53results about How to "Improve translation speed" patented technology

A method for building a multilingual translation database system includes the steps of: providing a plurality of multilingual sentence pairs or multilingual sentence-fragment pairs in a translation database, each of the multilingual sentence pairs or the multilingual sentence-fragment pairs formed with a source language and a target language; selecting repeated sentence structures or repeated sentence fragments from the multilingual sentence pairs or the multilingual sentence-fragment pairs; and defining or qualifying at least one repeated key sentence or key sentence fragment of the repeated sentence structures or the repeated key sentence-fragment structures with a predetermined degree of repeat frequency.

The invention belongs to the field of computer-aided human translation and particularly relates to a method for displaying a translation memory match result. The method is characterized in that a computer-aided translation software finds matched translation memory in a database, a source and a match value of a translated text are displayed through different marking methods, quality level and pricing level of the translated text can be judged according to a mark of the source of the translated text, and a translator can judge the quality of the translated text according to quality requirements or pricing strategies, wherein the marking methods can be different fonts, word sizes, colors, positions, animations, sounds and combination of the factors and the like, and also can be realized in other distinguishing ways such as sound and animation prompts and the like. According to the method, different sources and match values of the translation memory are displayed by marking, so that the translator can judge the quality and the pricing strategies of phrases of the translated text in a short time so as to quickly determine adoption or non-adaption of the matched translation memory, thereby accelerating a translation speed and improving translation efficiency and translation quality to the greatest extent.

The invention provides a machine translation method and translation system based on organism named entities. Firstly, a full translation text is analyzed to obtain initial segmentation and part-of-speech tagging information; the organism named entities are recognized, the initial position of the organism named entities is labeled with an NEOC label, segmentation is carried out again on the basis, part-of-speech tagging is carried out, and parts of speech of the recognized organism named entities are tagged as NP; the organism named entities are translated, and the organism named entities and corresponding translation results are stored in an organism named entity translation library; the full translation text is translated with sentence as a unit. By means of the method and system, the recognition rate of the organism named entities is improved, and accuracy of syntactic analysis and translation of sentences where organisms occur is improved. Meanwhile, the organism named entities which occur for many times are translated only once, and translation speed and accuracy of the system are improved.

The invention discloses a multi-language machine intelligent auxiliary processing method and a system. The method comprises the following steps of acquiring a translated interval of a source language according to the source language and a translated target language so as to expand an untranslated interval of the source language; generating a K-best prompting set according to the expansion of the untranslated interval of the source language; generating an input historical prompting set and a language model prompting set according to the untranslated interval of the source language and the translated target language; and generating a candidate set of a target language corresponding to the untranslated source interval of the source language according to the input historical prompting set, the language model prompting set and the K-best prompting set. Due to the adoption of the method and the system, the multi-language translation speed is accelerated, the user experience is good, the intelligent degree is high, and the translation accuracy is improved.

The invention provides a machine translation data processing method, which comprises the following steps of obtaining translation results output by at least two translation modules; determining weighting weighted values of the translation modules; calculating basic scores of the translation results; calculating final scores of the translation results on the basis of the basic scores of the translation results and the weighting weighted values of the corresponding translation modules; selecting a part of or all translation results to output according to the final scores of the translation results. The invention also provides a machine translation data processing system capable of realizing the method and a terminal provided with the system. By using the machine translation data processing method, the system and the terminal, the translation query efficiency and the system response speed are improved.

The invention discloses a multi-interpreter cooperation translation system and a method thereof. The system comprises a translation terminal, a user terminal and a data service terminal which are connected through Internet. The data service terminal stores a memory bank of the translation terminal and the user terminal. The memory bank is language pairs generated when an interpreter carries out translation for a client. Each language pair includes a source language and a corresponding target language. The memory bank also stores information data of the translation terminal and the user terminal. The translation terminal is used for receiving a task for the interpreter and the interpreter carries out translation work on the translation terminal. On the user terminal, the client searches the interpreter and sends a translation task to the interpreter. Multi-interpreter cooperation translation is realized, low error inspection can be automatically performed, a file can be automatically split and a contrast file can be automatically generated, and then technical effects of high translation efficiency and high translation quality are realized.

The invention discloses a text display method and device. The method comprises the following steps of: obtaining a sample; the method comprises the following steps: receiving after a to-be-translated target voice is received; , recognizing the target voice to obtain a recognized text; extracting each first keyword from the recognition text; therefore, when the recognition text is displayed, each first keyword and other words except each first keyword in the recognition text can be displayed differently, so that a translator can quickly obtain each first keyword from the displayed recognition text in the subsequent translation process. Each first keyword is a word which should be specially concerned when a translator performs instant translation on the target voice; therefore, each first keyword in the recognition text and other words are displayed differently, and a translator can quickly pay attention to the important information in the recognition text, so that the translation speed of the translator can be increased, and the accuracy of a translation result can be improved.

