JP3952009B2 - Translation memory system, translation method, and program for translation memory system - Google Patents

Description

  The present invention relates to a translation memory system that performs translation using a translation memory, a translation method, and a program for a translation memory system applied to the translation memory system.

  Translation memory systems that perform translation using a translation memory are known. The translation memory refers to a function of storing a pair of original sentences already translated and their translations in a storage device and extracting a translation of the newly input original sentence from the storage device. A system that performs translation using this function is called a translation memory system.

  As a translation memory system, for example, there is a translation system described in Patent Document 1. The translation system described in Patent Document 1 records an original text and its translation in a text recording unit. Then, it is determined whether or not the new original text given from the input section is the same as the original text recorded in the text recording section, and if it is equal, the translated text paired with the original text (original text recorded in advance) is newly created. Used as the translation result of the original text.

  A translation system that performs translation without using a translation memory is also known. Among such translation systems, there is also a translation system that outputs a translation in a display mode corresponding to an input original (for example, refer to Patent Document 2). The translation system described in Patent Document 2 analyzes an input original sentence and converts it into a target language translation sentence. The conversion result is converted and displayed according to the character attribute of the original text. At this time, for example, by using a character attribute conversion table that defines a rule such as “words displayed in“ underlined font ”in the original text are displayed in“ italicized font ”in the translated text”, character attributes for the translation result are used. Perform the conversion.

  Patent Document 3 describes an original image storage device for storing original font size, character style, and the like. The original image storage device is used together with a machine translation device having a dictionary, a grammar rule, and a knowledge base. When displaying a translation, the machine translation device displays a translation based on information stored in the original image storage device. Do. For example, when the original image storage device stores information that the word “royal” is expressed in Gothic, the translated word “royal” is also expressed in Gothic.

Japanese Patent Laid-Open No. 5-135095 (page 3-5) Japanese Patent Laid-Open No. 2-139666 (page 2, FIG. 2, FIG. 3) Japanese Patent Laid-Open No. 3-9472 (page 2-3, FIG. 1-4)

  As described in Patent Documents 2 and 3, in a translation system that translates without relying on a translation memory (translation system that translates by analyzing the original text or translates using a dictionary, grammar rules, knowledge base, etc.) The translated text can be displayed in a display mode corresponding to the character attribute of the original text.

  However, in the conventional translation memory system in which the original sentence that has already been translated and its translation are stored in the storage device as a set, only the translation stored in the storage device is extracted and output. The translation of the display mode according to the attribute could not be output. Therefore, in order to display the translated text in a display mode corresponding to the character attribute of the original text, the operator can change the text attribute of the translated text while displaying the original text on the display device or printing the reference text. Had to change. Therefore, the work burden on the operator is large, and the translated text may not be displayed in the same display mode as the original text due to an operator's mistake.

  Therefore, an object of the present invention is to provide a translation memory system, a translation method, and a translation memory system program applied to the translation memory system that can obtain a translation displayed in a display mode corresponding to the original sentence. To do.

A translation memory system according to the present invention includes a first language character string storage means for storing a first language character string described in a first language, and a character string obtained by translating the first language character string into a second language. A certain second language character string is stored in a storage area corresponding to the storage area of the first language character string storage means in which the first language character string is stored , and the font type and font size are displayed as the character string display mode. , the presence or absence of underline, and a second language character string storage means for storing character attribute information that defines a part or all of the display color for each individual second language character string, the first language character string of the new input An input means, a search means for searching a first language character string that is the same as a new first language character string from a first language character string stored in the first language character string storage means, and a search means First language sentence in which the first language character string is stored Extraction that extracts from the second language character string storage means the second language character string stored in the storage area corresponding to the storage area of the column storage means and the character attribute information that defines the display mode of the second language character string And an output means for outputting the second language character string and the character attribute information extracted by the extracting means.

Identical to the input means, the first language character string of the new, the character attribute information that defines a display form of the first language character string is input, the output means comprises a first language character string detecting means new When the first language character string cannot be searched, the new first language character string and its character attribute information may be output. According to such a configuration, when the first language character string cannot be searched and, as a result, the second language character string and the character attribute information thereof cannot be extracted, a new language character string is used instead of the second language character string. The first language character string and its character attribute information can be output.

  For example, the first language character string storage means stores the individual first language character strings so that the order in which the individual first language character strings are stored can be specified, and the search means is the same as the new first language character string. When the first language character string is stored in a plurality of storage areas of the first language character string storage means, the first language character string is stored last in each of the first language character strings stored in the plurality of storage areas. Alternatively, the first language character string may be searched.

  In addition, for example, when the display unit displays a character string in a display mode determined by the character attribute information and the display unit displays a plurality of character strings, one character string of the plurality of character strings is designated by the operator. The first language character string storage means stores the character attribute information for each first language character string, and the search means uses the same first language as the new first language character string. When character strings are stored in a plurality of storage areas of the first language character string storage means, each first language character string stored in the plurality of storage areas is converted into a character for each first language character string. Display on the display means in a display manner determined by the attribute information, and the designation receiving means searches for the first language character string designated by the operator from the first language character strings stored in the first language character string storage means. It may be configured. According to such a configuration, each first language character string is displayed on the display means in a display mode determined by the character attribute information for each first language character string, and the search target intended by the operator is displayed to the operator. Designation of the first language character string can be prompted.

Further, for example, the first language character string storage means stores character attribute information for each individual in the first language character string, the input means, the first language character string of the new, of the first language character string Character attribute information defining a display mode is input , the search means is a first language character string that is the same as the new first language character string, and the character attribute information is input to the input means. You may be comprised so that the 1st language character string which is the same may be searched.

The input unit, file describing a first language character string of the new display mode is determined by the character attribute information may be input.

