a bidding demand document; retrieving matched historical bidding document fragments and technical specifications from the industry knowledge base, and generating a retrieval enhanced data set; fusing the data through a dynamic weight distribution algorithm and generating an initial bidding document draft; checking conflict terms based on the legal semantic knowledge graph and marking correction suggestions; and optimizing the bidding document structure and the key content according to the score weight. According to the method, an online learning mechanism driven by sentence vector matching, timeliness weight calculation and user feedback is adopted, so that the problems of inaccurate technical parameter extraction, low legal conflict detection efficiency and non-optimized scoring rules in the bidding document generation process are solved.
Owner:SHENZHEN HIGHLAND BARLEY INFORMATION TECH CO LTD

Archive knowledge base construction and retrieval method and system based on multi-modal data fusion

The invention discloses an archive knowledge base construction and retrieval method and system based on multi-modal data fusion. The method comprises the steps that heterogeneous archive data are cleaned, image features are extracted through CNN, text features are extracted through Transform, audio is converted into text and then subjected to similarity, a unified feature vector is generated, and metadata is constructed according to archive code association; creating a graph database instance, defining nodes and relationship types, importing entities and relationships, and storing feature vectors and metadata; the features are mapped to a high-dimensional shared semantic space, positive and negative sample pairs are constructed to update embedded layer parameters, self-attention is used in modalities, a shared attention mechanism is used between modalities, weights are adjusted according to archive features, and unified knowledge representation is generated; segmenting the steering quantity of the multi-modal data, storing the steering quantity into a database, and adopting hierarchical indexing and optimizing as required; related document fragments are retrieved through RAG technology vectors, answers are generated with the help of a large language model, and session feedback is provided. The file retrieval efficiency and accuracy are improved.
Owner:GUANGDONG POWER GRID CO LTD +2

Intelligent operation and maintenance method for power grid equipment based on large language model and knowledge graph

The invention discloses a power grid equipment intelligent operation and maintenance method based on a large language model and a knowledge graph, and relates to the technical field of intelligent power grid operation and maintenance, and the method comprises the steps: carrying out the operation and maintenance of power grid equipment through a power grid operation and maintenance knowledge graph constructed through a large language model and a knowledge federation technology, the power grid equipment operation and maintenance comprises one or more of health state evaluation, fault risk prediction and early warning, intelligent operation and maintenance strategy generation and health degree dialogue query of the power grid equipment. According to the invention, a power grid operation and maintenance mode can be effectively promoted to be transformed and upgraded from a traditional manual experience type and a passive maintenance type to a data-driven, intelligent and active preventive maintenance mode. Key intelligent operation and maintenance technical support is provided for building a novel electric power system with new energy as a main body, the novel electric power system is assisted to achieve the development goals of being safer, more efficient, cleaner and lower in carbon, and important industry strategic significance and social contribution are achieved.
Owner:GANSU ZHENGPENG ELECTRIC POWER TECHNOLOGY CO LTD

Large language model (LLM) for enterprise applications developed by codeless platform

The present invention provides a large language model-based system and method for data processing in application developed by codeless platform. The invention includes identification of intent of a user to process procurement, supply chain, application integration, application restructuring or development scenarios.
Owner:NB VENTURES INC DBA GEP

Electric power work order intelligent processing method with RPA fused with multi-mode large model

The invention relates to the technical field of intelligent operation and maintenance and artificial intelligence crossing of a power system, in particular to an intelligent power work order processing method of an RPA fused multi-modal large model, which analyzes multi-modal work order data such as texts, voices, images and the like through a domain adaptation large language model, and realizes fault key information extraction and conflict resolution in combination with a dynamic knowledge graph; performing work order priority scoring and resource allocation by using space-time constraint reinforcement learning; an analysis result is converted into an automatic execution script through an RPA engine, and a whole-process closed loop of order sending, processing and feedback is achieved; meanwhile, a feedback optimization and conflict resolution cooperation mechanism is constructed, and the knowledge graph and the model precision are continuously iterated. The method improves work order processing efficiency and analysis precision, enhances decision scientificity, and is suitable for an intelligent operation and maintenance scene of a power system.
Owner:FUJIAN ZEYUAN INFORMATION TECHNOLOGY CO LTD

