JPH05324716A - Translation support system - Google Patents

Abstract

PURPOSE:To standardize equivalents to technical terms even at the time of divisional translation by inhibiting translators from translating technical terms registered in a term dictionary, and then substituting equivalents for the terms by referring the term dictionary at the time of editing after the divisional translation. CONSTITUTION:An original sentence when inputted from an input part 1 to an editing process part 4 is sent to a discrimination mark and abbreviation addition part 8 and collated with the term dictionary 2 to add discrimination marks for translation inhibition to the technical terms registered in the dictionary. Plural translators translate only parts other than the terms in the original sentence that the discrimination marks are added and leave the terms to which the discrimination marks are added as the original words. Translations obtained by the translators are inputted to an editing process part 4 through a communication part 6 or input part 1. The editing process part 4 The translations are put together and edited into one document, which is inputted to a translation substituting process part 9, which substitutes the equivalents for the untranslated technical terms. Equivalents to the technical terms registered in the term dictionary 2 are therefore standardized.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a translation support system applied to a translation process mainly composed of manual translation work.

[0002]

2. Description of the Related Art Recently, there is an increasing demand for rapidly translating a large amount of documents such as documents related to science and technology. In response to such a request, a form of so-called divided translation is adopted, in which a series of documents are divided by a plurality of translators to proceed with translation work. The translation result thus created by a plurality of translators is edited by one editor to obtain a final translated sentence.

[0003]

When the above-described divisional translation is performed, it is inevitable that the translator will unify the translation of the technical term. In particular, in the case of divisional translation, mistranslation of technical terms tends to occur because individual translators cannot grasp the content of the entire document, so the problem of unification of translated words is significant. Therefore, the editor has to perform a complicated and time-consuming work of unifying the unification of translated words when editing, which causes a reduction in the overall translation efficiency.

Further, when such editing work is performed, it is difficult to completely eliminate the incompatibility or mistranslation of technical terms due to a collation error between the original sentence and the translated sentence, and the quality of the translated sentence finally obtained deteriorates. Occurs. That is, in technical documents, if the translations of technical terms are inconsistent, the description of the translated text becomes unclear, and the correct meaning may not be grasped unless the translated text is compared with the original text.

Further, from the standpoint of individual translators,
Even though it takes a lot of time to look up the translation of a technical term, it is often not finally adopted, and the effort required to look up the translation is not rewarded.

According to the present invention, even when a plurality of translators divide and translate a series of documents, translations of technical terms can be almost completely unified, and at the same time, translation efficiency can be improved and translation speed can be increased. The purpose is to provide a support system.

[0007]

In order to solve the above problems, the translation support system of the present invention includes a first translation target system.
An original text input means for inputting an original text of a language, a term dictionary in which a set of a first language term and a corresponding translated word of a second language are registered as a dictionary, and an original text input by the original text input means is collated with the term dictionary. However, an identification mark adding means for adding an identification mark for prohibiting translation of the term to a term registered in the term dictionary in the original text, and a translator for the original text to which the identification mark is added by the identification mark adding means It is characterized by further comprising: a translated word replacement unit that obtains a final translation result by matching the translated sentence of the second language translated by the term dictionary with the term dictionary and replacing the untranslated term with the translated word.

Further, the present invention is characterized by further comprising an abbreviation adding means for adding a previously registered abbreviation to the term to which the identification mark adding means adds the identification mark.

[0009]

When the original sentence is input, the original sentence is collated with the term dictionary, and a term such as a technical term registered in the term dictionary is added with an identification mark for prohibiting translation. The original text to which the identification mark is added is divided into several pieces and delivered to a plurality of translators.

The translator translates only the portions of the original text other than the terms to which the identification mark is added, and leaves the terms to which the identification mark is added as the original language. At this time, if an abbreviation is also added to the term with the identification mark in the original sentence, when the translator uses a Japanese word processor to input the translated sentence, the Kana-Kanji conversion function will be applied to the term that remains as the original word. You can easily enter abbreviations using it, and spelling mistakes are less likely to occur.

The translated texts thus obtained by a plurality of translators are edited by the same person (editor) as one translated text, collated with the term dictionary, and untranslated terms are replaced with translated words to make a final translation. The translation result is obtained.
Therefore, of the technical terms appearing in the original text, at least the terms registered in the term dictionary are completely unified in translation.

