84 results about "Machine translation" patented technology

The invention discloses a cross-border logistics document multi-language machine translation method based on natural language processing, and relates to the technical field of machine translation, and the method comprises the steps: collecting original data of a cross-border logistics document, and carrying out text extraction and preprocessing through an OCR technology; performing language recognition and format analysis on the preprocessed document text to determine a source language and a target language; using a pre-trained cross-language alignment translation model to translate the document text into a target language text; the translation result is input into a compliance auditing module, and automatic compliance auditing is conducted through a knowledge graph and a double-check algorithm; and generating an output containing a translation result and a compliance audit report, and supporting manual review and revision. According to the method, the translation accuracy of the documents can be effectively improved, the manual auditing burden is reduced, and the globalization requirements of cross-border e-commerce and logistics enterprises are met.

The invention provides an intelligent glasses independent communication method and system based on RedCap and eSIM and a medium, and the method comprises the steps: after intelligent glasses are started, initiating a registration request to a 5G core network based on an embedded eSIM chip, and after the core network completes identity authentication and authentication, distributing an independent IP address for the intelligent glasses; collecting interaction data, and encrypting the interaction data; performing data processing on the interaction data based on a cloud AI platform to generate result data; analyzing the result data based on voice recognition and a machine translation model to obtain AR content or an output text or a voice result; the signal intensity of the 5G core network is monitored in real time, the signal intensity is compared with a set intensity threshold value, and a communication mode is dynamically switched based on a comparison result; the intelligent glasses are completely independently networked, free movement and use of the intelligent glasses are achieved, real-time translation, AR navigation, remote cooperation and cloud cooperation application are supported, and user experience and scene adaptability are improved.

This invention discloses a low-power edge-side simultaneous interpretation system and method based on audio-text synchronization and visual feature fusion, relating to the fields of multimodal human-computer interaction and machine translation technology. The system achieves high-precision audio-text timing alignment through an audio-text synchronization matching module, generates a lightweight visual feature stream using a visual feature processing module, and performs spatiotemporal fusion by a multimodal fusion inference module. Combined with edge-side heterogeneous computing power scheduling and dynamic power consumption control, it significantly reduces the power consumption of edge devices while ensuring low-latency translation of ≤20ms per frame. This invention fills the technological gap in edge-side low-power multimodal simultaneous interpretation and can be widely applied in edge scenarios such as mobile office, international communication, and smart wearables.

The invention discloses a translation hardware acceleration platform and method based on an SGRU, and relates to the technical field of machine translation, and the platform comprises an accelerator which comprises a control module, an input and output module, an Embedding layer, a distributed pulse cache module, a model parameter cache module, a calculation module and a gating calculation core; the input and output module is respectively connected with the control module, the Embedding layer, the model parameter cache module and the PC terminal; the calculation module is respectively connected with the control module, the Embedding layer, the distributed pulse cache module and the model parameter cache module; the gating calculation core is connected with the control module and the distributed pulse cache module. A calculation heterogeneous design of a calculation module and a gating calculation core is adopted, data conflicts are avoided through distributed caching, dynamic loading and updating are carried out based on a model parameter caching module, overall control is carried out according to an SGRU-based Seq2Seq model, and high-performance and low-power-consumption translation can be realized.

The present disclosure provides a method for knowledge distillation, an apparatus, an electronic device, a medium and a computer program product. The method includes: acquiring a training source text and a standard translation text corresponding to the training source text; inputting the training source text into a teacher translation model and a student translation model separately, to obtain a teacher distribution output by the teacher translation model and a student distribution output by the student translation model; obtaining a standard translation distribution according to the standard translation text and the training source text; and performing iterative training on the student translation model according to the teacher distribution, the student distribution, and the standard translation distribution, to obtain a target machine translation model.

The application discloses a machine translation software defect detection method based on combined semantics, comprising the following steps: S1, obtaining different Chinese translation sentences of the same English source sentence from different translation software; S2, using a word alignment model to correspond English words in the source sentence with segmented Chinese words in the Chinese translation sentences; S3, using a sentence compression model and a syntactic structure analysis method to obtain a main part and each additional part of the English source sentence respectively and form a sub-sentence set; S4, aligning each part of the English source sentence obtained in the step S3 with the corresponding translation part to obtain aligned translation of the main part and the additional part of the source sentence; and S5, detecting errors, including semantic similarity calculation and synonym checking.