Robot control method and device based on multi-modal data fusion

The invention discloses a robot control method and device based on multi-modal data fusion, and the method comprises the steps: responding to a control instruction inputted by a user, receiving and preprocessing sensing information fed back by a plurality of types of sensors preset on a robot, and obtaining multi-modal data; dynamically adjusting the weight of each modal data, and extracting the feature vector of each modal data; according to the dynamically adjusted weight of each modal data, fusing the extracted feature vectors of each modal data to generate a comprehensive feature vector for representing the comprehensive state of the current environment; according to the comprehensive feature vector, the control instruction is analyzed through a preset large language model, and a final decision result of the robot is obtained; and according to the final decision result, calling and executing an instruction set required to be executed by the robot so as to drive the robot to complete related actions. Therefore, by adopting the embodiment of the invention, the robot can effectively process and utilize information from different modes to the maximum extent, and better decision and performance of the robot are realized.
Owner:HANGZHOU FANJIA TECH CO LTD

Method to detect and fix hallucinations in generative large language models

A system and method to detect a generative model's output to see if it is hallucinating or not, and to check the facts listed in the model. Additionally, when a hallucination or incorrect fact is detected, the correct fact can be applied as context to ask the model to regenerate the output again, taking the fact into consideration while also optionally lowering the temperature or increasing the top-k values from which to choose. A method is provided comprising: obtaining output produced by the generative model based on input provided to the generative model; performing summarization and topic extraction on the output to obtain one or more topics; performing fact checking on each of the one or more topics to produce a consolidated ground truth context; and declaring that the generative model is hallucinating or not based on the consolidated ground truth context.
Owner:CISCO TECHNOLOGY INC

Systems and methods for multi-modal language models

Embodiments described herein provide a method of generating a multi-modal task output to a text instruction relating to inputs of multiple different modalities (e.g., text, audio, video, 3D). The method comprises receiving, via a data interface, a first input of a first modality, a second input of a second modality and the text instruction relating to the first and the second inputs; encoding, by a first multimodal encoder adapted for the first modality, the first input of the first modality into a first encoded representation conditioned on the text instruction; encoding, by a second multimodal encoder adapted for the second modality, the second input of the second modality into a second encoded representation conditioned on the text instruction; and generating, by a neural network based language model, the multi-modal task output based on an input combining the first encoded representation, the second encoded representation, and the text instruction.
Owner:SALESFORCE INC

Conference summary processing method and system using AI

The invention relates to the technical field of intelligent conference processing, and relates to a conference summary processing method and system using AI, and the method comprises the steps: carrying out the real-time noise suppression of a collected conference audio stream and associated text data through a noise suppression algorithm, and carrying out the cross-modal alignment of the denoised data through a cross-modal alignment algorithm; a domain-specific attention head is inserted into an attention layer of the pre-trained Transform model, a domain-enhanced speech recognition model is constructed, and audio is converted into a text sequence with a speaker tag; adopting a heterogeneous graph neural network to construct a structured topic evolution graph; key decision nodes in the structured topic evolution graph are extracted based on a reinforcement learning strategy, and a final conference summary document is generated. In the decoding stage, the fusion proportion of the acoustic model and the language model is dynamically adjusted based on the real-time acoustic confidence coefficient, the recognition rate of the vocabularies in the professional field is increased, and the problems of frequent term transcription errors and poor semantic coherence in the professional conference are effectively solved.
Owner:GUANGZHOU DAZZLE VIEW INTELLIGENT TECH CO LTD

Methods, systems, and computer readable media for network test configuration and execution using brokered communications with a large language model (LLM)

One example method for network test configuration and execution using brokered communications with a large language model (LLM) comprises receiving, at an LLM communication broker and from a user, input regarding a network test objective for a network test; generating, by the LLM communication broker, based on the input regarding the network test objective, and using an application programming interface (API) of the LLM, input for invoking the LLM to produce configuration instructions for the network test; providing, by the LLM communication broker and to the LLM, the input for invoking the LLM to produce the configuration instructions; receiving, by a network test system, the configuration instructions for the network test; using, by the network test system, the configuration instructions to configure a network test system to conduct the network test; and conducting, by the network test system, the network test.
Owner:KEYSIGHT TECHNOLOGIES INC

