CN113935339B

CN113935339B - Translation method, translation device, electronic equipment and storage medium

Info

Publication number: CN113935339B
Application number: CN202111014278.0A
Authority: CN
Inventors: 万星; 赵静璇; 王梦雪; 何中军; 吴华; 李芝; 李壮壮; 姚伟; 任云; 李国良
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2021-08-31
Filing date: 2021-08-31
Publication date: 2023-05-23
Anticipated expiration: 2041-08-31
Also published as: CN113935339A

Abstract

The disclosure provides a translation method, a translation device, electronic equipment and a storage medium, and relates to the technical fields of artificial intelligence such as big data, deep learning, natural language processing and the like. The specific implementation scheme is as follows: acquiring information of a document to be translated and a target language to be translated; identifying terms in the document to be translated based on the information of the document to be translated; and calling a professional term library based on the information of the document to be translated, the target language and the term, and translating the content of the document to be translated to obtain a first translation result. According to the technical scheme, the translation efficiency can be effectively improved.

Description

Translation method, translation device, electronic equipment and storage medium

Technical Field

The disclosure relates to the technical field of computers, in particular to the technical field of artificial intelligence such as big data, deep learning, natural language processing and the like, and particularly relates to a translation method, a translation device, electronic equipment and a storage medium.

Background

Translation is a very technical and specialized task, and traditional techniques rely on specialized translators to accomplish the translation. To improve translation efficiency, computer aided translation (Computer Aided Translation; CAT) tools have been created to assist translators in achieving high quality translations.

In the prior CAT use, in order to ensure the translation quality, a translator needs to prepare a term library by himself, upload and add the term library to the CAT before use.

Disclosure of Invention

The disclosure provides a translation method, a translation device, electronic equipment and a storage medium.

According to an aspect of the present disclosure, there is provided a translation method, wherein the method includes:

acquiring information of a document to be translated and a target language to be translated;

identifying terms in the document to be translated based on the information of the document to be translated;

and calling a professional term library based on the information of the document to be translated, the target language and the term, and translating the content of the document to be translated to obtain a first translation result.

According to another aspect of the present disclosure, there is provided a translation apparatus, wherein the apparatus includes:

the acquisition module is used for acquiring information of the document to be translated and a target language to be translated;

the identification module is used for identifying terms in the document to be translated based on the information of the document to be translated;

and the translation module is used for calling a professional term library based on the information of the document to be translated, the target language and the term, and translating the content of the document to be translated to obtain a first translation result.

According to still another aspect of the present disclosure, there is provided an electronic apparatus including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the aspects and methods of any one of the possible implementations described above.

According to yet another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the method of the aspects and any possible implementation described above.

According to yet another aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the method of the aspects and any one of the possible implementations described above.

According to the technology disclosed by the invention, the translation efficiency can be effectively improved.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.

Drawings

The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

FIG. 1 is a schematic diagram according to a first embodiment of the present disclosure;

FIG. 2 is a schematic diagram according to a second embodiment of the present disclosure;

FIG. 3 is a schematic diagram according to a third embodiment of the present disclosure;

FIG. 4 is a display interface provided by the present embodiment;

FIG. 5 is a view of another display interface provided by the present embodiment;

FIG. 6 is a further illustration of an interface provided by the present embodiment;

FIG. 7 is a schematic diagram according to a fourth embodiment of the present disclosure;

FIG. 8 is a schematic diagram according to a fifth embodiment of the present disclosure;

fig. 9 is a block diagram of an electronic device for implementing the translation method of an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

It will be apparent that the described embodiments are some, but not all, of the embodiments of the present disclosure. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments in this disclosure without inventive faculty, are intended to be within the scope of this disclosure.

It should be noted that, the terminal device in the embodiments of the present disclosure may include, but is not limited to, smart devices such as a mobile phone, a personal digital assistant (Personal Digital Assistant, PDA), a wireless handheld device, and a Tablet Computer (Tablet Computer); the display device may include, but is not limited to, a personal computer, a television, or the like having a display function.

In addition, the term "and/or" herein is merely an association relationship describing an association object, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship.

FIG. 1 is a schematic diagram according to a first embodiment of the present disclosure; as shown in fig. 1, the present embodiment provides a translation method, which specifically includes the following steps:

S101, acquiring information of a document to be translated and a target language to be translated;

s102, identifying terms in a document to be translated based on information of the document to be translated;

s103, based on the information of the document to be translated, the target language and the term, calling a special term library, and translating the content of the document to be translated to obtain a first translation result.

The execution subject of the translation method of the present embodiment is a translation apparatus. The translation means may be an electronic entity or may be an application employing software integration. For example, the translation device may be a CAT device.

When a user uses the translation device, the information of the document to be translated can be uploaded to the translation device, and the target language to be translated also needs to be uploaded. I.e. the user's purpose is to want to translate the document to be translated into a document described in the target language.

