JP2007087056A - Translation device and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a translation device capable of reasonably matching a translated text to the layout of an original text. <P>SOLUTION: Character data is extracted from image data of a document of the original text read by a reading part. The character data is translated to a designated language. The translated text is temporarily arranged on the layout of the original text to examine whether any arrangement error is present or not. When the arrangement error is present, character attribute change processing for changing the character attribute of the translated text is performed, and the result is temporarily arranged on the layout of the original text to examine the presence/absence of the arrangement error. When the arrangement error is still present, re-translation processing consisting of either or the both of translation method change processing for changing the method of translation or translated document change processing for changing the translated document is performed based on the translation result to change the translation content, whereby the translated text is matched to the layout of the original text. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a technique for translating an original sentence expressed in one language into a sentence in another language.

There is known a translation device that translates an original text expressed in one language into a text in another language. When the original sentence is translated, the number of characters and the character type of the original sentence and the translated sentence are different, so that there is a discrepancy in the length of the character string and the word before and after translation. For this reason, a translation device has been proposed that changes the character attributes such as the character type, character size, character pitch, or line pitch of the translated sentence so that the translated sentence fits in the same layout as the original sentence (for example, Patent Documents 1 to 6). ).
Japanese Patent Laid-Open No. 5-6396 JP-A-5-303589 JP-A-7-85215 JP-A-8-6948 Japanese Patent Laid-Open No. 9-114835 JP 2003-122751 A

  However, if the character attribute itself of such a translated sentence is changed to fit the original layout, the character becomes too fine and it may be difficult to read the translated sentence. Therefore, for example, there is a limit to reducing the size of characters, and characters of a certain size must be used. As a result, the translated sentence cannot be stored without difficulty in the layout of the original sentence, and a problem such as the translated sentence protruding from a frame line or the like in which the original sentence is stored occurs. In particular, when translating a manuscript with limited character areas, such as presentation materials containing tables and figures, the translated text protrudes from the tables and figures, resulting in a translation manuscript that is difficult to read and has a poor appearance. End up.

  Accordingly, an object of the present invention has been made in view of the above circumstances, and is to provide a technique capable of containing a translation that is easy for a viewer to read in a layout similar to the original.

  In order to achieve the above object, the present invention provides a storage means for storing a translation dictionary in which words having the same meaning are associated with each other in different languages, and a language database including information on expression contents that can be changed in each language. A layout analysis means for analyzing the layout of the original text expressed in the first language, and a translation expressed in the second language by translating the original text expressed in the first language using the translation dictionary A translation means for obtaining a sentence; a temporary placement means for fitting the translation sentence obtained by the translation means to the layout obtained by the layout analysis means; and a translation sentence fitted to the layout by the temporary placement means. Judgment means for judging whether or not the layout fits, and the judgment means judges that the translated sentence does not fit in the layout Expediently, by referring to the information about the expression contents contained in the language database, to provide a translation apparatus and a re-translation means for changing the representation of the translation to fit the layout.

  According to this translation apparatus, if a translated sentence fitted to the layout of the original sentence does not fit in the layout, retranslation processing for changing the translation content is performed to match the translated sentence with the layout of the original sentence. Therefore, for example, when a source document with a layout where the text area is restricted to a certain area, such as presentation materials including tables and figures, is translated, the translated text is stored in the layout, and a good-looking translation manuscript is prepared. Obtainable.

  In the present invention, the retranslation means may change the expression of the translated sentence by changing a method of expressing the translated sentence. For example, the retranslation means includes a process of changing the expression of the translated sentence to a shorter phrase expression, a process of omitting the modifier in the translated sentence, a process of deleting the subject in the translated sentence, and a sentence ending in the translated sentence. It is desirable to change the expression of the translated sentence by executing at least one of the processes for simplifying the above.

  In the present invention, the retranslation means may change the expression of the translation sentence by replacing the phrase of the translation sentence obtained by the translation means with another phrase. For example, the retranslation means performs a process of changing a phrase included in the translated sentence to an abbreviation, or a process of changing a phrase included in the translated sentence into a symbol and separately writing an annotation of the symbol in the translated sentence. Therefore, it is desirable to change the expression of the translated sentence.

