82 results about "Language module" patented technology

A method and apparatus are provided for updating the vocabulary of a speech translation system for translating a first language into a second language including written and spoken words. The method includes adding a new word in the first language to a first recognition lexicon of the first language and associating a description with the new word, wherein the description contains pronunciation and word class information. The new word and description are then updated in a first machine translation module associated with the first language. The first machine translation module contains a first tagging module, a first translation model and a first language module, and is configured to translate the new word to a corresponding translated word in the second language. Optionally, the invention may be used for bidirectional or multi-directional translation

The invention discloses a dialogue automatic reply system based on deep learning and reinforcement learning. The dialogue automatic reply system comprises a user interaction module which receives question information inputted by a user in a dialogue system interface; a session management module which records the active state of the user, wherein the active state includes historical dialogue information, user position transformation information and user emotion change information; a user analysis module which analyzes registration information and the active state of the user and portraits for the user so as to obtain user portrait information; a dialogue module which generates reply information through a language module according to the question information of the user with combination of the portrait of the user; and a model learning module which updates the language model through the reinforcement learning technology according to the reply information generated by the language model. According to the dialogue automatic reply system based on deep learning and reinforcement learning, the reply of the dialogue meeting the personality of the user can be given according to the dialogue text inputted by the user with combination of context information, the personality characteristics of the user and the intentions in the dialogue.

A method and apparatus are provided for updating the vocabulary of a speech translation system for translating a first language into a second language including written and spoken words. The method includes adding a new word in the first language to a first recognition lexicon of the first language and associating a description with the new word, wherein the description contains pronunciation and word class information. The new word and description are then updated in a first machine translation module associated with the first language. The first machine translation module contains a first tagging module, a first translation model and a first language module, and is configured to translate the new word to a corresponding translated word in the second language. Optionally, the invention may be used for bidirectional or multi-directional translation.

A system for use in developing a voice application, including a dialog element selector for defining execution paths of the application by selecting dialog elements and adding the dialog elements to a tree structure, each path through the tree structure representing one of the execution paths, a dialog element generator for generating the dialog elements on the basis of predetermined templates and properties of the dialog elements, the properties templates received from a user of the system, each of said dialog elements corresponding to at least one voice language template, and a code generator for generating at least one voice language module for the application on the basis of said at least one voice language template and said properties. The voice language templates include VoiceXML elements, and the dialog elements can be regenerated from the voice language module. The voice language module can be used to provide the voice application for an IVR.

A method and apparatus are provided for updating the vocabulary of a speech translation system for translating a first language into a second language including written and spoken words. The method includes adding a new word in the first language to a first recognition lexicon of the first language and associating a description with the new word, wherein the description contains pronunciation and word class information. The new word and description are then updated in a first machine translation module associated with the first language. The first machine translation module contains a first tagging module, a first translation model and a first language module, and is configured to translate the new word to a corresponding translated word in the second language. Optionally, the invention may be used for bidirectional or multi-directional translation.

Disclosed are a voice identification method and a voice identification system. The voice identification method comprises the steps of receiving voice input and extracting a voice frame characteristic; conducting voice decoding on input voice by utilizing decoding space to ensure a voice decoding result. The decoding space comprises multiple decoding paths constructed on the basis of syntax rules, the multiple decoding paths comprise three types of decoding paths, wherein one type of decoding path only comprises language type module nodes, another type of decoding path only comprises statistical language module nodes, the third type of decoding path comprises the language type module nodes and the statistical language module nodes, and a semantic parsing result is determined by recalling the nodes on the selected decoding paths. The voice decoding comprises the steps of enabling the input voice to traverse each decoding path in the decoding space, selecting a decoding path with the largest sum of a language layer score and an acoustic layer score, and determining the voice decoding result according to a triphone acoustic model of the nodes on the selected decoding path.

The invention relates to a parallelization security hole detecting method based on a function call graph, which is characterized in that a function set to be detected is determined by analyzing the function call relation in a C language module, instrumentation and assert analysis are performed for the function set by means of the preprocessing technology, and then security holes of a program are detected by the aid of model checking and the parallelization technology. The method mainly includes the steps: generating the function relation call graph, and determining C function information to be detected by analyzing the function relation call graph; extracting attribute information of a buffer area related to C program source codes by the aid of the constraint analysis technology, and inserting corresponding ASSERT statement information in a variable declaration, an assignment and a function call point by means of the attribute information of the buffer area; and performing accessibility judgment for the instrumented codes, and analyzing whether a dangerous point in the program includes an accessible path to discover the security holes in the source codes or not. The method detects the security holes such as buffer area overflow and the like by combining the constraint analysis static detecting technology, model checking and the parallelization technology, and the detecting precision of the method is higher than that of the general static detecting technology.