  The input means creates a copy of the input file, reads a new first language string from the input file, and the search means is identical to the new first language string read from the file by the input means. The first language character string storage means for searching the first language character string from the first language character string stored in the first language character string storage means, and the extracting means for storing the searched first language character string. The second language character string stored in the storage area corresponding to the storage area and the character attribute information defining the display mode of the second language character string are extracted from the second language character string storage means, and the output means The second language character string whose display mode is determined by the character attribute information extracted by the extracting means at the position where the new character string is described in the file copied by the input means Replace with Unishi Te, may be configured to describe the second language character string in the copied file. According to such a configuration, the layout when the input file is displayed and the layout when the file in which the second language character string is described can be aligned.

The input means, the file that contains the image data as well as describing the first language character string of the new display mode is determined by the character attribute information may be input. Since the image data is also copied, the image data included in the input file can be held as it is in the file in which the second language character string is described.

  The input means reads a new first language character string from the input file, and the search means converts the first language character string that is the same as the new first language character string read from the file by the input means to the first language character. The first language character string stored in the column storage means is searched, and the extraction means is stored in a storage area corresponding to the storage area of the first language character string storage means in which the searched first language character string is stored. The second language character string and the character attribute information defining the display mode of the second language character string are extracted from the second language character string storage means, and the output means is a file input to the input means. The second language character string whose display mode is defined by the character attribute information extracted by the extracting means may be described in a different file.

Further, the translation method according to the present invention includes a first language character string storage means for storing a first language character string described in a first language, and a character string obtained by translating the first language character string into a second language. Is stored in a storage area corresponding to the storage area of the first language character string storage means in which the first language character string is stored , and the font type, font as the character string display mode Second language character string storage means for storing character attribute information defining size, presence or absence of underline, part or all of display color for each second language character string, and a new first language character string are described A translation method applied to a translation memory system having an input means for inputting a file, wherein the input means creates a copy of the input file and creates a new first language character string from the input file. The search means First language searching from the first language character string for storing a first language character string first language character string identical to the new first language character string storage means, extracting means, retrieved by the retrieval means A second language character string stored in a storage area corresponding to the storage area of the first language character string storage means in which the character string is stored, and character attribute information defining a display mode of the second language character string . Extracted from the bilingual character string storage means , the output means extracts the new first language character string by the extracting means at the position where the new first language character string is described in the file copied by the input means. The second language character string is described in the copied file so as to be replaced with a second language character string whose display mode is determined by the extracted character attribute information .

The translation memory system program according to the present invention includes a first language character string storage means for storing a first language character string described in a first language, and the first language character string is translated into a second language. The second language character string, which is a character string, is stored in a storage area corresponding to the storage area of the first language character string storage means in which the first language character string is stored . Second language character string storage means for storing character attribute information defining the type, font size, presence or absence of underline, part or all of the display color for each second language character string, and a new first language character string A process of reading a new first language character string from the input file, creating a copy of the file input to the input means in a computer having an input means for inputting a file describing Language characters Processing for searching for the same first language character string from the first language character string stored in the first language character string storage means, and first language character string storage means for storing the searched first language character string A process for extracting from the second language character string storage means a second language character string stored in a storage area corresponding to the storage area and character attribute information defining the display mode of the second language character string, and copying The second language character string whose display mode is determined by the character attribute information extracted by the extraction means at the position where the new first language character string is described in the file. In this case, a process of describing the second language character string in the copied file is executed.

  According to the present invention, in addition to the second language character string, the character attribute information of the second language character string can also be output according to the new first language character string. Therefore, the operator does not need to change the attribute of the second language character string to be output, and the operator's workload is reduced.

  The best mode for carrying out the present invention will be described below with reference to the drawings. In the following description, “character attribute information” is information defining a display mode of a character string. The character attribute information defines, for example, the font type, font size, presence / absence of underline, and the like. It should be noted that the font type, font size, presence / absence of underline, etc. mentioned here are examples of items defined by the character attribute information, and other items may be defined. For example, the character attribute information may define the display color of the character string. The character attribute information is preferably defined for many items such as font type, font size, presence / absence of underline, display color, etc. For example, for example, only one type of font is used. It may be determined.

  In the following description, a character string described in a language before translation (first language) is referred to as an original text. A character string obtained by translating an original sentence into a target language (second language) is referred to as a translated sentence.

  FIG. 1 is a block diagram showing an example of the configuration of a translation memory system according to the present invention. The translation memory system includes a data processing device 2 and a storage device 3. Further, the data processing device 2 includes a character string acquisition unit 21, a character string search unit 22, a file output unit 23, and a registration information acquisition unit 24. The storage device 3 includes an original sentence storage unit 31 and a translated sentence storage unit 32.

  The original text storage unit 31 stores character attribute information that defines a character string to be an original text and a display mode of the character string. The translation storage unit 32 stores character attribute information that defines a character string to be translated and a display mode of the character string. In this example, the original sentence storage unit 31 and the translated sentence storage unit 32 store a character string and character attribute information for each sentence. In addition, all the characters which comprise one sentence are not necessarily displayed with the same display mode. For example, some fonts of characters constituting one sentence may be different from fonts of other characters, or some characters may be underlined. In this case, of each character constituting one sentence, which character is represented in which font, which character is underlined, and the like are stored as character attribute information.

  The registration information acquisition unit 24 receives the registration document file 13 created by the system administrator. The registration document file includes an original text file 13a describing the original text and a translated text file 13b describing the translation of the original text. In the original text file 13a, an original text whose display mode is defined by the character attribute information is described. Similarly, the translation file 13b describes a translation whose display mode is determined by the character attribute information. In the combination of the original sentence and the translated sentence that are paired, for example, the display mode of the original sentence and the translated sentence is determined to be the same. Note that the same display mode means that the character attributes such as the font type, font size, and presence / absence of underline are the same. However, the combination of the original text and the translated text may be determined so that the display mode of the original text and the translated text is different.

  2A is an explanatory diagram illustrating an example of a character string described in the original sentence file 13a, and FIG. 2B is an explanatory diagram illustrating an example of a character string described in the translated sentence file 13b. FIG. 2 shows an example in which the original text file 13a is described in Japanese and the translated text file 13b is described in English.