  Further, the present invention uses the layout analysis function for analyzing the layout of the original text expressed in the first language and the translation dictionary in which words having the same meaning in different languages are associated with each other in the computer. A translation function for translating an original sentence expressed in a language to obtain a translated sentence expressed in a second language, and a provisional function for fitting the translation sentence obtained by the translation function to the layout obtained by the layout analysis means An arrangement function; a determination function for determining whether or not the translated sentence fitted to the layout by the temporary arrangement function is included in the layout; and the determination function determines that the translated sentence is not included in the layout If so, the language database containing information on the contents of expressions that can be changed in each language is referred to, and fits in the layout. Providing a program for implementing the re-translation function to change the sea urchin the translated sentence representation.

Hereinafter, embodiments of a translation apparatus according to the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing the configuration of the translation apparatus 11 according to this embodiment. The translation device 11 is, for example, a multifunction machine, and includes a control unit 1, an operation unit 2, a reading unit 3, a display unit 4, a printing unit 5, and a storage unit 6, as shown in FIG. The operation unit 2 includes various keys and can be operated by an operator. The reading unit 3 scans an image of a document placed on a reading surface (not shown) and outputs it as image data. The display unit 4 includes, for example, a liquid crystal display. The display unit 4 displays image data output from the reading unit 3, image data read by the control unit 1 from the recording medium, the storage unit 6, and the like, and the operation unit 2. The image is displayed in accordance with data input by the constituent keys. The printing unit 5 includes a photoconductor, an exposure unit, a development unit, a transfer unit, a fixing unit, and the like (not shown), and forms a toner image based on image data supplied from the control unit 1 and is a recording material. The toner image is fixed on the paper.

  The storage unit 6 includes a translation dictionary in which words having the same meaning in different languages are associated with each other, a language database including information on expression contents that can be changed in each language, and a translation program executed by the control unit 1. Stored. The contents of the language database will be described in detail later. The control unit 1 displays an image on the display unit 4 based on image data read by the reading unit 3 or image data read from the storage unit 6 based on an instruction from the operator using the operation unit 2. . In addition, the control unit 1 executes a translation program stored in the storage unit 6 to convert a document expressed in a certain language (first language) into a language (second language) designated by the operator. ) Translated into a document.

Next, the translation process by the control unit 1 will be described with reference to the flowchart shown in FIG. Here, a case where the Japanese original text shown in FIG. 3 is translated into English and a case where the English original text shown in FIG. 4 is translated into Japanese will be described as examples.
First, the operator places the original text on the reading surface of the reading unit 3 and operates the operation unit 2 to start the reading operation. In response to this operation, the control unit 1 causes the reading unit 3 to read the original text placed on the reading surface, and obtains image data representing the contents as an image (step S01). Thereby, the control unit 1 obtains image data representing, for example, a Japanese original text shown in FIG. 3 or an English original text shown in FIG.

  Next, the control unit 1 performs layout analysis on the image represented by the image data (step S02). Further, the control unit 1 performs a process of separating character information and design information from the image represented by the image data (step S03). As a result, a pattern area (symbol L in FIGS. 3 and 4) and a character area (symbol W in FIGS. 3 and 4) are separated from the image data. Next, the control unit 1 performs OCR (Optical Character Reader) processing on the character region W separated from the image data, and performs character information recognition processing for converting the image data into character data (step S04). And the control part 1 performs the translation process using a translation dictionary with respect to the character data obtained by this character information recognition process, and converts the character data of an original sentence into the character data of the translated sentence of the designated language. Conversion is performed (step S05). Thus, for example, the original Japanese character data shown in FIG. 3 is translated into English as the translation language, and the original English character data shown in FIG. 4 is translated into Japanese as the translation language. Become.