[0012]

EXAMPLES Examples of the present invention will be described below. Figure 1
It is a block diagram showing a schematic structure of a translation support system by one example of the present invention. This translation support system includes an input unit 1 for inputting an original sentence and various commands, a term dictionary 2 described later, an external term dictionary 3, an edit processing unit 4, a display unit 5 for displaying an original sentence and a translated sentence, an external unit. The communication unit 6 is an interface for exchanging the original sentence and the translated sentence with the translator of the above by communication, and an output unit 7 for outputting the original sentence and the translated sentence as a hard copy or as a medium such as a floppy disk.

The edit processing unit 4 has a function of editing the original sentence and the translated sentence such as division / combination, and includes an identification mark / abbreviation adding unit 8 and a translated word replacement processing unit 9.
The identification mark / abbreviation adding unit 8 performs a process of adding an identification mark and an abbreviation indicating that translation is prohibited to the term registered in the term dictionary 2 in the original sentence to be translated. The translated word replacement processing unit 9 performs processing of converting an untranslated term in a translated sentence translated by a translator into a translated word by referring to the term dictionary 2.

FIG. 2 shows an example of the contents of the term dictionary 2.
As shown in this figure, the term dictionary 2 has a large number of registered pairs of technical terms in the first language (English in this example) and corresponding translated words in the second language (Japanese in this example). , Abbreviations are registered for each technical term. Abbreviations are three-digit numbers in this example. Further, it is desirable that the term dictionary 2 be capable of coping with the ending change of technical terms (singular / plural difference, present tense, past tense, progressive tense, etc.).

Next, the operation of this embodiment will be described with reference to FIGS.
Will be described. FIG. 3 is a flow chart showing a flow of a series of translation processes using the translation support system of FIG. 1, and FIG. 4 is an original sentence (a), an original sentence with an identification mark and an abbreviation (b), and a translation divided by a translator. Translated text (c)
And the final translation result (d) are shown.

First, the original sentence to be translated as shown in FIG. 4 (a) provided by the translation request source is input to the edit processing unit 4 by the input unit 1 (S1). The input original text is sent to the identification mark / abbreviation adding unit 8. The identification mark / abbreviation addition unit 8 performs automatic term selection (S2). That is, in the identification mark / abbreviation adding unit 8, the input original sentence is converted into the term dictionary 2
Then, the technical term registered in the term dictionary 2 is searched from the original sentence, and an identification mark for prohibiting translation by the translator is added to the term. An external term dictionary 3 provided by, for example, a translation request source is transplanted into the term dictionary 2 as needed. The external term dictionary 3 may include a term without a corresponding translation.

Next, the editor manually selects a term (S3). That is, the term dictionary 2 is included in the input original text.
Since there are cases where technical terms that are not registered in the external term dictionary 3 are included, the editor adds a similar identification mark to such unregistered terms.

After this, abbreviations are further added to the technical terms to which the identification marks have been added in S2 and S3 (S4).
The abbreviation to be added is selected by referring to the term dictionary 2 for the term to which the identification mark is automatically added in S2, and is selected by the editor for the technical term in which the identification mark is manually added in S3. To be selected.

As shown in FIG. 4B, the original text to which the identification mark and the abbreviation are added is plural (2 in this example).
One). As shown in FIG. 4B, the technical terms have {} as an identification mark added to the beginning and end of the term, and a three-digit number that is an abbreviation is added after the term.

The original sentence divided into a plurality of pieces by adding the identification mark and the abbreviation in this way is delivered to a plurality of translators through the communication section 6 or the output section 7 and is divided and translated (S5). Each translator does not translate the term to which the identification mark is added, inputs it in English as it is, and translates only the other parts into Japanese, so that Japanese with English as shown in FIG. Get a translation of a word.

Generally, a translator inputs a translated sentence by using a Japanese word processor. For a technical term with an identification mark, an abbreviation added to the term is keyed, and then an abbreviation is input. By operating the keys, it is possible to input easily. In order to enable this abbreviation input, each translator is asked to provide in advance a conversion dictionary that stores the relationship between abbreviations and terms together with the original sentence shown in FIG. Just add it to the dictionary.