The translation method of the embodiment can be used for realizing the translation of the technical document. Translation of a technical document may include translation of grammar and translation of terms. As for grammar, grammar is generic regardless of the domain of the document term to be translated. The terms of the document to be translated have certain professionals and technologies, and the different fields can be translated or interpreted with different professionals, which belongs to the key points and difficulties in translation. In order to ensure the accuracy of translation, in this embodiment, terms in a document to be translated need to be recognized in order to ensure the accuracy of translation of terms in translation later.

And finally, calling a professional term library based on the information, the target language and the terms of the document to be translated, and translating the content of the document to be translated to obtain a first translation result. Wherein the term library may include terms described in a source language and translations of the terms in a target language. That is, in the translation process, the translation device of the embodiment obtains the translation of the term by calling the external, very specialized and authoritative specialized term library, and then the translation device translates the whole document to be translated from the source language to the target language based on the translation of the term.

The terms in this embodiment may refer to words, and in particular, may be understood as words that are necessary for repeated use in translation. The professional term library stores a plurality of terms and translations of target languages corresponding to the terms so as to ensure that translations of the same term are consistent throughout professional translations. At each translation, a term library may be invoked to obtain translations of terms.

The term library of art in this embodiment may be understood as an external information platform, which can obtain professional translations of terms, and store the terms and corresponding translations in the term library of art. Specifically, specialized translations of the terms may be obtained based on big data, or specialized translations of the terms by specialized translators may also be obtained. The term of art library may provide an application program interface (Application program interface; API) for the translation device of the present embodiment. When the translation device translates, for the terms needing to be translated, the API of the special term library can call the special term library to acquire the corresponding translation of the terms needing to be translated; and in the process of translating the document to be translated, translating the terms into translations acquired from the technical term library, and further translating the document to be translated.

Compared with the prior art, the translation method of the embodiment does not need the user to prepare and upload the term library by himself, and the translation device can automatically translate after uploading the information of the document to be translated, so that the operation of the user can be effectively reduced, and the translation efficiency is improved.

FIG. 2 is a schematic diagram according to a second embodiment of the present disclosure; the translation method of the present embodiment further describes the technical solution of the present application in more detail on the basis of the technical solution of the embodiment shown in fig. 1. As shown in fig. 2, the translation method of the present embodiment may specifically include the following steps:

s201, acquiring content, source language, field of a document to be translated and target language to be translated, which are uploaded by a user;

this step is a specific implementation manner of step S101 in the embodiment shown in fig. 1. Specifically, in step S201, the content of the document to be translated, the source language, and the field of the document to be translated are included in the information of the document to be translated in step S101 as examples. In practical application, if only one external technical term library exists, and the fields are not distinguished, the information of the document to be translated can also only include the content of the document to be translated, and the information does not need to carry field information and source language information, even does not need target language information, and corresponding translation information is directly obtained based on terms. Or the target languages of the external technical term libraries are the same, and when the technical term libraries need to be positioned according to the source language, only the source language is needed to be acquired at the moment, and the target language and the domain information are not needed. In short, based on at least one of the source language, the domain and the target language of the document to be translated, and combining the content of the document to be translated, the translation requirements of different scenes can be met.

In addition, optionally, other information, such as author, year, and other auxiliary information, etc., may be carried in the information of the document to be translated, so that the translation is performed together during the translation. Or when more technical term libraries are included, the corresponding technical term libraries can be called more accurately by referring to the auxiliary information. For example, the domain has more terms, and the term library can be split based on other auxiliary information.

The field of the present embodiment may be a professional field, such as a field with relatively strong professionals of communication, optics, nuclear physics, sports, financial accounting, and the like. Or may be non-professional, biased toward a general purpose general-purpose domain.

The source language of the document to be translated in this embodiment refers to a description language of the document to be translated, and the source language and the target language are different, but may be languages such as chinese, english, french, korean, japanese, etc.

S202, word segmentation is carried out on each sentence in the content of a document to be translated, and a plurality of word segmentation is obtained;

s203, extracting terms from the plurality of segmented words;

steps S202-S203 are a specific implementation of identifying terms in the document to be translated in step S102 of the embodiment shown in fig. 1 described above.

Specifically, when the document to be translated is segmented, each sentence in the document to be translated can be segmented in sequence according to the sequence from front to back, when the word is segmented, each word in the sentences can be used as a segmentation segment according to word segmentation strategies, and after all the sentences in the document to be translated are segmented, a plurality of segmented words can be obtained.

In this embodiment, extracting terms from a plurality of segmented words may include the following ways:

(1) Filtering word fragments with word frequencies greater than a preset threshold value in the general field from a plurality of word fragments to obtain terms;

in practical application, word frequency of each word in the general field and each professional neighborhood in the network can be counted based on big data technology. For example, the word frequency of each word appearing is obtained by analyzing all information in the network in a certain area.