  Next, the control unit 1 performs provisional placement processing for fitting the character represented by the character data of the translated sentence to the layout determined from the image data of the original sentence (step S06), and the translated sentence fitted to the layout is It is determined whether or not it fits in the layout (step S07). That is, the control unit 1 may refer to the result of provisional arrangement of the translation sentence in the original sentence layout and determine whether or not there is an arrangement error such that each character fitted to the layout protrudes from the frame line or the like. As a result, when it is determined that the translated text does not protrude from the frame or the like and the translated text is included in the layout (step S07; Yes), the control unit 1 performs an arrangement process for determining the temporary arrangement ( In step S08, the translated text is printed by the printing unit 5 (step S09).

  Here, FIG. 5 shows a state in which an arrangement error has occurred when character data translated from Japanese to English is provisionally arranged in the original text layout. In the example shown in FIG. 5, the number of lines increases because the phrase of the translated sentence is longer than before translation, and the translated sentence protrudes from the frame line that is the pattern area L. FIG. 6 shows the result of provisional arrangement of character data translated from English into Japanese in the original text layout. Also in this case, the number of lines increases because the phrase of the translated sentence is longer than before translation, and the translated sentence protrudes from the frame line that is the pattern area L.

  As illustrated in FIGS. 5 and 6, when it is determined that the translated sentence does not fit in the original sentence layout (step S <b> 07; No), the control unit 1 performs a character attribute change process (step S <b> 10). In this character attribute changing process, the character attributes such as the character size, the character spacing, the line spacing, or the font represented by the character data of the translated text are changed so that the translated text can be accommodated in the original text layout as much as possible. Note that the change of the character attribute is performed within a certain range so that the character is not too small or the character interval and line interval are too narrow so that it is difficult to decipher.

  Next, the control unit 1 performs a temporary arrangement process in which the character data of the translated sentence whose character attribute has been changed is again applied to the layout of the original sentence (step S11). The control unit 1 determines whether or not the result of the temporary placement process is acceptable (step S12). That is, the control unit 1 refers to the result of provisional arrangement of the translation sentence in the original sentence layout, and determines whether or not there is an arrangement error such that a word or phrase fitted to the layout protrudes from a frame or the like. As a result, when it is determined that the translated text is within the layout of the original text (step S12; Yes), the control unit 1 performs an arrangement process for finalizing the temporary arrangement (step S08), and the printing unit 5 performs the translation. The sentence is printed (step S09).

  On the other hand, if the control unit 11 determines that there is still an arrangement error and the translated sentence does not fit in the layout of the original sentence (Step S12; No), the re-translation processing is performed on the translated sentence (Step S12). S13). This retranslation process is a process of shortening a phrase in a translated sentence by using the following method, and is roughly divided into a change in a translation sentence expression method and a replacement of a phrase after translation.

(1) Change of translation method First, the change of the translation sentence expression method will be specifically described.
(1-1) Process for changing to simple or plain expression The control unit 1 refers to a language database stored in the storage unit 6 for a partial expression of the translated sentence, and performs a simple expression or a simple expression ( Change to a shorter expression. For example, if the translated sentence is an expression “please write down”, it is changed to a simpler expression “please write”. With regard to what expression can be changed to what expression, for example, expressions before and after the change may be associated in advance and included in the language database.

(1-2) Change processing to synonym The control unit 1 refers to the language database stored in the storage unit 6 for a partial expression of the translated sentence, and changes the expression to a shorter expression using a synonym. For example, when the translated sentence is an expression “make a decision”, the expression is changed to “decide”. With regard to what expression can be changed to what expression, for example, expressions before and after the change may be associated in advance and included in the language database.

(1-3) Qualifier Omission Processing The control unit 1 omits this redundant modifier when a modifier such as an adjective appears repeatedly in the translated sentence. For example, if the expression “technical support person in charge” appears repeatedly in the translation, “technical support” is omitted from this expression and changed to “person in charge”. The algorithm of what kind of expression is omitted in what case is included in the contents of the language database in advance.

(1-4) Subject deletion process The control unit 1 deletes unnecessary subjects in the translated sentence. For example, when the translated sentence is Japanese, if the subject is clearly indicated at the beginning of the sentence, it is not always necessary to indicate the subject in the subsequent sentences. Algorithms such as when the subject can be omitted are included in the contents of the language database in advance.