  The first sentence of the original document file is a sentence “It is fine when raining.” In the character attribute information of the character string constituting this sentence, the character “rain” and the character “clear” are Gothic, the character “sometimes” is bold Gothic, italic, and underlined. It is stipulated. The first sentence of the translation file is a sentence “Showers with sunny intervals.” The character attribute information of the character string constituting this sentence specifies that the character “with” is bold Gothic and italic, underlined, and other characters are Gothic. Comparing the first sentence of the original and the translated sentence, "with" corresponding to the original "sometimes" is determined to be the same display form as the original "sometimes", and all other characters are displayed in Gothic It is determined to be. Therefore, in the first sentence, the display form of the original sentence and the translated sentence is determined to be the same.

  The second sentence of the original text file is a sentence “Sunny and Rainy”. In the character attribute information of the character string constituting this sentence, the character “sunny” and the character “rain” are Gothic, the character “after” is bold Gothic, italic, and underlined. It is stipulated. The second sentence of the translation file is a sentence “Rain after sunshine.”. The character attribute information of the character string constituting this sentence defines that the character “after” is bold Gothic and italic, underlined, and other characters are Mincho. Comparing the second sentence of the original and the translated sentence, “after”, which corresponds to “after” of the original, is determined to be the same display form as “after” of the original, and all other characters are Mincho. It is determined to be displayed. Therefore, in the second sentence, the display form of the original sentence and the translated sentence is determined to be partially different.

  The registration information acquisition unit 24 causes the original text storage unit 31 to store a character string included in the original text file 13a and character attribute information that defines a display mode of the character string. Also, the registration information acquisition unit 24 stores the translation of the translation paired with the original text in the storage area in the translation storage section 32 corresponding to the storage area in the original text storage section 31 in which the character string and the character attribute information of the original text are stored. Character strings and character attribute information are stored. As already described, in this example, the original sentence storage unit 31 and the translated sentence storage unit 32 store a character string and character attribute information for each sentence. Therefore, the registration information acquisition unit 24 stores a character string constituting one sentence included in the original sentence and character attribute information of the character string in the storage area in the original sentence storage unit 31. Then, in the storage area in the translation storage unit 32 corresponding to the storage area, the character string constituting one sentence that is the translation of the sentence and the character attribute information of the character string are stored.

  The source text storage unit 31 and the translation source storage unit 32 may be configured such that the storage area in the source text storage unit 31 and the storage area of the translation storage unit 32 corresponding to the storage area are adjacent to each other. FIG. 3 is an explanatory diagram showing a corresponding storage area in such a configuration. As shown in FIG. 3, a character string constituting one sentence included in the original sentence and character attribute information of the character string are stored in a storage area of the original sentence storage unit 31. Then, in the storage area adjacent to the character string, a character string constituting one sentence serving as a translation of the sentence and character attribute information of the character string are stored. In the adjacent storage area, the original character string and character attribute information are also stored. As shown in FIG. 3, the storage area of the original text storage unit 31 and the storage area of the translation text storage unit 32 are alternately arranged, and a pair of original text and a character string of the translation are stored in a pair of adjacent storage areas. It may be a configuration.

  The registered information acquisition unit 24 extracts the translation for the newly input original from the translation storage 32 by storing the paired original and translation together with the character attribute information in the original storage 31 and the translation storage 32. Is possible.

  A new document file 11 is input to the character string acquisition unit 21. The new document file 11 is a file that is input after the registration information acquisition unit 24 stores the original character string and character attribute information and the translated character string and character attribute information in the storage device 3. The new document file 11 describes a new original text whose display mode is determined by the character attribute information. The character string acquisition unit 21 acquires the character attribute information and the new original text by reading the new original text whose display mode is determined by the character attribute information from the new document file 11.

  For each sentence included in the original sentence described in the new document file 11, the character string search part 22 uses the same character string as the character string constituting the sentence in the character string stored in the original sentence storage part 31. Search from. Further, the character string search unit 22 extracts from the translation storage unit 32 a character string that is a translation of the searched character string and character attribute information of the character string.

  The file output unit 23 outputs the character string extracted by the character string search unit 22 and the character attribute information of the character string. Here, outputting the character string whose display mode is determined by the character attribute information is one of the output modes of the character string and the character attribute. The file output unit 23 describes (outputs) the character string whose display mode is determined by the character attribute information extracted by the character string search unit 22 in the translated text file 12. The translated file 12 may be stored in the storage device 3 illustrated in FIG. 1 or may be stored in another storage device (not shown).

  In the following description, it is assumed that the character string acquisition unit 21 creates a copy of the new document file 11. Then, the file output unit 23 converts the new original text into a translated text whose display mode is determined by the character attribute information extracted by the character string search unit 22 at the position where the new original text is described in the copied file. The description will be made on the assumption that the character string of the translated sentence is described in the copied file by replacing. In this case, the translation file 12 is first created as a copy of the new document file 11.

  The character string acquisition unit 21, the file output unit 23, and the registration information acquisition unit 24 are, for example, a CPU that operates according to a program and a drive device that reads / writes data from / to an information storage medium (floppy (registered trademark) disk drive, CD -ROM drive etc.). Moreover, the character string search part 22 is implement | achieved by CPU which operate | moves according to a program, for example. This program is stored in a storage device such as a RAM.

The program includes a first language character string storage unit that stores a first language character string described in a first language, and a second language that is a character string obtained by translating the first language character string into a second language. The character string is stored in a storage area corresponding to the storage area of the first language character string storage unit in which the first language character string is stored , and the font type, font size, and presence / absence of an underline as a character string display mode Second character string storage means for storing character attribute information defining part or all of the display color for each second language character string and a file describing a new first language character string are input. A process of making a copy of the file input to the input means on a computer provided with the input means, and reading a new first language character string from the input file , the same first as the new first language character string Language string as first language The process of searching from the first language character string stored in the character string storage means, and the storage area corresponding to the storage area of the first language character string storage means in which the searched first language character string is stored Processing for extracting a second language character string and character attribute information defining a display mode of the second language character string from the second language character string storage means, and the new first language character string in the copied file The copied first file is replaced with the second language character string whose display mode is determined by the character attribute information extracted by the extracting means at the position where is described. Is a program for a translation memory system for causing a process to describe the second language character string .