Considering that the term of art cannot be high in word frequency in the general field, a certain word frequency occurs only in the corresponding field. Therefore, the word fragments having the word frequency greater than the preset threshold value in the general field can be filtered out, and the word fragments cannot be terms in the professional field. In this way, the accuracy of the resulting term can be improved.

(2) Filtering the word fragments belonging to the stop word list from a plurality of word fragments to obtain a term;

in practical applications, the stop word list including all stop words can be counted in advance based on big data technology. The stop words are then filtered out of the plurality of word segments to improve the accuracy of the resulting term.

Alternatively, steps (1) and (2) may be performed sequentially.

(3) A pre-trained term recognition model is employed to recognize terms from the plurality of segmentations.

When the scheme is implemented, a plurality of segmented words can be input into a term recognition model, the term recognition model can predict and output the probability that each segmented word belongs to a term, and if the probability is larger than a preset probability threshold value such as 0.6, 0.7 or other preset numerical values, the segmented word is considered to belong to the term.

Specifically, a plurality of terms and non-term words in some general fields can be collected in advance, and the term recognition model can be trained. Where the table of terms is 1 and the table of words of non-terms is 0. During training, each term is input into a term recognition model, the term recognition model predicts and outputs the probability of the term, and if the probability is smaller than 1, parameters of the term recognition model are adjusted so that the probability tends to 1. For each non-term, the non-term is input into a term recognition model, the term recognition model outputs the probability corresponding to the non-term, and if the probability is not 0, the parameters of the term recognition model are adjusted so that the probability tends to 0. According to the method, the term recognition model is continuously trained by adopting the collected terms and the non-terms until the predicted result is consistent with the table identification result, and the training is finished, so that the parameters of the term recognition model are determined, and the term recognition model is further determined.

In addition, optionally, in order to improve accuracy of term recognition, a term recognition model may be trained correspondingly for each domain. During training, a plurality of terms in the field and some vocabularies not belonging to the field are collected to train the term identification model, and the training principle is the same as that described above, and the description is omitted here. In addition, in this embodiment, the terms may be extracted from the plurality of word segments in other manners, which is not described herein in detail.

In this embodiment, the above term identification scheme is adopted, so that accuracy and comprehensiveness of the identified term can be effectively improved in any mode.

S204, according to the source language of the document to be translated, the field of the document to be translated, the target language to be translated and the obtained term, calling a professional term library, and translating the content of the document to be translated to obtain a first translation result;

specifically, according to the source language of the document to be translated, the field of the document to be translated and the target language to be translated, a technical term library can be accurately positioned. The term library may then be invoked to obtain an interpretation, i.e., a translation, of the target language to which each term corresponds. Then, in the process of translating the document to be translated, the translation of the term can be replaced by the translation obtained from the professional term library, and based on the translation, all translations of the document to be translated are completed, so that a first translation result is obtained.

S205, performing column display based on a document to be translated and a first translation result;

in this embodiment, after the first translation result is obtained, the translation device uses the first translation result as only one preliminary translation result, and there may be a problem of translation error, so the preliminary translation result is not the final translation result. The user is required to check the first translation result, and the first translation result can be edited in the user checking process. Therefore, in order to facilitate the user to check and edit, in this embodiment, the document to be translated and the first translation result may be displayed in columns, so that one column displays the document to be translated, and the other column displays the first translation result, so that the user can completely check which sentence of the first translation result corresponds to each sentence of the document to be translated, which is very convenient to use.

S206, receiving editing information of a user on the first translation result;

specifically, if a user finds that a certain translation is not accurate during the verification process, the first translation result may be edited, where the editing may include adding, deleting, or modifying. The user may input editing information to the translation device through the human interface model. In practical application, the first translation result may be a final translation result, but since the first translation result completely belongs to a translation result of the translation device, in order to further improve the accuracy and the translation efficiency of the translation, in this embodiment, the user may further calibrate the first translation result, and edit the first translation result when the first translation result is inaccurate.

S207, editing the first translation result based on editing information of the user on the first translation result to obtain a second translation result.

After receiving the editing information input by the user through the man-machine interface module, the translation device can modify the first translation result based on the editing information to obtain a second translation result. This process may be referred to as post-translation editing for improving translation accuracy and translation efficiency. After the user's collation and editing, the second translation result may be used as the final target translation result.

According to the translation method, by adopting the scheme, the corresponding technical term library can be called based on the source language of the document to be translated, the field of the document to be translated, the target language to be translated and the term of the document to be translated, and the document to be translated is translated, so that the first translation result can be accurately obtained. Further, the document to be translated and the first translation result can be displayed in columns, so that the user can effectively edit the first translation result. And further based on editing information of the user, editing the first translation result to obtain a second translation result, so that a more accurate translation result can be obtained, the translation efficiency can be effectively improved, and the accuracy of the translation result can be improved.