An integrated circuit modeling method 100 implementable on computer, which has an executable software model 145 having modules 140 of reusable functional cores 105 coded in a high level language and a virtual platform of the integrated circuit employable in an architecture exploration step 115. A modeling library of modules coded in high level languages and hardware level languages are provided and instantiated according to user input in a functional verification step 120 having a co-simulation environment, with interface code 170 between modules automatically generated by an interface generator 130 based on a two dimensional data array of hardware specification inputs 205, the interface code 170 further interfacing with wrappers engaged between high and hardware level language modules.

A system for use in developing a voice application, including a dialog element selector for defining execution paths of the application by selecting dialog elements and adding the dialog elements to a tree structure, each path through the tree structure representing one of the execution paths, a dialog element generator for generating the dialog elements on the basis of predetermined templates and properties of the dialog elements, the properties received from a user of the system, each of said dialog elements corresponding to at least one voice language template, and a code generator for generating at least one voice language module for the application on the basis of said at least one voice language template and said properties. The voice language templates include VoiceXML elements, and the dialog elements can be regenerated from the voice language module. The voice language module can be used to provide the voice application for an IVR.

A text inputting method includes: obtaining a user identifier, finding a user language module corresponding to the user identifier according to the user identifier; obtaining a user input, generating a candidate sentence list according to the user input; obtaining a universal language model, calculating an on-screen probability of each candidate sentence in the candidate sentence list according to the user language model and the universal language model; sorting candidate sentences in the candidate sentence list according to their on-screen probabilities; and outputting the sorted candidate sentence list. The method increases a text inputting accuracy ratio and an inputting speed. In addition, a text inputting system and an apparatus are also provided.

The invention discloses a comprehensive mathematics teaching and learning system. The system comprises a man-machine operation module, a classic mathematical model database, a simulation training model database, an explanation video database, a mathematical modeling interest advancing module, a simulation training module, a puzzling question collection module, a central processor, a teacher module, a measurement and control module, a data sharing unit and a language module. Data in the class can be acquired and used for after-class learning; exercise training is combined with games, the learning interest of students is improved, and the learning efficiency is improved; the students use different measurement and control test questions for training, individualized instruction is realized truly, and the pertinence is high; the test questions in simulation training are all random, the students cannot answer the questions by remembering answers mechanically, and accordingly, the students are driven to truly understand knowledge points related to the test questions.

A system for transforming domain specific unstructured data into structured data including an intake platform controlled by feed back from a control platform. The intake platform includes an intake acquisition module for acquiring data building baseline data related to a domain and problem of interest, an intake pre-processing module, an intake language module, an intake application descriptors module, and an intake adjudication module. The control platform includes a control data acquisition module, a control data consistency collator, a control auditor, a control event definition and policy repository, an error resolver, and an output that outputs results of the workflow into structured data enabled to be used in data analytics.

A method including: providing a mobile device (e.g. cellular phone) with a core engine for performing speech recognition; providing a plurality of sets of language-specific modules, each set of the plurality of sets for enabling the core engine to recognize a different language; selecting one set of language-specific modules among the plurality of sets of language-specific modules; and loading into memory within the mobile communication device the selected set of language-specific modules so as to enable the mobile communication device to recognize speech spoken in the language of the selected set.

The invention discloses a simulation verification system. The simulation verification system comprises a Verilog verification module, a System Verilog interface module, a System C module and a high-level language module. The Verilog verification module is used for instantiating RTL codes and obtaining first state information and further used for receiving time sequence control information returned by the System Verilog interface module to achieve simulation verification on a chip. The System Verilog interface module is used for processing the first state information to obtain second state information and further used for sending the time sequence control information returned by the System C module. The System C module is used for providing multiple language models and obtaining transaction-level data information and language call instructions according to the second state information and the corresponding language models and further used for generating and sending the time sequence control information according to an algorithm returned by the high-level language model. The high-level language model is used for obtaining and returning corresponding algorithms according to the transaction-level data information and the language call instructions. The application range is wide, and the testing efficiency and the precision of the testing result are improved.

A method and system are provided for updating an utilizing the contents of a non-essential region in a memory device in a computer system. A computer program is utilized to build an image file. The image file may include program code and data for operating a computer system. The image file includes an essential region for storing program code required to boot the computer system and a non-essential region for storing program code and data which is not required to boot the computer system. The non-essential program code may include graphics data, a language module, or diagnostic tools for the computer system. After building the image file, the image file is then copied to a memory device in the computer system. Finally, only the non-essential region in the memory device is updated. A method may also be utilized to execute program code contained in the non-essential region prior to executing program code contained in the essential region.