Described is a system and method for training a multilingual semantic parser. A method includes receiving, by a multilingual semantic parser, a multilingual training dataset, wherein the multilingual training dataset includes pairs of utterances and meaning representations from at least one high-resource language and at least one low-resource language and wherein the multilingual training dataset is initially a machine-translated dataset, training, the multilingual semantic parser, by translating the utterances in the multilingual training dataset to a target language; and iteratively performing selecting, by an acquisition functions estimator, a subset of the multilingual training dataset for human translation, updating the multilingual training dataset with the human-translated subset of the multilingual training dataset with, and retraining, the multilingual semantic parser, with the updated multilingual training dataset.

The invention discloses an ancient modern text machine translation method based on deep learning, and belongs to the technical field of natural language processing. The invention provides a multi-task collaborative optimization framework aiming at the problems of domain knowledge deficiency, frequent word activation, corpus scarcity and the like of an existing neural network model in ancient language translation. Precise domain knowledge injection is realized by constructing a closed ancient language parallel corpus retrieval library and combining historical term alignment and logic chain reasoning; a probability weighted part-of-speech embedding mechanism is adopted, and grammar feature distribution of virtual word ambiguity and word type activity is dynamically fused, so that the grammar analysis robustness is improved; and semantic generation, retrieval alignment and part-of-speech constraint tasks are jointly optimized through a fixed weight strategy. According to the method, the semantic accuracy and the logic continuity of ancient language translation are improved, and extended application of complex ancient language styles is supported. Experimental results verify that the performance of the method is improved in terms of BLEU and ROUGE-L indexes, and efficient technical support is provided for ancient book digitization and cross-time explanation.

The application provides a kind of two-stage visual fusion multimodal translation method and system, to solve the problem of insufficient utilization of visual information in existing multimodal translation, single fusion mechanism, model prone to ignore visual information.The method comprises: obtaining source language text and corresponding image, extracting fine-grained regional visual features and coarse-grained global visual features;In the first stage, the fine-grained regional features are spliced with the text representation and encoded through guided self-attention to realize cross-modal alignment;In the second stage, the coarse-grained global features are projected and broadcasted through gated residual modulation to inject scene-level prior information;The multimodal machine translation target and the visual conditional mask language modeling target are jointly optimized for training.The application effectively improves the translation quality through two-stage multi-granularity fusion, and the BLEU and METEOR indicators on Multi30K and other data sets are significantly better than existing methods, which can be applied to multimedia translation, cross-language retrieval and other fields.

The invention, which relates to the technical field of English parsing, discloses an intelligent parsing system for a long and difficult English sentence structure in combination with context features, comprising a context feature hierarchical extraction module, a dynamic coupling association module, an adaptive weight adjustment module and a syntactic structure parsing module. According to the method, a three-layer context feature layered extraction model of a syntactic structure layer, a chapter semantic layer and a sentence pattern function layer is constructed, the multi-dimensional context features of English long and difficult sentences are dynamically coupled and associated, and the feature analysis weight of each layer is optimized in real time according to the context feature distribution through an adaptive weight adjustment module; synchronous linkage of syntactic structure analysis and context semantics is achieved, the problems that in the prior art, due to the fact that multi-dimensional context association is ignored, nested subordinate sentence splitting errors, logic confusion of master and slave sentences, core component positioning deviation and the like are caused are solved, and the accuracy of English long-difficult sentence structure analysis is improved; and a technical support is provided for application scenes depending on long and difficult sentence analysis, such as machine translation, academic literature research and reading, language teaching and the like.

The invention relates to the technical field of natural language processing and artificial intelligence application, and discloses a multilingual spelling error correction method based on big language modeling, which comprises the following steps: generating multilingual training data through a disturbance strategy, and supervising and finely tuning a big language model; zero sample error detection is realized based on lexical-level probability deviation; and a detection result and a generative error correction result are integrated, and only consistent errors are replaced. According to the method, the excessive error correction tendency of the generative model is effectively inhibited, the error correction performance of low-resource languages such as Vietnamese and Indonesia is improved, and meanwhile, the principle of'minimum change 'is kept. The method does not need manual data annotation, has good mobility and cross-language adaptability, and can be widely applied to scenes such as text cleaning, language learning and machine translation.

The invention discloses a multilingual speech semantic intelligent translation method and system applied to an exhibition scene, and belongs to the technical field of machine translation, and the method comprises the following steps: S1, obtaining a multi-person question judgment result; s2, if the multi-person questioning judgment result is multi-person questioning, audio identification information is obtained through analysis, and otherwise, the audio identification information is directly obtained through analysis; s3, obtaining each storage question keyword, each contrast question keyword and a key matching weighting factor corresponding to each question keyword; s4, obtaining a same-group evaluation result, if the same-group evaluation result is the same group, analyzing to obtain the comprehensive matching similarity of the storage groups, otherwise, analyzing the comprehensive matching similarity of each parallel storage group; s5, obtaining a comprehensive matching judgment result, if the comprehensive matching judgment result is unqualified, performing secondary refining processing to obtain a question and answer, and otherwise, directly obtaining the question and answer; and S6, voice broadcasting is carried out, and accurate separation and language recognition of voice sources of different questioning users are achieved.