Next, the operation will be described.
First, an operation in which the registration information acquisition unit 24 stores the paired original sentence and translated sentence in the storage device 3 based on the registration document file 13 will be described. FIG. 4 is a flowchart showing an example of processing progress of an operation in which the registration information acquisition unit 24 stores the paired original sentence and translated sentence in the storage device 3.

  When the registration document file 13 including the original text file 13a and the translated text file 13b is input, the registration information acquisition unit 24 acquires the character string and the character attribute information of the character string from the original text file 13a (step S11). . In step S11, the registration information acquisition unit 24 acquires, for each sentence, a character string constituting the sentence and character attribute information of the character string. The registration information acquisition unit 24 may determine a sentence break based on the presence of a symbol used for the sentence break.

  For example, in Japanese, punctuation marks are used for sentence breaks. Therefore, if the registration information acquisition unit 24 determines that the first sentence to the punctuation of the original sentence is a single sentence, and the subsequent characters from the next punctuation to the next punctuation are a single sentence. Good. For example, in the case of English, a period is used for sentence breaks. Therefore, if the registration information acquisition unit 24 determines that the first character to the period in the original sentence is one sentence, and thereafter, the character from the next character to the next period is one sentence. Good. Here, the case of Japanese and English has been described as an example, but the same applies to the case of other languages.

  The registration information acquisition part 24 acquires a character string and the character attribute information of the character string from the translation file 13b after step S11 (step S12). Also in step S12, the registration information acquisition part 24 acquires the character string which comprises a sentence, and the character attribute information of the character string for every sentence. As already described, the registration information acquisition unit 24 may determine a sentence break based on the presence of a symbol used for the sentence break.

  Each character string acquired in steps S11 and S12 does not necessarily constitute one sentence. For example, there is a case where a character string indicating a title is described at the beginning of an original sentence or a translated sentence, followed by a line break after the title and the sentence starts. In this case, it is determined that one sentence is from the first character of the title to a symbol such as a period or a period. However, in this case, it is preferable that the title part is regarded as one sentence separately from the following sentence. In the case of English, a period is used not only for sentence breaks but also for notation of words. Therefore, the registration information acquisition unit 24 may determine that the sentence ends at the position of the period used in the middle of the sentence.

  Therefore, after Steps S11 and S12, the registration information acquisition unit 24 displays each character string acquired in Steps S11 and S12 on a display unit (not shown in FIG. 1) such as a display device. The operator is prompted to confirm that the character strings are paired as the original text and the translated text. Then, the operator confirms that the displayed character strings are paired as an original sentence and a translated sentence, for example, via an input device (not shown in FIG. 1) such as a keyboard or a mouse (step 1). S13).

  When the operator who viewed the displayed character string determines that the character strings that should be two sentences have been acquired as one sentence, the registration information acquisition unit 24 breaks the sentence in the displayed character string. Is input from the operator via the input device. When the character string that should be two sentences is acquired as one sentence, for example, a title that should be regarded as one sentence and a sentence that follows the title are acquired as one sentence. Is the case. In the case of this example, the registration information acquisition unit 24 may receive a designation from the operator that the last character of the title becomes a sentence break. The operator may make such designation. The registration information acquisition unit 24 regards the part before the sentence break designated by the operator as the character string acquired in step S11 or step S12. The part after the break of the sentence is acquired again when the process of step S11 or step S12 is performed again.

  If the operator who viewed the displayed character string determines that only a part of one sentence has been acquired, the registration information acquisition unit 24 further acquires a character string in accordance with the operation of the operator. The case where only one half of a sentence is acquired is a case where, for example, up to a period used in the middle of a sentence is acquired as one sentence. In this example, the registration information acquisition unit 24 may receive an instruction to further acquire a character string from the operator through the input device to the next period (a symbol used in a sentence break). The operator may input such an instruction. In response to this instruction, the registration information acquisition unit 24 further acquires a character string and character attribute information of the character string from the original text file 13a or the translated text file 13b. And the registration information acquisition part 24 determines with the character string already acquired by step S11 or step S12, and the acquired character string comprising one sentence.

  As a result of the processing as described above, if the operator inputs confirmation that the displayed character strings are paired as the original text and the translated text, the registration information acquisition unit 24 performs the process of step S14.

  In step S <b> 14, the registration information acquisition unit 24 stores the character string (character string constituting one sentence) acquired from the original text file 13 a and the character attribute information of the character string in the original text storage unit 31. Furthermore, the registration information acquisition unit 24 stores the character string (a character string constituting one sentence) acquired from the translation file 13b and the character attribute of the character string in the storage area of the translation storage unit 32 corresponding to the storage area. Remember information.

  After step S14, the registration information acquisition unit 24 determines whether or not the acquisition of the character strings from the original text file 13a and the translated text file 13b is completed (step S15). If there is a character string that has not yet been acquired in the original file 13a and the translated file 13b, it is determined that the acquisition of the character string has not been completed, and the processes in and after step S11 are repeated. If all the character strings have been acquired from the original text file 13a and the translated text file 13b, it is determined that the acquisition of the character strings has ended, and the processing is ended.

  For example, it is assumed that an original text file 13a and a translated text file 13b describing a character string illustrated in FIG. 2 are input. Then, a character string “slightly raining” is stored in the original text storage unit 31 by the processing of the registration information acquisition unit 24 illustrated in FIG. 4. In addition, character attribute information that defines that the characters “rain” and “sunny” are Gothic, the character “sometimes” is bold Gothic, italic, and underlined is also stored in the original text. 31. Then, the character string “Showers with sunny intervals.” Is stored in the storage area of the translation storage unit 32 corresponding to the storage area. In addition, character attribute information defining that the character “with” is bold Gothic and italic, underlined, and that other characters are Gothic is also stored. The second sentence of the original sentence and the second sentence of the translated sentence are also stored in the same manner.