FIG. 3 is a schematic diagram according to a third embodiment of the present disclosure; the translation method of the present embodiment further describes the technical solution of the present application in more detail on the basis of the technical solution of the embodiment shown in fig. 1. As shown in fig. 3, the translation method of the present embodiment may specifically include the following steps:

s301, acquiring the content of a document to be translated uploaded by a user and a target language to be translated;

s302, based on the content of a document to be translated, identifying the field of the document to be translated by adopting a pre-trained field identification model;

unlike the embodiment shown in fig. 2, in this embodiment, when the user does not inform the domain of the document to be translated, the domain recognition model may be used to recognize the domain of the document to be translated.

Specifically, since the content of a document is generally briefly summarized in the abstract of the document, the domain of the document can be identified based on the abstract of the document. Alternatively, the abstract, or the abstract heading, of the document to be translated may be input into a domain identification model, which may predict the domain of the document to be translated based on the input information. Further, in order to improve accuracy of the domain identification model, the whole document to be translated can be input into the domain identification model to predict the domain of the document to be translated. Compared with the previous mode, as the input information is increased, the time for identifying the domain is prolonged, and the efficiency of identifying the domain is reduced.

Prior to use, the domain identification model needs to be trained in advance. Specifically, taking only the abstract of the input document at the time of domain identification as an example. Before training, a plurality of pieces of training data can be collected in advance, and each piece of training data comprises a abstract of a training document and a labeled field of the document. During training, the abstract of the training document in each piece of training data is input into a domain identification model, and the domain identification model predicts and outputs the domain of the corresponding document. And then detecting whether the predicted result and the marked result are consistent, and if not, adjusting parameters of the field identification model to enable the predicted result and the marked result to be consistent. According to the mode, the collected multiple pieces of training data are adopted to train the domain identification model continuously until the predicted result is consistent with the marked result, parameters of the domain identification model are determined, and then the domain identification model is determined.

The method can effectively improve the accuracy and the intelligence of the domain identification in the domain of the document to be translated identified by adopting the method.

S303, identifying the source language of the document to be translated by adopting a pre-trained language identification model based on the content of the document to be translated;

the language recognition model in this embodiment is similar to the implementation principle and training principle of the above-mentioned domain recognition model, and details of the description of the above-mentioned domain recognition model may be referred to, which is not described herein again.

The source language of the document to be translated identified by adopting the method can effectively improve the accuracy and the intelligence of source language identification.

S304, acquiring a term extraction model corresponding to the field based on the field of the document to be translated;

in this embodiment, different term extraction models are adopted in different fields, so that accuracy of term extraction can be improved.

S305, extracting terms of the document to be translated from the content of the document to be translated by adopting a term extraction model corresponding to the field;

specifically, each sentence in the content of the document to be translated is respectively input into a term extraction model, and the term extraction model can extract a corresponding term from the sentence. If there is no term in a sentence, the extraction may be empty.

The term extraction model may collect a plurality of training sentences in advance before training, and annotate corresponding terms in the sentences. During training, any training sentence is input into a term extraction model, and the term extraction model can predict and output the terms in the training sentence. And then comparing the predicted result with the marked result, and if the predicted result and the marked result are inconsistent, adjusting parameters of the term extraction model to enable the predicted result and the marked result to be consistent. And training the term extraction model continuously by adopting a plurality of training sentences according to the mode until the predicted result is consistent with the marked result, and determining parameters of the term extraction model after the training is finished, so as to determine the term extraction model.

Similarly, optionally, a term extraction model may be trained for each domain. The training principle is the same as above and will not be described in detail here.

Alternatively, in this embodiment, the steps S202 to S203 in the embodiment shown in fig. 2 may be used to implement the term extraction.

S306, based on the source language of the document to be translated, the target language to be translated, the field of the document to be translated and the terms, calling a professional term library by referring to a local term library and a memory library, and translating the document to be translated to obtain a first translation result;

when the step is specifically implemented, when each sentence in the document to be translated is translated, firstly, whether the sentence has corresponding translation or not can be detected based on a local memory bank, if not, whether the translation of the corresponding term in the sentence exists in the term bank is detected by referring to the local term bank, and if so, in the translation process of the sentence, the translation of the term adopts the translation of the corresponding term in the local term bank. If the specific term is not available, a special term library can be accurately positioned according to the source language of the document to be translated, the field of the document to be translated and the target language to be translated. Next, translations of the target language for which the term corresponds may be obtained from the term store. Then, in the process of translating the document to be translated, the translation of the term can be replaced by the translation obtained from the professional term library, and based on the translation, all translations of the document to be translated are completed, so that a first translation result is obtained. That is, in the translation process, the priority of the memory bank is highest, the local term bank is next, and the priority of the professional term bank is lowest. Wherein, the local term library and the local memory library are stored locally based on the user's historical translation information or the user's requirements. The term of art library can be understood as an external term of art library which is more specialized and authoritative, and is created without reference to any user's historical translation information. The priority of the term library is lowest at the time of translation.