(1-5) Ending Change Process The control unit 1 changes the end of the translated sentence to a simple one. For example, when the ending of the translated sentence is “Study”, the ending is changed to “Study.” Such as verbal stop. Information such as what kind of ending is changed to what expression is included in the contents of the language database in advance.

(2) Substitution of words after translation Next, the substitution of words after translation will be described.
(2-1) Process for changing to general abbreviation The control unit 1 changes a part of the phrase of the translated sentence to a general abbreviation. For example, when the phrase of the translated sentence is “Department”, the phrase is changed to “Dept.”. In addition, when the phrase of the translated sentence is “First fiscal half” or “Second fiscal half”, these words are changed to “1H” or “2H”. This is also the case when “AND” is changed to “&”. Information such as what common abbreviations are in what words is included in the language database in advance.

(2-2) Change processing to special abbreviation According to this method, a phrase that does not have a general abbreviation as described above can be replaced with a special abbreviation that is used only within a certain organization. The control unit 1 changes a part of the phrase of the translated sentence to a special abbreviation defined in advance in the language database. For example, when the phrase of the translated sentence is “Information Management” or “Data Storage”, these phrases are changed to “IM” or “DS”. Information such as what special abbreviations exist for what words may be included in the language database in advance.

(2-3) Separate processing by annotation The control unit 1 changes a part of the phrase of the translated sentence into a symbol, and additionally writes the annotation of this symbol in a place where no trouble occurs in the document layout. For example, in the case of “Information Management” or “Data Storage”, these phrases are changed to the symbols “IM” and “DS”. For example, “IM: Information Management” or “DS: Annotation “Data Storage” is described. As for what words are changed to symbols, an algorithm based on the appearance frequency of words in the translated sentence, the length of the words, etc. is prepared in advance and included in the language database. .

  Here, FIG. 7 shows the contents of the retranslation processing for the English translation. The English translation is subjected to a process for changing to a general abbreviation and a process for changing to a special abbreviation in the translation document change process described above. Specifically, as shown in FIG. 7, “And” is changed to “&”, “Data Storage” is changed to “DS”, “Information Management” is changed to “IM”, and “First fiscal” is changed. “half” is changed to “1H”, and “Second fiscal half” is changed to “2H”.

  FIG. 8 shows the contents of retranslation processing for a Japanese translation. For this Japanese translation, subject deletion processing and synonym changing processing in the translation method change processing are performed. Specifically, as shown in FIG. 8, the subject “You are” has been deleted, and “Clarify” has been changed to a shorter synonym “Confirm”.

  Returning to FIG. 2 again, the control unit 1 performs a provisional arrangement process in which the character data of the translated sentence subjected to the retranslation process is applied to the layout of the original sentence (step S14). And the control part 1 judges whether the result of a temporary arrangement | positioning process may be sufficient (step S15). When the control unit 1 determines that the layout is not good (step S15; No), the control unit 1 returns to the retranslation processing in step S13 again and repeats the above-described processing until there is no arrangement error (steps S13 to S15).

  Here, FIG. 9 translates the original Japanese text in FIG. 3 into English, and further translates the original text layout by performing the retranslation processing in step S13 on the result of the character attribute change processing in step S10. It is a figure which shows the state which could contain the sentence. As shown in the figure, the translated sentence (character area W) is contained in the frame line which is the picture area L, as in the original sentence shown in FIG. 10 translates the original English text in FIG. 4 into Japanese, and further performs the retranslation processing in step S13 on the result of the character attribute change processing in step S10, thereby translating the translated text into the original text layout. It is a figure which shows the state which was able to be accommodated. As shown in the figure, the translated text (character area W) is contained within the frame line that is the picture area L, as in the original text shown in FIG. As shown in FIGS. 9 and 10, when it is determined that the translated sentence fitted to the layout of the original sentence is included in the layout (step S15; Yes), the control unit 1 performs an arrangement process for finalizing the provisional arrangement. (Step S08), and the translated text is printed by the printing unit 5 (step S09).