  If the system administrator intends to obtain a translation with a display form different from the original text when translating the original text with a certain display style, the character attribute information differs between the paired original text and the translated text. Just do it. For example, the second sentence of FIG. 2B (character “Rain” etc. is Mincho) is output as the translation result of the second sentence of FIG. 2A (letter “rain” etc. is Gothic). If the system administrator intends, as illustrated in the second sentence of FIG. 2, the original text file 13a and the translated text file 13b may be created with different character attribute information.

  Next, the operation of outputting the translation file 12 describing the translation of the original sentence described in the new document file 11 will be described. FIG. 5 is a flowchart showing an example of processing progress of this operation.

  When a new document file 11 is input, the character string acquisition unit 21 creates a copy of the file. The copied file is used as the translated text file 12, and the translation memory system describes the translated text in the translated text file 12 by replacing the original text described in the copied file with the translated text.

  In addition, the character string acquisition unit 21 acquires a character string and character attribute information of the character string from the original text described in the new document file 11 (step S21). In step S21, the character string acquisition unit 21 acquires, for each sentence, a character string constituting the sentence and character attribute information of the character string. The character string acquisition unit 21 may determine a sentence break based on the presence of a symbol used for the sentence break.

  However, as already explained, if the determination is made only with the symbols used in sentence breaks, a character string that should be two sentences is acquired as one sentence, or one sentence is partway through as one sentence. It may be acquired. Therefore, the character string acquisition unit 21 displays the acquired character string on a display unit (not shown in FIG. 1) such as a display device, as in the process of step S13. When the operator who viewed the displayed character string determines that the character strings that should be two sentences are acquired as one sentence, the character string acquisition unit 21 breaks the sentence in the displayed character string. The position is input from the operator via an input device (not shown in FIG. 1) such as a keyboard or a mouse. The character string acquisition unit 21 regards a part before the sentence break designated by the operator as the character string acquired in step S21. The part after the break of the sentence is acquired again when the process of step S21 is performed again. In addition, when the operator who has viewed the displayed character string determines that only part of a sentence has been acquired, the character string acquisition unit 21 further follows the character string and the character string in accordance with the operator's instruction. Get character attribute information of. Then, the character string acquisition unit 21 determines that the already acquired character string and the acquired character string constitute one sentence. The operation until one sentence is confirmed according to the operation of the operator is the same as the operation when one sentence is confirmed in step S13.

  As a result of the above processing, if the operator inputs confirmation that the displayed sentence is a single sentence, the process proceeds to step S22 by the character string search unit 22.

  When the character string acquisition unit 21 acquires the character string in step S21, the position where the character string is described (for example, the display position of the character string when the contents of the new document file 11 are displayed). ).

  In step S21, the character string constituting the sentence and the character attribute information of the character string are acquired for each sentence. Therefore, the processes in steps S21 to S25 described below are also performed for each character string of one sentence. Done.

  In step S22, the character string search unit 22 selects the same character string as the character string acquired by the character string acquisition unit 21 (character string constituting one sentence) from the character strings stored in the original text storage unit 31. Search is performed (step S22). If only one character string identical to the character string acquired by the character string acquisition unit 21 is stored in the original text storage unit 31, the character string search unit 22 may search for the character string.

  However, the same character string as the character string acquired by the character string acquisition unit 21 may be stored in a plurality of storage areas of the original text storage unit 31. For example, a character string stored in a storage area of the original text storage unit 31 and a character string stored in another storage area of the original text storage unit 31 are the same, and character attribute information of each character string is It can be different. Thus, when the same character string as the character string acquired by the character string acquisition unit 21 is stored in the plurality of storage areas of the original text storage unit 31, the character string search unit 22 is stored in each storage area. Each character string is displayed on display means (not shown in FIG. 1) such as a display device. At this time, the character string search unit 22 displays each character string in a display mode determined by the character attribute information of each character string. In this way, the same character string having a different display mode is displayed, and the operator is prompted to specify a character string to be searched that is intended by the operator. An input device (not shown in FIG. 1) such as a keyboard and a mouse is designated by the operator as one of the displayed character strings. The character string search unit 22 searches for a character string designated by the operator via the input device.

  The character string search unit 22 determines whether or not the search for the same character string as the character string acquired by the character string acquisition unit 21 is successful (step S23). If the same character string as the character string acquired by the character string acquisition unit 21 can be searched in step S22, the character string search unit 22 sets the character string of the translated sentence paired with the searched character string and the character string of the character string. Character attribute information is extracted from the translated text storage unit 32. At this time, the character string and the character attribute information of the translation may be extracted from the storage area of the translation storage unit 32 corresponding to the storage area of the searched character string. Then, the file output unit 23 outputs the character string and character attribute information (step S24). In step S24, the file output unit 23 determines the display mode of the extracted character string based on the extracted character attribute information, and when displaying, the extracted character string is displayed in the display mode. Write to the translated text file 12. Since the translated text file 12 is a file created as a copy of the new document file 11, the translated text file 12 describes the character string (original text) acquired in step S21. The file output unit 23 replaces the original text with the translated text whose display mode is determined by the character attribute information extracted by the character string search section 22 at the position where the original text is described in the translated text file 12. The translation is described in the translation file 12. It should be noted that the position where the original text to be replaced is described in the translated text file 12 is the same position as the position determined in step S21.

  If the character string search unit 22 cannot search for the same character string as the character string acquired by the character string acquisition unit 21 in step S22, the file output unit 23 reads the character string acquired by the character string acquisition unit 21, and The character attribute information of the character string is output (step S25). The character string acquired by the character string acquisition unit 21 is not stored in the original text storage unit 31 because the same character string as the character string acquired by the character string acquisition unit 21 cannot be searched in step S22. Is the case. In step S25, the file output unit 23 describes the character string acquired in step S21 in the translated text file 12, as in the case where the translated text is described in step S24. That is, the translated text file 12 describes the character string (original text) acquired in step S21. In step S21, the file output unit 23 replaces the original text with the character string whose display mode is determined by the character attribute information acquired in step S21 at the position where the original text is described in the translated text file 12. The acquired character string is described in the translation file 12. In this case, the character string before replacement and the character string after replacement are the same.