Wherein the local term library can be generated by the user and uploaded to the translation device by the user in a prior art manner. The terms appearing in the term library can be any vocabulary appearing in the translation, if the terms are reused, the user can save the terms as terms, and meanwhile, the corresponding translations of the terms are saved, and the saved multiple terms and the corresponding translation set are called a term library. The term library generated in the embodiment can be reused, not only in the translation, but also in the translation work of later projects or other people, so as to improve the efficiency and solve the translation consistency problem. Alternatively, the term library may be generated by means of sedimentation accumulation during use without being generated by the user beforehand.

The memory bank is generated based on historical usage of the user. In the translation process, some sentences which are strong in technical skill and possibly can be reused and the translation of the sentences are stored in the memory library, so that the subsequent translation efficiency is improved, and the problem of translation consistency is solved. The memory bank is generated by means of precipitation accumulation. The memory bank may be empty when first used.

S307, based on the document to be translated and the first translation result, performing column display; meanwhile, at the interface of column display, the terms in the document to be translated and the translations of the terms corresponding to the terms in the professional term library are displayed;

the difference from S205 in the embodiment shown in fig. 2 described above is that in this step, in the column display interface, terms in the document to be translated and translations of the terms corresponding to those in the term library are also displayed. For example, the term in the document to be translated and the translation of the term corresponding to the term in the term of art library may be used as the third column. Alternatively, based on the user's review habits, the document to be translated and the first translation result may be presented in the first two columns of the interface, and the terms in the document to be translated and the translations of the terms corresponding to the terms in the term store may be presented in the right-most column. In this way, when the user checks, clicking on each sentence in the document to be translated in the interface, the translation device may identify the translation corresponding to the sentence in the first translation result, for example, may identify the translation by using a highlighting, underline, or other means. Meanwhile, the translation device can display the terms in the sentence and the corresponding interpretations in the professional term library in a third column for the user to check.

Optionally, in this embodiment, the translation device may also call a term library of the other fields to obtain translations of the terms in the other fields, and display the translations in the third column. When the user checks, according to the translation of the term in the current field displayed in the third column, the translation of the term in other fields can be referred to at the same time to determine whether the translation of the term is accurate and proper, if not, the translation of the corresponding term in the first translation result can be edited by referring to the translation of the other fields, so that the translation of the term is modified to be considered to be accurate by the user. For example, fig. 4 is a display interface provided in this embodiment. The display interface shown in fig. 4 is the display interface corresponding to this step.

S308, at least one of the number of the identified terms, the number of the local memory banks used in translation and the number of the local term banks used in translation is displayed;

according to the translation method of the embodiment, a user does not need to upload the term library by himself, and only the document to be translated is uploaded to realize translation. Therefore, the perception of the user is less. By displaying the information, the perception of the user can be enhanced, and the user can know the specific condition of the translation.

It should be noted that, the information displayed in step S308 may be displayed to the user in a form of a frame above the column display interface in S307, so as to inform the user of the translation condition. After the user finishes reading, the bullet frame is closed, and the candidate post-translation editing of the user is not affected. For example, fig. 5 is another display interface provided in this embodiment. The display interface shown in fig. 5 is a display interface corresponding to the step, and is used for simultaneously displaying the number of recognized terms, the number of local memory banks used in translation, and the number of local term banks used in translation.

S309, receiving editing information of a user on the first translation result on a column display interface;

s310, editing the first translation result based on editing information of the user on the first translation result to obtain a second translation result;

similarly, the first translation result in this embodiment is a preliminary translation result, and the second translation result is a translation result after user verification and editing, which may be used as a final target translation result.

The specific implementation manner of step S309 and step S310 is the same as the implementation manner of step S206 and step S207 in the embodiment shown in fig. 2, and the details of the foregoing embodiment may be referred to the description of the foregoing embodiment, which is not repeated herein.

S311, updating a local term library based on terms in the document to be translated and the translation of the terms in the second translation result;

in particular, the method comprises the following steps:

(a) Detecting terms in a document to be translated, which are not included in a term library;

(b) Popping up a prompt interface, wherein the prompt interface comprises terms which are not included in a term library in a document to be translated and corresponding translated prompt interfaces so as to prompt a user whether to update the term library locally;

it should be noted that, the same prompt interface may include a plurality of terms, and each term may be separately provided with a determination button and a deletion button, so that the user agrees to update the term to the local term library, clicks to determine, and if not easy, clicks to delete. After all terms are operated, the update operation is triggered by determining the click interface. Fig. 6 is a further illustration of an interface provided by the present embodiment. The display interface shown in fig. 6 is the prompt interface corresponding to the step.

(c) If the user is detected to determine updating, the terms and corresponding translations which are not included in the term library in the document to be translated are updated into the local term library.

If the user does not prepare the term library by himself, the term library is empty when the term library is used for the first time. Through multiple uses, the local term library can be continuously updated, so that a great number of terms and corresponding explanations can be deposited in the term library, thereby effectively improving the subsequent translation efficiency.