This invention relates to a prompt-based machine translation method, belonging to the field of natural language processing technology. It solves the problems of inaccurate, omitted, and mistranslated nouns and proper nouns in existing machine translation models. By constructing a set of nouns and their translations from the text to be translated, the input text and adjustment matrix of the translation model are obtained. The translation model is then used to translate the input text, and the attention calculation of the model is adjusted using the adjustment matrix M, ultimately outputting the translation. Based on the input data containing noun translation prompts and the adjustment matrix, the accuracy of the noun translation by the translation model is guaranteed to a certain extent, solving the problems of omitted and mistranslated nouns, and improving the accuracy of noun translation in machine translation models.

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

The application discloses a neural machine translation Internet of Things remote proof method for data flow attacks, which comprises an offline stage and a runtime verification stage. In the offline stage, a verifier and a prover first complete the negotiation of a symmetric key for subsequent remote proof, and at the same time, complete static plugging for a target program. A program control flow dataset is pre-constructed through fuzzy testing, a neural machine translation model is trained to establish the mapping of program input to an execution path, and a control flow graph is embedded to provide a structured prior for decoder attention. In the runtime verification stage, the verifier initiates a proof challenge to the prover, the prover provides the program input and the control flow path of the last execution, the verifier predicts a benign path from the program input by the neural machine translation model, and the difference between the benign predicted path and the actual path is used to judge the legitimacy of the prover. The application realizes accurate modeling of the program execution path, and shows effective detection capability for the abnormal path triggered by malicious input containing real vulnerabilities, and a good balance is achieved between detection coverage and running overhead.

The invention relates to the technical field of online conferences, and discloses a collaboration system and method for an online conference, and the method comprises the following steps: a user logs in the system through authentication, the authentication comprises AzureActiveDirect-based login and secondary verification, and the secondary verification is realized through a mobile phone short message code or a security verification code; the user creates a new conference and submits conference information, wherein the conference information comprises a conference theme, a conference date, a conference owner and a remark; the user uploads a video file to the new conference, wherein the video file supports multiple formats; and performing translation analysis on the uploaded video file. The efficient multi-language speech recognition and translation system is constructed by integrating the acoustic model, the language model and the neural machine translation model with the attention mechanism, accurate speech recognition and real-time translation can be automatically carried out on conference videos including Cantonese, Mandarin and English, synchronous subtitles are generated, and the speech recognition and translation efficiency is improved. And the language barrier in the multi-language conference scene is effectively eliminated.

The application discloses a kind of electric power communication service data translation method, system, device and storage medium, method includes steps: obtaining electric power communication service data, obtaining service routing data from the electric power communication service data;The service routing data is converted into structured linked list data;The structured linked list data is input to pre-trained bidirectional LSTM network and carries out machine translation, and the bidirectional LSTM network output restores link sequence.The application does not need to rely on character level matching large-scale data set, and fully utilizes data context semantic information, solves the problem that site meaning is not clear due to data multi-sourcing, improves the accuracy and reliability of data matching, reduces the labor cost of carrying out data cleaning work.

The application discloses a drawing format conversion and machine translation method and system based on data recognition, relates to the technical field of computer-aided design file processing, and comprises the following steps: S1, a unified coordinate baseline is established, the absolute position, rotation angle and size ratio of each text object are recorded, and an adjacent signature sequence is generated according to the recording result; S2, the adjacent signature sequence is used to construct an adjacent chain with consistent direction for each text object, a unique chain sequence number is allocated according to the space features of the previous nodes, and a loop-free candidate set is generated. The application generates an adjacent signature sequence by establishing a unified coordinate baseline, constructs an adjacent chain with consistent direction and forms a loop-free candidate set; overlapping fuses are set to eliminate overlapping paths, the text order is rearranged according to a difference set, paragraph splicing is completed, and a width threshold is calculated; and then a breathing detection gate and a reverse loop guide structure are used to dynamically control the detection range, so that the automatic suppression of circular reference and the stable control of paragraph merging are realized.