  If the process of step S24 or step S25 is complete | finished, the character string acquisition part 21 will determine whether acquisition of the character string from the new document file 11 was complete | finished (step S26). If there is a character string that has not yet been acquired in the new original text file 11, it is determined that the acquisition of the character string has not been completed, the process of step S 21 is executed again, and the character string search unit 22 and the file output unit 23 again executes the processing of step S22 to step S24 (or step S25). If all the character strings have been acquired from the new document file 11, it is determined that the acquisition of the character strings has been completed, and the process ends. As a result, the translated text file 12 in which the translated text is described in a display mode corresponding to the original text is obtained.

  Further, the new document file 11 may include image data (including graphic data drawn in a diagram). The image data included in the new document file 11 is copied to the translated sentence file 12 as it is. Further, the file output unit 23 describes the translated text by replacing the original text with the translated text at the position where the original text is described in the copied file (translated text file 12). Therefore, the graphic data included in the new document file 11 can be held in the translated text file 12 as it is. Further, the layout when the new document file 11 is displayed can be held in the translated text file 12 as it is.

  Below, the specific example of the process of step S21-S26 is shown. However, the character string and the character attribute information are stored in the original sentence storage unit 31 based on the original sentence shown in FIG. 2A, and the character string and the character attribute information are converted into the translated sentence storage part 32 based on the translated sentence shown in FIG. It is assumed that it is stored in FIG. 6A shows an example of the original text described in the new document file 11. When the new document file 11 in which the original text shown in FIG. 6A is described is input to the character string acquisition unit 21, the character string acquisition unit 21 reads the character string “1” of the first sentence shown in FIG. A character string “rainy and sometimes sunny” and its character attribute information are acquired (step S21). This character attribute information defines that the character “rain” and the character “clear” are Gothic, the character “sometimes” is bold Gothic, italic, and underlined. Further, the character string acquisition unit 21 creates a copy of the new document file 11.

  Next, the character string search unit 22 searches the character string stored in the original text storage unit 31 for the same character string as the character string “rainy sometimes fine” acquired in step S21 (step S22). In this example, the original text storage unit 31 stores a character string “rainy and sunny” (see FIG. 2A). Therefore, the character string search unit 22 can search the original text storage unit 31 for a character string “rainy and sunny” (Y in step S23). Therefore, the process proceeds to step S24.

  The character string search unit 22 extracts the character string and the character attribute information of the translated sentence from the storage area of the translated sentence storage unit 32 corresponding to the storage area in which the searched character string “rainy and sunny” is stored. A character string “Showers with sunny intervals.” And character attribute information of the character string are stored in the storage area of the translation storage unit 32 corresponding to the storage area of the character string “rainy and sunny.” The character attribute information defines that the character “with” is bold Gothic and italic, underlined, and that other characters are Gothic (see FIG. 2B). Therefore, the character string search unit 22 extracts the character string “Showers with sunny intervals.” And its character attribute information. Then, the file output unit 23 describes the character string “Showers with sunny intervals.” Whose display mode is determined by the extracted character attribute information in the translated text file 12 (a copy of the new document file 11). At this time, "Showers with sunny intervals." Is described by replacing "Showers with sunny intervals." With "Showers with sunny intervals." As a result, the first sentence “Showers with sunny intervals.” Shown in FIG. 6B is written as the first sentence in the translated sentence file.

  Next, the character string acquisition unit 21 determines whether or not the acquisition of the character string from the new document file 11 is completed (step S26). Since the second sentence “Sunny Day” that has not yet been acquired remains in the new document file 11 (see FIG. 6A), the character string acquisition unit 21 finishes acquiring the character string. It is determined that it has not been performed (N in step S26). Then, the character string acquisition unit 21 acquires a character string “sunny day” of the second sentence shown in FIG. 6A and character attribute information of the character string (step S21). The character attribute information defines that the character “sunny” and the character “day” are Gothic, the character “no” is bold Gothic, italic, and underlined.

  Next, the character string search unit 22 searches the character string stored in the original text storage unit 31 for the same character string as the character string “Sunny Day” acquired in step S21 (step S22). In this example, the original text storage unit 31 does not store a character string “sunny day” (see FIG. 2A). Therefore, the character string search unit 22 cannot search the original text storage unit 31 for the character string “sunny day” (N in step S23). Therefore, the process proceeds to step S25.

  The file output unit 23 displays the character string “Sunny Day” whose display mode is determined by the character attribute information acquired in step S21 (this character string is also acquired in step S21) as the translated sentence file 12. (Copy of new document file 11). At this time, “sunny day” is described in such a manner that “sunny day” is replaced with “sunny day” at the position where “sunny day” is described. In this case, the character string before replacement and the character string after replacement are the same. As a result, the second sentence “sunny day” shown in FIG. 6B is written as the second sentence in the translation file.

  Next, the character string acquisition unit 21 determines whether or not the acquisition of the character string from the new document file 11 is completed (step S26). Since all the character strings have been acquired from the new document file 11 (Y in step S26), the process ends. Through the above processing, the translated text file 12 describing the character string shown in FIG. 6B is output. The second sentence “Sunny Day” cannot be translated because the combination of the original sentence and the translated sentence is not stored in the storage device 3. In this case, the operator may edit the translated text file 12 and change the attribute of the translated character string after translation. The operator does not need to change the attribute of the character string extracted from the translated sentence storage unit 31, and only edits the sentence that has not been translated, such as the second sentence in FIG. 6B. . Therefore, the editing work burden on the operator is reduced.

  A conventional translation memory system stores only a pair of original text and translated text in advance, and does not store character attribute information. Therefore, the display mode when the output translation is displayed is a display mode determined by default, and is not according to the display mode of the input original text. Therefore, even if, for example, the character string shown in FIG. 7A is input to the conventional translation memory system, the output result is a character represented in a predetermined display form as shown in FIG. It becomes a column and does not display according to the input original text.