S312, identifying technical sentences in the document to be translated;

in particular, the recognition process may employ a pre-trained technical sentence recognition model to recognize. Or other preset rules, such as a screening template with a plurality of technical sentences, can be set to identify the technical sentences.

Specifically, when the technical sentence recognition model is adopted for recognition, each sentence in the document to be translated can be input into the technical sentence recognition model, the technical sentence recognition model can predict and output the probability that the sentence belongs to the technical sentence, if the probability is larger than a preset probability threshold, the sentence belongs to the technical sentence, and otherwise, the sentence does not belong to the technical sentence.

Before training, the technical sentence recognition model needs to collect a plurality of pieces of training data in advance, wherein each piece of training data comprises a training sentence and the probability that the labeled sentence belongs to the technical sentence. Specifically, there are positive examples and negative examples of the training data collected. The probability corresponding to the positive example is 1, and the probability corresponding to the negative example is 0. During training, training sentences in the training data are input into a technical sentence recognition model, the technical sentence recognition model can predict and output the probability that the corresponding training sentences belong to technical sentences, then whether the predicted probability is consistent with the labeling probability is detected, and if not, parameters of the technical sentence recognition model are adjusted so that the predicted probability tends to the labeling probability. According to the mode, the technical sentence recognition model is continuously trained by adopting a plurality of pieces of training data until the predicted probability is consistent with the labeling probability, the training is finished, the parameters of the technical sentence recognition model are determined, and then the technical sentence recognition model is determined.

S313, updating a local memory bank based on the technical sentences and the translation of the technical sentences in the second translation result.

The method is used for realizing a sediment accumulation memory bank so that technical sentences can be translated rapidly by means of the memory bank in subsequent translation, the consistency of the technical sentence translation is ensured, and the translation efficiency can be effectively improved.

By adopting the scheme, the translation method of the embodiment not only can obtain more accurate translation results, but also can effectively improve the translation efficiency and the accuracy of the translation results. Meanwhile, a local term library and a memory library can be updated in a precipitation accumulation mode, so that the consistency of terms and technical sentence translation in subsequent translation can be effectively ensured, and the subsequent translation efficiency can be effectively improved.

FIG. 7 is a schematic diagram according to a fourth embodiment of the present disclosure; as shown in fig. 7, the present embodiment provides a translation apparatus 700, including:

an obtaining module 701, configured to obtain information of a document to be translated and a target language to be translated;

an identifying module 702, configured to identify terms in a document to be translated based on information of the document to be translated;

the translation module 703 is configured to invoke a term library based on information, a target language, and terms of the document to be translated, and translate content of the document to be translated to obtain a first translation result.

The translation device of the present embodiment, by adopting the implementation principle and the technical effect of implementing the translation by using the above modules, is the same as the implementation of the above related method embodiments, and details of the above related embodiments may be referred to the description of the above related embodiments, which is not repeated herein.

FIG. 8 is a schematic diagram according to a fourth embodiment of the present disclosure; the translation apparatus 700 of the present embodiment further describes the technical solution of the present application in more detail on the basis of the technical solution of the embodiment shown in fig. 7.

As shown in fig. 8, in the translation apparatus 700 of the present embodiment, the identification module 702 includes:

a word segmentation unit 7021, configured to segment each sentence in the content of the document to be translated, so as to obtain a plurality of segmented words;

an extraction unit 7022 for extracting terms from the plurality of segmentations.

Further optionally, the extracting unit 7022 is configured to:

filtering word fragments with word frequencies greater than a preset threshold value in the general field from a plurality of word fragments to obtain terms; and/or

And filtering the word fragments belonging to the stop word list from the plurality of word fragments to obtain the term.

Further alternatively, the extracting unit 7022 is configured to:

a pre-trained term recognition model is employed to recognize terms from the plurality of segmentations.

As a further alternative, as shown in fig. 7, in the translation apparatus 700 of the present embodiment, the obtaining module 701 includes:

A first acquisition unit 7011 for acquiring the content of a document to be translated;

a second acquisition unit 7012 configured to acquire a target language to be translated;

a third acquiring unit 7013 configured to acquire a source language of a document to be translated; and

a fourth acquisition unit 7014 is used for acquiring the field of the document to be translated.

Wherein the third acquisition unit 7013 is used for

Acquiring the field of a document to be translated uploaded by a user; or alternatively

Identifying the field of the document to be translated by adopting a pre-trained field identification model;

wherein the fourth acquisition unit 7014 is configured to:

acquiring a source language of a document to be translated uploaded by a user; or alternatively

And identifying the source language of the document to be translated by adopting a pre-trained language identification model.

Further optionally, a translation module 703 is configured to:

and calling a professional term library based on the source language, the field, the target language and the terms of the document to be translated, and translating the content of the document to be translated to obtain a first translation result.