This invention discloses a method and system for fine-tuning gradient correction of synthetic data in asymmetric machine translation, relating to the field of neural machine translation technology. The method includes the following steps: constructing a hybrid training dataset; initializing the neural machine translation model and training parameters; calculating the gradients of real data and synthetic data respectively; performing unidirectional judgment based on the real data gradient to detect gradient conflict states; performing asymmetric gradient correction based on the gradient conflict state detection results; updating the parameters and verifying the convergence of the neural machine translation model; constructing a hybrid dataset and calculating the real data gradients of real parallel data and synthetic data respectively; when gradient conflict is detected, unidirectionally projecting the synthetic data gradient to the orthogonal direction of the real data gradient, preserving the semantic anchoring effect of the real data, avoiding semantic drift introduced by the synthetic data, improving translation quality and robustness, increasing computational efficiency, and exhibiting good adaptability and economy in low-resource scenarios.

The application provides a text classification model training method, a text classification method, a device, an electronic equipment and a computer readable storage medium; and relates to the technical field of artificial intelligence; the method comprises the following steps: calling a machine translation model based on a plurality of first text samples in a first language, so as to obtain a plurality of second text samples corresponding to the plurality of first text samples one by one; wherein the plurality of second text samples are in a second language different from the first language; training a first text classification model for the second language based on a plurality of third text samples in the second language and corresponding category labels; performing confidence-based screening processing on the plurality of second text samples by using the trained first text classification model; and training a second text classification model for the second language based on the second text samples obtained through the screening processing. Through the application, cross-language text samples can be automatically obtained, and the accuracy of text classification is improved.

The invention relates to a neural machine translation selective knowledge distillation method based on dependency constraint self-attention, and belongs to the technical field of machine translation. An existing knowledge distillation method has the problems that only vocabulary-level probability distribution is transmitted, syntactic structure constraints are ignored, and the capacity of a student model is reduced, so that the complex syntactic modeling capacity is insufficient. Therefore, according to the method, a syntactic matrix converted through a source language dependency syntactic tree is provided, linear combination is adopted to dynamically correct self-attention weight distribution of an encoder, and explicit syntactic constraints are synchronously injected into a teacher-student model. Through a selective distillation strategy of syntax perception, deep knowledge effective for training in a teacher model is screened, a cross-language syntax corresponding relation is obtained in combination with structure alignment distillation, and model compression and translation performance enhancement is realized through attention optimization and a selective knowledge transmission mechanism guided by a dependency syntax.

A system is provided.SOLUTION: A system comprising: means for electronically scanning and storing books as digital data; means for analyzing collected book data and extracting categories, abstracts, and key topics; means for inputting a user's favorite genre and past reading history; means for analyzing information input by a user and creating a profile; means for recommending books based on a user profile; means for displaying recommended books; means for providing books in multiple languages by automatic translation; means for automatically generating illustrations and videos; and means for converting stories of novels into video contents.SELECTED DRAWING: Figure 1

The application discloses a kind of automatic conversion method and system from Chinese to braille based on pre-training model, wherein the method comprises the following steps: constructing pre-training corpus, Chinese blind parallel corpus and machine translation model;The pre-training corpus and the Chinese blind parallel corpus are encoded, to obtain the encoded pre-training corpus and the encoded Chinese blind parallel corpus;The machine translation model is pre-trained based on the encoded pre-training corpus, to obtain pre-training model;The pre-training model is parameter fine-tuning based on the encoded Chinese blind parallel corpus, to obtain conversion model;Chinese is input into the conversion model for translation, to obtain braille sequence, complete Chinese blind translation.The application can convert Chinese into corresponding braille in one step, and greatly reduces the dependence of model on parallel data, and good effect can also be achieved using a small amount of data for training.

The invention discloses an ancient language machine translation method for syntactic perception and knowledge enhancement, and belongs to the technical field of natural language processing. Aiming at the difficulties of complex sentence patterns and scarcity of historical knowledge of ancient Chinese, three innovation modules are designed: firstly, a dynamic dependency syntactic analysis module explicitly models an ancient Chinese syntactic structure by utilizing probability analysis and a graph network; secondly, a retrieval enhancement generation module accurately extracts related knowledge fragments from the high-quality ancient language corpus, and semantic comprehension is enhanced; and finally, the stream grammar constraint decoder fuses the source language syntax and the target language generation process through a double-stream mechanism and a stream mechanism, and loyalty and smooth modern text output is realized. The three modules are deeply fused, so that the accuracy and interpretability of ancient language translation are effectively improved, and a reliable technical path is provided for ancient book digitization.