The invention discloses a text data processing method for English translation. Implementation by a translation system, The translation system comprises a user login module, a data acquisition module,a data input module, a database and character recognition module, an emotion recognition module, a processing module, an output module, a display module and a sounding module. The data processing method comprises the following steps that S1, a database containing an English library and a Chinese library is established, modeling classification management is conducted on vocabularies and grammar inthe database through the processing module, a corresponding source code library is generated, and the source code library comprises a character source code library and a voice source code library; According to the method, the source code is generated for the information in the database, the root code is generated for the input data needing to be translated, the information needing to be translatedcan be quickly searched for through matching of the root code and the source code, the translation speed is increased, the translated voice is endowed with the emotion of a speaker, and the translated word and sentence better conform to the real meaning of the current speaker.

The invention relates to a label-based text translation system, which comprises: a marking module used for adding a start character and an end character to texts in different formats, and identifyinga position needing to be translated through the start character and the end character; the escape module marks the first row of spaces and symbols in the original text, so that the translation moduledoes not process the characters as formats; the translation module identifies an area needing to be translated in the text, compiles a translation program or calls a network interface to translate content needing to be translated in the mark; a lexicon records the proper nouns and is used for recording the one-to-one correspondence relationship between the proper nouns and the replacement character strings. According to the automatic translation system provided by the invention, the label is added into the original document, and a program automatically recognizes the area to be translated through label positioning, so that correct text content is translated in a specific format, and the text translation speed is increased.

A translation assistance system includes a storage unit for storing at least one font, each of the monosheets each including at least one monolingual list having n layers, each of the monolingual lists being stored separately A plurality of substitution sources and a plurality of substitution results, and the plurality of substitution sources of the mth monolingual list in the monochness comprises the plurality of substitution results of the mth monolingual word table; a user interface including a substitution process And an article display unit that imports and displays an article, the substitution processing unit searches for and replaces the sentence with at least one substitution means based on the alphabet, the substitution means includes first to n'th substitution means , Respectively corresponding to the first to n-th layer of the word list; and a processing unit, electrically connected to the memory unit and the user interface, and searching, according to the plurality of replacing means corresponding to the substitution processing unit, Substitutes the same character as the source, and replaces the character with the multiple substitutions.

The present invention relates to an electronic language translation aid. The aid includes a receiver for receiving a language sentence in a first language. A translator translates the language sentence into both a literal translation and a correct translation in a second language. A display displays the literal translation and the correct translation. Preferably, the display is configured to display both translations concurrently so that an early language learner can obtain an immediate comprehension of individual word meaning through literal translation, and overall contextual meaning through the correct translation.

A sawing or reducing assembly which includes at least one saw or reducing head such as a chipping head, at least one drive driving the saw or reducing head, and at least one drive line cooperating between the drive and the saw or reducing head, so as to rotatably drive the saw or reducing head. The saws or reducing heads, or at least some of them, are at least selectively transversely actuable so as to selectively translate transversely into a workpiece feed path as a workpiece is translated along the feed path into engagement with the workpiece. The drive is fixed in relation to the feed path. The drive line includes a corresponding slip joint allowing for relative movement between the drive and the saw or reducing head as the saw or reducing head is selectively translated transversely into the feed path.

The embodiment of the invention provides a speech translation method, a device, equipment and a storage medium. According to the method, an input voice signal is recognized and judged, at least one true meaning part is accumulated to form a judgment unit, a true meaning unit is obtained based on a BERT classification model and the judgment unit, and then the true meaning unit is segmented to form a target word segment; and finally, on the basis of a Transform translation model, translating and outputting the target word segment. By adopting the scheme, the length of the sentence needing to be translated is greatly shortened, the translation problem caused by complex semantics and statement parts in a long sentence is avoided, and the translation accuracy is improved; Meanwhile, voice input, target word segment formation and translation can be performed synchronously, so that the translation speed is increased.

The invention relates to a translation quality-based translation method, which comprises the following steps of: obtaining a to-be-translated file which comprises a plurality of to-be-translated statements; Obtaining an audit file and a first translation file according to the to-be-translated statement of the to-be-translated file; And obtaining a final translation file and a scoring result according to the audit file and the first translation file. According to the translation method, the mode of monitoring the quality of the first translation file in the translation process and monitoring the quality of the first translation file in the final examination process is adopted, the objectivity of the quality standard of the final examination translation is guaranteed, meanwhile, the qualityof the final examination translation is improved, and the translation speed is increased on the premise that the quality of the final examination translation is guaranteed.