As a further alternative, as shown in fig. 8, the translation apparatus 700 of the present embodiment further includes:

and the display module 704 is used for performing column display based on the document to be translated and the first translation result.

Further optionally, the display module 704 is further configured to:

And displaying the terms in the document to be translated and the translations of the terms in the professional term library corresponding to the terms in the interface displayed in columns.

a receiving module 705, configured to receive edit information of a first translation result from a user;

the editing module 706 is configured to edit the first translation result based on the editing information of the first translation result by the user, so as to obtain a second translation result.

Further optionally, a translation module 703 is configured to:

and based on the information of the document to be translated, the target language and the terms, referring to a local term library and a memory library, calling a special term library, and translating the document to be translated to obtain a first translation result.

Further optionally, the display module 704 is further configured to:

at least one of the number of recognized terms, the number of memory banks used locally in translation, and the number of term banks used locally in translation is presented.

an updating module 707 for updating the local term library based on the terms in the document to be translated and the translation of the terms in the second translation result.

Further optionally, the updating module 707 is configured to:

detecting terms in a document to be translated, which are not included in a term library;

popping up a prompt interface, wherein the prompt interface comprises terms which are not included in a term library and are included in a document to be translated, and corresponding translated prompt interfaces so as to prompt a user whether to update the terms in the local term library;

if the user is detected to determine updating, the terms and corresponding translations which are not included in the term library in the document to be translated are updated into the local term library.

Further optionally, the identifying module 702 is further configured to identify technical sentences in the document to be translated;

the updating module 707 is further configured to update the local repository based on the technical sentence and the translation of the technical sentence in the second translation result.

In the technical scheme of the disclosure, the acquisition, storage, application and the like of the related user personal information all conform to the regulations of related laws and regulations, and the public sequence is not violated.

According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

Fig. 9 shows a schematic block diagram of an example electronic device 900 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 9, the apparatus 900 includes a computing unit 901 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 902 or a computer program loaded from a storage unit 908 into a Random Access Memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the device 900 can also be stored. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other by a bus 904. An input/output (I/O) interface 905 is also connected to the bus 904.

Various components in device 900 are connected to I/O interface 905, including: an input unit 906 such as a keyboard, a mouse, or the like; an output unit 907 such as various types of displays, speakers, and the like; a storage unit 808, such as a magnetic disk, optical disk, etc.; and a communication unit 909 such as a network card, modem, wireless communication transceiver, or the like. The communication unit 909 allows the device 900 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunications networks.

The computing unit 901 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 901 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 performs the respective methods and processes described above, such as a translation method. For example, in some embodiments, the translation method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 900 via ROM 802 and/or communication unit 909. When the computer program is loaded into RAM 903 and executed by the computing unit 901, one or more steps of the translation method described above may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform the translation method by any other suitable means (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel or sequentially or in a different order, provided that the desired results of the technical solutions of the present disclosure are achieved, and are not limited herein.

The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. A translation method, wherein an execution subject of the method is a translation apparatus, the method comprising:

based on the information of the document to be translated, the target language and the term, calling a professional term library to translate the content of the document to be translated to obtain a first translation result;

The method further comprises the steps of calling a professional term library based on the information of the document to be translated, the target language and the term, translating the content of the document to be translated, and obtaining a first translation result, wherein the method further comprises:

positioning the external professional term library to be used according to the source language of the document to be translated, the field of the document to be translated and the target language to be translated;

the method comprises the steps of calling a professional term library based on information of the document to be translated, the target language and the term, translating the content of the document to be translated to obtain a first translation result, wherein the first translation result comprises the following steps:

calling the special term library through an application program interface of the external special term library to acquire the translation of the target language of the term;

and translating the content of the document to be translated based on the information of the document to be translated and the target language, and translating the term into the translation obtained from the professional term library in the translation process, so as to realize the translation of the document to be translated and obtain the first translation result.

2. The method of claim 1, wherein identifying terms in the document to be translated based on information of the document to be translated comprises:

Word segmentation is carried out on each sentence in the content of the document to be translated, so that a plurality of word segments are obtained;

the term is extracted from the plurality of segmentations.

3. The method of claim 2, wherein extracting the term from the plurality of tokens comprises:

filtering word fragments with word frequencies greater than a preset threshold value in the general field from the plurality of word fragments to obtain the term; and/or

Filtering the word fragments belonging to the stop word list from the plurality of word fragments to obtain the term.

4. The method of claim 2, wherein extracting the term from the plurality of tokens comprises:

the term is identified from the plurality of segmentations using a pre-trained term identification model.

5. The method of claim 1, wherein obtaining information of the document to be translated comprises:

acquiring the content of the document to be translated; or alternatively

And acquiring the content of the document to be translated, the source language of the document to be translated and the field of the document to be translated.