The invention discloses a drawing format conversion and machine translation method and system based on data recognition, and relates to the technical field of computer aided design file processing, and the method comprises the following steps: S1, establishing a unified coordinate baseline, recording the absolute position, the rotation angle and the size proportion of each character object, and generating an adjacent signature sequence according to a recording result; s2, utilizing the adjacent signature sequence to construct an adjacent chain with the same direction for each character object, distributing a unique chain sequence number according to the spatial characteristics of the preorder nodes, and generating an acyclic candidate set; according to the method, an adjacent signature sequence is generated by establishing a unified coordinate baseline, adjacent chains with consistent directions are constructed, and an acyclic candidate set is formed; overlapping fusing lines are set to eliminate overlapping paths, a character sequence is rearranged according to the difference value set to complete paragraph splicing, and a width threshold value is calculated; and the detection range is dynamically regulated and controlled through the breathing type detection gate and the reverse ring guide structure, so that automatic restraint of circular reference and stable control of paragraph combination are realized.

The application provides a machine translation model generation method and device, comprising: obtaining training data; training a machine translation model based on a pre-stored deep neural network using the training data; the machine translation model is used to translate a text to be translated into a text translation; the machine translation model comprises a memory enhancement adapter layer, a first memory and a second memory; the memory enhancement adapter layer is used to retrieve information in the first memory and / or the second memory and utilize the information; the first memory is used to save source language data obtained by performing reverse translation and forward calculation processing on the training data; and the second memory is used to save target language data obtained by performing reverse translation and forward calculation processing on the training data. The application saves the training data in a form acceptable to the model, reads information from the training data to assist the model when performing a translation task, and achieves the effect of adapting to various translation scenarios while taking into account lower algorithm time complexity and better model performance.

The invention discloses a small word block micro-context vocabulary learning system. The system comprises an access layer, a monitoring and log layer, an AI and translation layer, a service module layer, a data storage layer, a middleware layer and a terminal layer which are coordinated in sequence. The access layer distributes user requests through Nginx load balancing and guarantees service stability; the monitoring layer realizes operation and maintenance monitoring by means of service monitoring and an ELK log system; the AI layer is fused with a large model to generate a micro-context, and machine translation provides multi-scene translation; the business layer supports the whole vocabulary learning process through user and order modules; the data layer adopts multiple databases to adapt to different types of data storage; the middleware layer guarantees continuous integration, distributed transactions, asynchronous communication and timed tasks; and the terminal layer supports multi-terminal access of the Web and the mobile terminal. The system solves the problems of load, monitoring, micro-context, data storage, service reliability, multi-terminal adaptation and the like in the prior art, and improves the system stability, the vocabulary learning effect and the user experience.

The invention provides a translation processing method, a related device and a medium, and the method comprises the steps: putting a key matrix and a value matrix corresponding to a to-be-translated text into a first memory area, and generating an initial output lexical element corresponding to the to-be-translated text; executing a plurality of iteration steps; in each iteration step, generating a query vector, a key vector and a value vector of the iteration step based on the input of the iteration step, respectively updating a key matrix and a value matrix in the first memory area by using the key vector and the value vector, and performing attention processing based on the query vector and the updated key matrix and value matrix to obtain an output lexical element of the iteration step; generating a first threshold value based on the query vector of the iteration step and the updated key matrix, and performing vector filtering on the key matrix and the value matrix based on the first threshold value; and integrating the initial output lexical units and the output lexical units of the iteration steps into a translated text. According to the method, the storage pressure of machine translation on the memory can be reduced on the premise of not influencing the translation quality. The method is used for machine translation and other scenes.

The application provides a translation quality analysis method, device, equipment, medium and program product, and relates to the technical field of information processing. The method comprises the following steps: obtaining translation task information, wherein the translation task information comprises at least one of machine translation text, first modification mark information of the machine translation text, post-translation translation text, second modification mark information of the post-translation translation text and standard translation text; determining modification mark quality information of the translation task information according to a first similarity between the machine translation text and the standard translation text; determining post-translation translation quality information of the translation task information according to a second similarity between the post-translation translation text and the standard translation text and the second modification mark information; and determining translation quality information corresponding to the translation task information according to the post-translation translation quality information and the modification mark quality information.

The disclosure provides a text classification model training method, a text classification method, an apparatus, an electronic device, and a computer-readable storage medium, and relates to artificial intelligence technology. The text classification model training method includes: performing machine translation on a plurality of first text samples in a first language to obtain a plurality of second text samples in a second language different from the first language; training a first text classification model for the second language based on a plurality of third text samples in the second language and corresponding class labels; performing confidence-based filtering on the plurality of second text samples by the trained first text classification model; and training a second text classification model for the second language based on the filtered second text samples.