An embodiment of the invention provides a text correction method and user equipment and relates to the language processing field. The text correction method and the user equipment can reduce correction mistakes and improve correction flexibility and correction accuracy. The text correction method comprises the steps of obtaining more than two text types of text to be corrected in the preset text classification standard, obtaining a sub-language-module to be combined, corresponding to each text type, of the text to be corrected in a correction knowledge base, mixing more than two obtained sub-language-modules to be combined into a hybrid language model, and correcting the text to be corrected according to the hybrid language model so as to obtain correction suggested text. The text correction method and the user equipment are used for correcting wrong text.

The invention discloses a student learning integrated management system. The system comprises a plurality of teacher terminals and a plurality of student terminals which are connected with a management terminal through a communication module. Each student terminal comprises an exercise database, a review data storage module, a difficult question gathering module, a daily task module, a reciting training module, a simulation training module, a man-machine operation module, a display screen, a central processor, a voice module, a language module, a data sharing module and a database. According to the invention, all subject learning, reviewing and training are integrated in one system, management of learning conditions of students is greatly facilitated, each student uses customized reviewing data and tasks, the relevance is high, through design of the difficult question gathering module, on one hand, learning of the students is facilitated, on the other hand, teachers can know better about the conditions of the students, individualized teaching can be conducted, and the teaching efficiency is greatly improved.

An integrated circuit modeling method 100 implementable on computer, which has an executable software model 145 having modules 140 of reusable functional cores 105 coded in a high level language and a virtual platform of the integrated circuit employable in an architecture exploration step 115. A modeling library of modules coded in high level languages and hardware level languages are provided and instantiated according to user input in a functional verification step 120 having a co-simulation environment, with interface code 170 between modules automatically generated by an interface generator 130 based on a two dimensional data array of hardware specification inputs 205, the interface code 170 further interfacing with wrappers engaged between high and hardware level language modules.

A system is provided for generating an optimized speech model by training a knowledge module on a knowledge graph. A language module is trained on unlabeled text data and a speech module is trained on unlabeled acoustic data. The knowledge module is integrated with the language module to perform semantic analysis using knowledge-graph based information. The speech module is then aligned to the language module of the integrated knowledge-language module. The speech module is then configured as an optimized speech model configured to leverage acoustic and language information in natural language processing tasks.

The invention provides a query expansion method based on semi-supervised clustering. The query expansion method includes the steps: (1) initially retrieving user queries by a query likelihood estimation language module and returning n front documents of retrieved results; (2) manually annotating k front documents in the initial retrieved results and dividing the k front documents into a relevant document set and an irrelevant document set; (3) analyzing the n front documents by a semi-supervised clustering algorithm for constraint and distance integration and extracting the documents related to the queries as feedback documents; (4) selecting expansion words by an expansion word selection module according to the feedback documents and forming new queries by the aid of the expansion words and original queries. By learning relevancy of a small number of annotated documents and query, relevancy of a large number of unknown documents and query can be accurately estimated, the quality of the feedback documents is improved, and accordingly, the recall ratio and precision of retrieval are effectively improved.

The invention discloses a string segmentation method and device. According to the technical scheme, a character string segmentation language module is established according to the word frequency of word segments in each numeric and English character string in a corpus of linguistic data and English character strings; for any linguistic data and English character string to be segmented, the segmentation result of the numeric and English character string to be segmented is obtained by being based on the character string segmentation language model corresponding to a category of the numeric and English character string to be segmented, using a dynamic programming algorithm, and determining an optimal segmentation path of the numeric and English character string to be segmented. Therefore, the method and device solve the problems that when a dictionary and matching mode is used to conduct numeric and English character string segmentation, it is impossible to segment unlisted character strings, and when post-processing rules are used to conduct segmentation correction, coverage is limited. The accuracy of numeric and English character string segmentation is improved.

The invention discloses a culture dissemination and promotion system, which comprises a database, an infrared thermal radiation detecting device, a human-machine operating module, a central processing unit, a 3D holographic projector, a voice module, a language module and a data sharing module. The culture dissemination and promotion system disclosed by the invention, in the mode of an advertising board, can conduct sensing-type playing, so as to avoid the waste of electric quantity; meanwhile, since 3D holographic projection technology is adopted for playing, a sense of vivid reality is offered to viewers, so that the interesting of the viewers is greatly improved; a data sharing request is automatically transmitted after viewing, so as to achieve data sharing under the circumstance that the viewers agree; therefore, the effect of culture promotion is greatly improved; and with the design of the dialect module, the entire system is more user-friendly and is broader in application scope.

The present invention discloses a system for detecting and eliminating computer virus and its method. It includes: scenario language module for writing said scenario; scenario compiler for compiling the scenario language in said scenario into instruction code; and scanario executing machine for executing said instruction code. The scenario language module is formed from self-defined basic instruction set, and the scenario executing machine includes virtual CPU for executing said byte instruction code and self-defined API function. Said invention can call the self-defined API function and basic instruction and use them to write scenario, and compile it into byte instruction, and uses the engine containing scenario executing machine to execute said byte instruction to implement virus-detecting and virus-killing work.