Large-Scale, Privacy Preserving Personalized Large Language Models (LLMs)

A method for providing personalized responses to textual prompts using a large scale, privacy preserving, large language model (LLM) includes receiving a textual prompt from a user specifying a task for an LLM to perform, and obtaining a set of user features associated with the user. The method also includes determining, using the set of user features associated with the user, a user prompt embedding for the user, and processing, using the LLM, the textual prompt conditioned on the user prompt embedding for the user to generate a personalized response to the textual prompt. The method further includes providing the personalized response to the textual prompt for output from a user device associated with the user.
Owner:GOOGLE LLC

Chatbot

A computer-implemented chatbot including a prompt generation module, a large language model (LLM) module, and an answer generation module. The prompt generation modules generates an initial prompt based on a prompt template combined with a received user query. The initial prompt includes information source data specifying sources of factual information, conversation history data, and failed response data. The initial prompt is input to the LLM module, which is configured to generate an output and communicate the output to the answer generation module. The answer generation modules determines if the output is a plan to answer a user query. If so, relevant data is retrieved from external database or more APIs specified in the information source data. A further prompt is generated for answering the user query and input to the LLM module. The answer generation module repeats its tasks until a suitable answer to the user query is output.
Owner:SAGE GLOBAL SERVICES LTD

Personalized retrieval-augmented generation system

The present disclosure relates to systems, non-transitory computer-readable media, and methods for generating personal responses through retrieval-augmented generation. In particular, the disclosed systems can generate a query embedding from a query generated by an entity and determine data context specific to the entity by comparing the query embedding with a plurality of vectorized segments of content items associated with the entity. The disclosed systems can provide the data context to a large language model and generate a personalized response informed by the data context. Subsequently, the disclosed systems can provide the personalized response for display on a client device associated with the entity.
Owner:DROPBOX INC

Knowledge management system and method for constructing large voice model

The embodiment of the invention provides a method for constructing a large language model knowledge management system. The method comprises a knowledge construction step, a knowledge retrieval step and an answer production step. According to the scheme, the artificial intelligence large language model technology can be utilized. External knowledge is obtained from multiple data sources, a search technology and a cue word function of a large language model are combined, RAG proposes a problem to the large model, knowledge obtained by searching the problem in the multiple data sources is used as a background context, and information of the problem and the background context is integrated into cue words of the large language model. And therefore, the large language model can make a final accurate answer.
Owner:BEIJING XJ ELECTRIC

Multi-modal interview automatic quality analysis and evaluation method and system based on large model

The invention discloses a multi-modal interview automatic quality analysis and evaluation method and system based on a large model, and the method comprises the steps: collecting and storing multi-modal data, such as texts, audios, videos and behavior interaction, and carrying out the preprocessing of the multi-modal data to form a standardized data set; utilizing a preset interview structure and a large model to dynamically guide the process, adjusting the topic rhythm according to real-time feedback, and recording stage conversion information to form logic trajectory data for process coherence management; automatically coding text data through a large language model, extracting features such as keywords and performing topic clustering, performing cross validation and semantic fusion in combination with data analysis results of each modal, and generating deep analysis results such as psychological states; and generating a comprehensive assessment report containing qualitative description, quantitative score and psychological abnormality or cognitive disorder risk prompts based on a deep analysis result, thereby providing a basis for psychological health assessment and cognitive competence evaluation. According to the method, automatic analysis of multi-modal data is realized, and evaluation scientificity and efficiency are improved.
Owner:BEIJING NORMAL UNIVERSITY +1

Enterprise-level classification of data-items in an enterprise repository and prevention of leakage of personally identifiable information (PII)

Device, system, and method for automatically detecting and classifying personally identifiable information (PII) in documents and files. A method includes performing a deterministic rule-based search, in a plurality of stored documents, for PII data-items. If the deterministic rule-based search indicates that a particular document is more likely than not to contain a PII data-items then the method includes: extracting a textual snippet from the particular document, wherein the textual snippets surrounds the PII data-item; adding the textual snippet and the particular document to one or more training datasets utilized for training a Large Language Model (LLM) configured to find PII data-items in documents for Named Entity Recognition (NER) in those documents.
Owner:VARONIS SYSTEMS INC