  According to this invention, the translation memory | storage part 32 memorize | stores the character attribute information of the character string with the character string of a translation. Therefore, the character string of the translated sentence can be written in the translated sentence file 12 so that it is displayed in a display mode corresponding to the original sentence at the time of display. As a result, the operator can obtain the translated text file 12 describing the character string whose display mode is determined according to the original text. The operator does not need to perform an operation to change the attribute of the character string, and only needs to perform an editing operation on a sentence that has not been translated. Therefore, the work burden on the operator can be reduced.

  In addition, when the original text storage unit 31 cannot search the original text storage unit 31 for a character string that is the same as the newly input original text string, the newly input character string is displayed. Is written in the translated text file 12 so that is stored as it is. Therefore, when the translated text cannot be output, the input character string can be output as it is.

  In addition, the character string acquisition unit 21 creates a copy of the new document file 11, and the file output unit 23 describes the translated text in the translated text file 12 that is the copied file. Then, the file output unit 23 describes the translated sentence by replacing the original sentence with the translated sentence at the position where the original sentence is described in the translated sentence file 12. Therefore, the graphic data included in the new document file 11 can be held in the translated text file 12 as it is. Further, the layout when the new document file 11 is displayed can be held in the translated text file 12 as it is.

  The original text storage unit 31 also stores character attribute information of the character string along with the original character string. Therefore, when a plurality of the same character strings are stored in the original text storage unit 31, the plurality of the same character strings are displayed in a display mode corresponding to the character attribute information, respectively, and the operator intends to the operator. The user can be prompted to specify the search target character string.

  In the above description, when the same character string as the character string acquired by the character string acquisition unit 21 is stored in the plurality of storage areas of the source text storage unit 31 in step S22, the operator is requested to specify the search target character string. Explained when to encourage. Instead of prompting the operator to specify a search target character string, the character string search unit 22 may specify one character string from the same character strings stored in a plurality of storage areas. For example, when the same character string is stored in a plurality of storage areas of the original text storage unit 31, the character string search unit 22 may search for the last stored character string. In this case, the character search unit 22 displays a plurality of identical character strings in their respective display modes, and performs a character search without prompting the operator to specify. Therefore, there is no need to display a plurality of identical character strings in their respective display modes. In this case, the character attribute information need not be stored in the original text storage unit 31. However, the original text storage unit 31 stores individual character strings so that the order in which the individual character strings are stored can be specified. For example, when a storage area for storing character strings is set in advance and the registration information acquisition unit 24 stores the character strings in the original text storage unit 31, the character strings are stored in a storage area corresponding to the order of the character strings. Just do it. Therefore, when the same character string is stored in a plurality of storage areas of the original text storage unit 31, the character string search unit 22 is stored in a storage area in which the last order is determined among the plurality of storage areas. May be searched. Further, as a configuration in which the original text storage unit 31 stores individual character strings so that the order in which the individual character strings are stored can be specified, the character string and the time when the character string is stored are stored. Also good.

  In the search process of step S22, the search unit 22 is the same character string as the character string acquired by the character string acquisition unit 21, and the character attribute information is the character attribute information acquired by the character string acquisition unit 21. You may extract the character string which is the same. That is, the search unit 22 determines not only the identity of the character string but also the identity of the character attribute information, and when the identity is confirmed in both the character string and the character attribute information, it is stored in the original text storage unit 31. That character string may be specified as a search target.

  In the above description, the file output unit 23 describes a case where a translation is described for a copy of the new document file 11. Instead of copying the new document file 11, the translation may be described in a new file different from the new document file 11. When the search in step S22 is successful, the file output unit 23 newly creates a translated sentence (character string extracted by the search unit 22) whose display mode is determined by the character attribute information extracted by the character string search unit 22. It should be described in a simple file. If the search in step S22 is not successful, the original text whose display mode is determined by the character attribute information acquired in step S21 (the original text acquired in step S21) may be described in a new file.

  In the above description, the translation is output as a file. However, it may be displayed on a display unit (not shown in FIG. 1) such as a display device. That is, the translated text may be displayed and output. In this case, the file output unit 23 may display the translated text on a display device or the like in a display mode determined by the character attribute information. In addition, the translated text may be displayed together with the original text described in the new document file 11. Further, an editing operation for the displayed character string may be received from the operator via an input device (not shown in FIG. 1).

  In the above description, a case has been described in which a character string is stored for each sentence in the original sentence storage unit 31 and the translated sentence storage unit 32, and the processes of steps S21 to S25 are performed for each sentence. Processing may be performed in units of phrases or words instead of one sentence. In that case, a character string may be stored for each phrase (or word) in the original sentence storage unit 31 and the translated sentence storage unit 32, and the processes of steps S21 to S25 may be performed for each phrase (or word). Also, in the original text file 13a, the translated text file 13b, and the new document file 11, a symbol indicating a break of a phrase (or word) is described, and the registration information acquisition unit 24 and the character string acquisition unit 21 What is necessary is just to determine the break of a phrase (or break of a word).

  In the above description, the original text storage unit 31 corresponds to first language character string storage means. The translation storage unit 32 corresponds to a second language character string storage unit. The character string acquisition unit 21 corresponds to an input unit. The character string search unit 22 corresponds to search means and extraction means. The file output unit 23 corresponds to an output unit. The display device (not shown in FIG. 1) corresponds to a display unit. An input device (not shown in FIG. 1) such as a keyboard and a mouse corresponds to a designation receiving unit.

  The present invention is applicable to a translation memory system that performs translation using a translation memory. Further, the present invention can be applied to a parallel translation editor system or the like in which the original sentence and the translated sentence are displayed side by side and the displayed original sentence and the translated sentence can be edited.

It is a block diagram which shows the example of a structure of the translation memory system by this invention. It is explanatory drawing which shows the example of the character string described in the original text file and the translation file. It is explanatory drawing which shows the example of a corresponding storage area. It is a flowchart which shows the example of the process progress of the operation | movement which memorize | stores an original sentence and a translation into a memory | storage device. It is a flowchart which shows the example of the process progress of the operation | movement which outputs the translated text file which described the translated text. It is explanatory drawing which shows the example of the character string described in the new document file and the translation file to be output. It is explanatory drawing which shows the example of the display mode of the translation output in the conventional translation memory system.