The invention discloses a mathematical modeling teaching learning system which comprises a student learning unit, a teacher evaluation unit, a display, a teaching management unit, a modeling method summarizing unit, a language module and a wireless communication module, wherein the student learning unit is used for counting scores of daily training questions, final exam results and puzzling questions encountering during learning; the teaching management unit is used for recording lesson preparation information of teachers, information of courses previewed by students and classroom discipline information in real time. According to the system, data in the process of taking classes by the students and teachers can be collected to be used for after-school learning; due to the student learningunit and the teacher evaluation unit, student practice and teacher evaluation are well combined together, and the targeting property is high. Moreover, due to a data sharing module, test questions areprinted to be practiced again; the language module and a video explanation module are combined together, so that the learning quality of the students is effectively improved.

The invention provides customer premises equipment. The customer premises equipment is accessed by an electronic device; a browser is installed in the electronic device; multiple national language environments are pre-stored in the customer premises equipment; the customer premises equipment comprises a language module, a matching module and a setting module, wherein the language module is used for judging the language version of the browser of the electronic device; the matching module is used for judging whether the customer premises equipment has the same language environment as the language version of the browser and outputting the judgment result; and the setting module is used for setting the language environment of the customer premises equipment as the same national language environment as the language version of the browser when the matching module judges that the same national language environment exits. The customer premises equipment can set the language environment as the same national language environment as the language environment of the browser of the accessed electronic device, so that a customer conveniently accesses the customer premises equipment. In addition, the invention also provides a configuration method of the customer premises equipment.

The invention provides a language model training method and system, a mobile terminal and a storage medium, and the method comprises the steps: obtaining a training text and a training vocabulary, carrying out the classification of the training text so as to obtain a plurality of language modules, and constructing a language dictionary corresponding to the language modules according to the training vocabulary; performing model training on a module language model in the language module according to the language dictionary, and training the training text to obtain a text language model; obtaining to-be-recognized voice to perform phoneme recognition to obtain a phoneme string, and matching the phoneme string with the module language model to obtain a phoneme matching result; and performing probability calculation on the phoneme matching result through a text language model, and outputting the sentence corresponding to the maximum probability value. According to the method, the training efficiency and accuracy of the language model are improved by classifying the training texts and constructing and designing the language dictionary, and the language model can be effectively expanded on the basis of the training design of the module language model and the training texts.

The invention provides an important-field intelligent robot character broadcast and reading check and proofreading method and system. The system comprises three database modules pre-established in a system load module: a broadcast language module and a voice reading identification module, which are formed by collecting various knowledge base and specific symbol broadcast corpuses and reading corpuses; and a word list scientific and practical module formed by a system word bank database, an error type database and a scientific and practical word list. During broadcast and reading, an original manuscript is imported, the broadcast language module carries out broadcast on the original manuscript through a player, and a proofreader carries out proofreading on a proofreading manuscript according to broadcast; during voice reading proofreading, the proofreading manuscript is imported, the proofreader reads the original manuscript through a microphone to the system, and the voice reading identification module carries out proofreading on the proofreading manuscript; and during the proofreading above, the word list scientific and practical module is utilized to carry out reminding on the suspicious places simultaneously, and stores the modified information to the scientific and practical word list. A speech database comprises symbol corpus, and accurate proofreading of a text can be realized.

The invention discloses Fortran simulation model system based on high level architecture (HLA), and aims to solve the problem that the conventional Fortran simulation model cannot be added into an HLA distributed system. The Fortran simulation model system based on the HLA comprises a response time index (RTI) system unit, an adapter unit and a Fortran simulation model unit, wherein the adapter unit comprises a file encapsulation module, a function recalling module and an RTI service module. According to the Fortran simulation model system based on the HLA, a Fortran language module can be communicated with a C language module through a C language program on an object file layer; and the integration demand of the Fortran model during cooperative simulation can be met, so that the generality of cooperative simulation and the reusability of the existing Fortran model are improved.

A method for and a pre-compiling device capable of easily changing a software structure in compliance with a hardware characteristic. The pre-compiling device includes a source code scanner for scanning for a source code file, retrieving a configuration directive, analyzing a syntax of the directive, and outputting a tree of a lexical unit corresponding to the retrieved configuration directive; a core for outputting the configuration directive in a configuration state; code generation modules for outputting a configuration state output from the core as a programming language that has no correlation with a flow of the lexical unit; and programming language modules for transferring information on a comment syntax of the corresponding language to the source code scanner and the code generation modules, and receiving the programming language that has no correlation with the flow of the lexical unit outputted from the code generation modules and outputting source codes.