6. The method of claim 5, wherein obtaining the source language of the document to be translated comprises:

acquiring a source language of the document to be translated uploaded by a user; or alternatively

Identifying the source language of the document to be translated by adopting a pre-trained language identification model;

the obtaining of the field of the document to be translated comprises the following steps:

acquiring the field of the document to be translated uploaded by the user; or alternatively

And identifying the field of the document to be translated by adopting a pre-trained field identification model.

7. The method according to claim 5 or 6, wherein, based on the information of the document to be translated, the target language and the term, invoking a term library to translate the content of the document to be translated, to obtain a first translation result, including:

and calling the professional term library based on the source language, the field, the target language and the term of the document to be translated, and translating the content of the document to be translated to obtain the first translation result.

8. The method according to claim 6, wherein the method comprises:

and performing column display based on the document to be translated and the first translation result.

9. The method of claim 8, wherein the method further comprises:

10. The method according to claim 8 or 9, wherein the method further comprises:

receiving editing information of the user on the first translation result;

and editing the first translation result based on the editing information of the user on the first translation result to obtain a second translation result.

11. The method of claim 10, wherein, based on the information of the document to be translated, the target language and the term, invoking a term of art library to translate the document to be translated to obtain a first translation result, including:

and based on the information of the document to be translated, the target language and the term, referring to a local term library and a memory library, calling a special term library, and translating the content of the document to be translated to obtain the first translation result.

12. The method of claim 11, wherein, based on the information of the document to be translated, the target language, and the term, referring to a local term library and a memory library, invoking a professional term library to translate the document to be translated, and after obtaining the first translation result, the method further comprises:

at least one of the number of the recognized terms, the number of the memory banks used locally in translation, and the number of the term banks used locally in translation is displayed.

13. The method of claim 11, wherein the method further comprises:

updating the local term library based on the term in the document to be translated and the translation of the term in the second translation result.

14. The method of claim 13, wherein updating the local term library based on the term in the document to be translated and the translation of the term in the second translation result comprises:

detecting terms in the document to be translated, which are not included in the term library;

popping up a prompt interface, wherein the prompt interface comprises terms which are not included in the term library in the document to be translated and corresponding translated prompt interfaces so as to prompt a user whether to update the term library locally;

if the user is detected to determine updating, updating the terms and corresponding translations which are not included in the term library in the document to be translated into the local term library.

15. The method according to any one of claims 11-14, wherein the method further comprises:

identifying technical sentences in the document to be translated;

Updating the local memory bank based on the technical sentence and the translation of the technical sentence in the second translation result.

16. A translation apparatus, wherein the apparatus comprises:

the translation module is used for calling a professional term library based on the information of the document to be translated, the target language and the term, and translating the content of the document to be translated to obtain a first translation result;

the translation module is used for positioning the external professional term library to be used according to the source language of the document to be translated, the field of the document to be translated and the target language to be translated;

the translation module is used for:

17. The apparatus of claim 16, wherein the identification module comprises:

the word segmentation unit is used for segmenting each sentence in the content of the document to be translated to obtain a plurality of segmented words;

and an extraction unit for extracting the term from the plurality of segmented words.

18. The apparatus of claim 17, wherein the extraction unit is configured to:

19. The apparatus of claim 17, wherein the extraction unit is configured to:

20. The apparatus of claim 16, wherein the acquisition module comprises:

the first acquisition unit is used for acquiring the content of the document to be translated;

the second acquisition unit is used for acquiring a target language to be translated;

a third obtaining unit, configured to obtain a source language of the document to be translated; and

and the fourth acquisition unit is used for acquiring the field of the document to be translated.

21. The apparatus of claim 20, wherein the third acquisition unit is configured to

Acquiring the field of the document to be translated uploaded by a user; or alternatively

the fourth acquisition unit is configured to:

acquiring a source language of the document to be translated uploaded by the user; or alternatively

22. The apparatus of claim 20 or 21, wherein the translation module is configured to:

23. The apparatus of claim 21, wherein the apparatus further comprises:

and the display module is used for carrying out column display based on the document to be translated and the first translation result.

24. The apparatus of claim 23, wherein the presentation module is further to:

25. The apparatus of claim 23 or 24, wherein the apparatus further comprises:

the receiving module is used for receiving editing information of the first translation result from the user;

and the editing module is used for editing the first translation result based on the editing information of the user on the first translation result to obtain a second translation result.

26. The apparatus of claim 25, wherein the translation module is configured to:

27. The apparatus of claim 26, wherein the presentation module is further configured to:

28. The apparatus of claim 26, wherein the apparatus further comprises:

and the updating module is used for updating the local term library based on the terms in the document to be translated and the translation of the terms in the second translation result.

29. The apparatus of claim 28, wherein the update module is configured to:

30. The apparatus of any one of claims 28-29, wherein:

the identification module is also used for identifying technical sentences in the document to be translated;

the updating module is further configured to update the local memory bank based on the technical sentence and the translation of the technical sentence in the second translation result.

31. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-15.

32. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-15.