  As described above, according to the translation apparatus according to the present embodiment, the translated sentence is further matched with the layout of the original sentence by further performing retranslation processing on the translation result. Thus, for example, when a source document with a layout where the text area is restricted to a certain area, such as presentation materials including tables and figures, is translated, the translated text is stored in that layout, and a good-looking translation manuscript Can be obtained.

  Note that the translation device 11 is not limited to a multi-function device including a reading unit and a printing unit, and may be realized by a computer connected to such a multi-function device as a host device. In this case, the translation apparatus 11 does not include the reading unit 3 and the printing unit 5 illustrated in FIG. Further, the original text to be translated is not limited to the document read by the reading unit 3, but may be text data, for example. In this case, the processing in steps S01 and S04 in FIG. 2 is not necessary.

It is a block diagram which shows the structure of the translation apparatus concerning this embodiment. It is a flowchart explaining the translation process by a translation apparatus. It is a figure which shows the original document which consists of Japanese. It is a figure which shows the original document which consists of English. It is a figure which shows the layout of the state which produced the arrangement | positioning error at the time of the temporary arrangement | positioning of the character data translated from Japanese into English. It is a figure which shows the layout of the state which produced the arrangement | positioning error at the time of the temporary arrangement | positioning of the character data translated from English into Japanese. It is a figure which shows the retranslation process with respect to an English translation. It is a figure which shows the retranslation process with respect to a Japanese translation. It is a figure which shows the layout after the arrangement | positioning error elimination at the time of the temporary arrangement | positioning of the character data translated from Japanese to English. It is a figure which shows the layout after the arrangement | positioning error elimination at the time of temporary arrangement | positioning of the character data translated from English to Japanese.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Control part, 2 ... Operation part, 3 ... Reading part, 4 ... Display part, 5 ... Printing part, 6 ... Memory | storage part, 11 ... Translation apparatus, L ... Picture area, W ... Character area.

Claims (6)

Storage means for storing a translation dictionary in which words having the same meaning in different languages are associated with each other, and a language database including information on expression contents that can be changed in each language;
Layout analysis means for analyzing the layout of the original text expressed in the first language;
Translation means for translating an original sentence expressed in the first language using the translation dictionary and obtaining a translated sentence expressed in a second language;
Temporary placement means for fitting the translation obtained by the translation means to the layout obtained by the layout analysis means;
A judging means for judging whether or not the translation sentence applied to the layout by the temporary placement means fits in the layout;
When the determination means determines that the translated sentence does not fit in the layout, the expression of the translated sentence is included so as to fit in the layout with reference to information on the contents of the expression included in the language database. And a retranslation means for changing.   The translation device according to claim 1, wherein the retranslation unit changes the expression of the translated sentence by changing a method of expressing the translated sentence.   The re-translation means includes a process for changing the expression of the translated sentence to a shorter phrase expression, a process for omitting the modifier in the translated sentence, a process for deleting the subject in the translated sentence, and a concise ending of the sentence in the translated sentence. The translation apparatus according to claim 2, wherein the expression of the translated sentence is changed by executing at least one of the processes to be performed.   The translation device according to claim 1, wherein the retranslation means changes the expression of the translation sentence by replacing a phrase of the translation sentence obtained by the translation means with another phrase.   The retranslation means performs a process of changing a phrase included in the translation sentence to an abbreviation, or a process of changing the phrase included in the translation sentence into a symbol and separately writing the annotation of the symbol in the translation sentence, The translation apparatus according to claim 4, wherein expression of the translated sentence is changed. On the computer,
A layout analysis function for analyzing the layout of the original text expressed in the first language;
A translation function that translates the original sentence expressed in the first language using a translation dictionary that associates words of the same meaning in different languages, and obtains a translated sentence expressed in the second language;
A provisional arrangement function for fitting the translation obtained by the translation function to the layout obtained by the layout analysis means;
A determination function for determining whether or not the translated sentence fitted to the layout by the temporary placement function is included in the layout;
When it is determined by the determination function that the translated sentence does not fit in the layout, a language database including information on expression contents that can be changed in each language is referred to so as to fit in the layout. A re-translation function for changing the expression of the translated sentence.

Images