Next, the translated texts divided and translated by a plurality of translators are input to the edit processing unit 4 through the communication unit 6 or the input unit 1. The edit processing unit 4 collects a plurality of input translations mixed with English, synthesizes them and edits them into one document (S5), and inputs them to the translation replacement unit 9 to convert untranslated technical terms into translations. By the replacement, the final translation result as shown in FIG. 4D is displayed and output on the display unit 5, for example (S8).

The translated word replacement processing unit 9 refers to the term dictionary 2 for the term registered in the term dictionary 2 among the untranslated terms (the term to which the identification mark is automatically added in S2). The terms that are not registered in the term dictionary 2 (the terms to which the identification mark is manually added in S3) are automatically converted into the translated words using the translated words obtained in S7. Therefore, as for the terms registered in the term dictionary 2, the translated words can be unified 100% regardless of which translator is responsible for the translation.

In S7, a term which has not yet been translated is added to the terms registered in the term dictionary 2 (this corresponds to a term ported from the external term dictionary 3) or an identification mark is manually added in S3. This is a process in which the same person (for example, an editor) associates a translated word with the defined term. The translated word obtained by this processing is appropriately fed back to the term dictionary 2 and newly registered in the term dictionary 2. By performing such a process, the number of words registered in the term dictionary 2 gradually increases, so that the subsequent translation process can be performed more efficiently.

Finally, the editor reviews the final translation result obtained in S8 as necessary, corrects the inappropriate translation, and then makes a hard copy from the output unit 7 or a floppy disk. Store and output. This correction is performed, for example, when it is desired to use different translated words for the same term appearing in the original sentence. The output final translation result is delivered to the translation request source.

The present invention is not limited to the above embodiment, and various modifications can be made. For example, a symbol other than {} may be used as an identification mark, and the display of the target term may be reversed. The method does not matter. Further, although the above embodiment has been described by taking the translation from English to Japanese as an example, conversely, the present invention can be applied to translation from Japanese to English or translation between any other languages. Furthermore, it is possible to incorporate machine translation as part of the divided translation.

[0027]

As described above, according to the present invention, when a series of technical documents is divided and translated by a plurality of translators, translation of the technical terms registered in the term dictionary is prohibited by the translators. By referring to the term dictionary and substituting the translated words when editing after the divided translation, the translated words of the technical terms can be almost completely unified.

Therefore, when the editor does not need to unify the translations of the technical terms after the translation, or the work is remarkably small, and the original sentence is extremely long, and a large number of translators perform divided translations. However, a certain level of reliability and quality can be guaranteed because the translations are unified. On the other hand, from the standpoint of a translator, it is not necessary to look up and translate a difficult technical term having various meanings in a dictionary or the like, so that the burden is greatly reduced. As a result, it is possible to reduce the labor required for the editor to unify the translated words and to improve the efficiency of the entire translation process, and it is possible to obtain a high-quality translation result in a short time.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a translation support system according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of contents of a term dictionary in FIG.

FIG. 3 is a flowchart showing the flow of translation processing using the translation support system of FIG.

FIG. 4 is a diagram showing an example of an original sentence, an original sentence to which an identification mark and an abbreviation are added, a translated sentence translated by a translator, and a final translation result.

[Explanation of symbols]

1 ... Input part 2 ... Term dictionary 3 ... External term dictionary 4 ... Editing processing part 5 ... Display part 6 ... Communication part 7 ... Output part 8 ... Identification mark
Abbreviation addition unit 9 ... Translation replacement unit

Claims (2)

[Claims] 1. A source text inputting means for inputting a source text of a first language to be translated, a term dictionary in which a set of a term of the first language and a corresponding translated word of a second language are registered as a dictionary, and the source text. Identification mark adding means for collating the original text input by the input means with the term dictionary and adding an identification mark for prohibiting translation of the term to a term registered in the term dictionary in the original text, the identification mark A translated word replacement unit that obtains a final translation result by matching a translated sentence of a second language translated by a translator with respect to the original sentence added with the identification mark by the adding unit with the term dictionary and replacing an untranslated term with a translated word. A translation support system comprising: 2. The translation support according to claim 1, further comprising an abbreviation adding means for adding a pre-registered abbreviation to the term to which the identification mark is added by the identification mark adding means. system.