Explanation of symbols

2 Data Processing Device 3 Storage Device 11 New Document File 12 Translation File 13 Registration Document File 13a Original File 13b Translation File 21 Character String Acquisition Unit 22 Character String Search Unit 23 File Output Unit 24 Registration Information Acquisition Unit 31 Original Text Storage Unit 32 Translation storage

Claims (11)

First language character string storage means for storing a first language character string described in a first language;
A storage area corresponding to the storage area of the first language character string storage means in which the second language character string, which is a character string translated from the first language character string into the second language, is stored. A second language that stores character attribute information that defines a font type, font size, presence / absence of an underline, and part or all of a display color as a character string display mode for each second language character string A character string storage means;
An input means for inputting a new first language character string;
Search means for searching a first language character string stored in the first language character string storage means for the same first language character string as the new first language character string;
A second language character string stored in a storage area corresponding to a storage area of the first language character string storage means in which the first language character string searched by the search means is stored, and a display mode of the second language character string Extracting from the second language character string storage means the character attribute information that defines
A translation memory system comprising: output means for outputting a second language character string and character attribute information extracted by the extraction means. The input means comprises a first language character string of the new, and the character attribute information that defines a display form of the first language character string is input,
The output means outputs the new first language character string and the character attribute information when the detecting means cannot search for the same first language character string as the new first language character string. The translation memory system described in 1. The first language character string storage means stores the individual first language character strings so that the order in which the individual first language character strings are stored can be specified,
When the first language character string identical to the new first language character string is stored in a plurality of storage areas of the first language character string storage means, the search means stores the first language character string in the plurality of storage areas. The translation memory system according to claim 1 or 2, wherein the first language character string stored last is searched among the first language character strings. Display means for displaying a character string in a display mode determined by character attribute information;
When the display means displays a plurality of character strings, it comprises a designation receiving means for designating one of the plurality of character strings by an operator,
The first language character string storage means stores character attribute information for each first language character string,
When the first language character string identical to the new first language character string is stored in a plurality of storage areas of the first language character string storage means, the search means stores the first language character string in the plurality of storage areas. Each first language character string is displayed on the display means in a display manner determined by the character attribute information for each first language character string, and the first language character string designated by the operator by the designation receiving means is displayed in the first language. The translation memory system according to claim 1 or 2, wherein a search is performed from the first language character string stored in the character string storage means. The first language character string storage means stores character attribute information for each first language character string,
The input means comprises a first language character string of the new, and the character attribute information that defines a display form of the first language character string is input,
The search means searches for a first language character string that is the same first language character string as the new first language character string and whose character attribute information is the same as the character attribute information input to the input means. The translation memory system according to claim 1 or 2. The input means, the translation memory system according to any one of claims 1 to 5 in which file describing first language character string of the new display mode is determined by the character attribute information is input . The input means creates a copy of the input file, reads a new first language string from the input file,
The search means searches the first language character string stored in the first language character string storage means for the same first language character string as the new first language character string read from the file by the input means,
The extraction means includes a second language character string stored in a storage area corresponding to the storage area of the first language character string storage means in which the searched first language character string is stored, and display of the second language character string Extracting the character attribute information defining the aspect from the second language character string storage means;
The output means has a display mode defined by the character attribute information extracted by the extraction means at the position where the new character string is described in the file copied by the input means. The translation memory system according to claim 6, wherein the second language character string is described in the copied file so as to be replaced with a second language character string. 8. The translation memory system according to claim 7, wherein a file containing image data and a new first language character string whose display mode is defined by character attribute information is input to the input means. The input means reads a new first language character string from the input file,
The search means searches the first language character string stored in the first language character string storage means for the same first language character string as the new first language character string read from the file by the input means,
The extraction means includes a second language character string stored in a storage area corresponding to the storage area of the first language character string storage means in which the searched first language character string is stored, and display of the second language character string Extracting the character attribute information defining the aspect from the second language character string storage means;
The translation memory system according to claim 6, wherein the output means describes the second language character string whose display mode is defined by the character attribute information extracted by the extraction means in a file different from the file input to the input means. . A first language character string storage means for storing a first language character string described in a first language; a second language character string that is a character string obtained by translating the first language character string into a second language; The first language character string is stored in a storage area corresponding to the storage area of the first language character string storage means , and the character string display mode includes font type, font size, presence of underline, display color A second language character string storage means for storing part or all of character attribute information for each second language character string; and an input means for inputting a file describing a new first language character string. A translation method applied to the translation memory system provided,
The input means creates a copy of the input file, reads a new first language string from the input file,
Search means searches the first language character string stored in the first language character string storage means for a first language character string that is the same as the new first language character string,
A second language character string stored in a storage area corresponding to the storage area of the first language character string storage means in which the extraction means stores the first language character string searched by the search means; and the second language character Character attribute information defining the display mode of the column is extracted from the second language character string storage means ;
The output means displays the new first language character string at the position where the new first language character string is described in the file copied by the input means using the character attribute information extracted by the extraction means. A translation method , wherein the second language character string is described in the copied file so as to be replaced with a predetermined second language character string . A first language character string storage means for storing a first language character string described in a first language; a second language character string that is a character string obtained by translating the first language character string into a second language; The first language character string is stored in a storage area corresponding to the storage area of the first language character string storage means, and the character string display mode includes font type, font size, presence of underline, display color A second language character string storage means for storing part or all of character attribute information for each second language character string; and an input means for inputting a file describing a new first language character string. Computer
A process of making a copy of the file input to the input means and reading a new first language character string from the input file;
A process of searching a first language character string stored in a first language character string storage means for a first language character string identical to the new first language character string;
The second language character string stored in the storage area corresponding to the storage area of the first language character string storage means in which the searched first language character string is stored, and the display mode of the second language character string are defined Processing for extracting character attribute information from the second language character string storage means; and
A second language character whose display mode is determined by the character attribute information extracted by the extracting means at the position where the new first language character string is described in the copied file. A program for a translation memory system for executing a process of describing the second language character string in the copied file so as to be replaced with a string .

Images