JP4248549B2 - Machine translation editing apparatus, machine translation editing program, and machine translation editing method - Google Patents

Description

  The present invention relates to a machine translation editing apparatus that generates a translation by translating a text original text in a first language including a markup symbol into a second language, and in particular, a machine translation apparatus and a machine translation editing program suitable for editing a translation. And a machine translation editing method.

  In general, a machine translation apparatus has a translation module. The translation module analyzes the input source language document (text) using the basic dictionary, the user dictionary, and the grammar rules to generate a structural analysis tree. The translation module also uses the tree structure conversion rules to convert the structure parse tree of the source language text to the structure parse tree of the target language, assigns an appropriate translation to each word, and supplements the necessary parts. To generate text in the target language.

  On the other hand, recently, source language documents described in a markup language are increasingly subject to translation. In this type of source language document, by including a markup symbol in the text of the document, it is possible to display it in a format according to the markup symbol. Well-known markup languages include HTML (HyperText Markup Language), SGML (Standard Generalized Markup Language), and XML (Extensible Markup Language), which have spread rapidly with the spread of the WWW.

  For example, a markup symbol included in an HTML document provides a layout, a specific character display method, a link function, and the like. According to this link function, a user can call a linked document by clicking a portion surrounded by markup symbols. When the markup language included in the source language is translated, it is required to reflect the markup symbol included in the markup language at an appropriate position in the translated sentence.

  A machine translation technique for reflecting a markup symbol in a translation is described in, for example, Patent Document 1. According to the machine translation technique described in Patent Document 1 (hereinafter referred to as the prior art), machine translation is executed in the following procedure.

  First, a start tag of a markup symbol is detected from the original text. Then, the position of the word to which the markup symbol is attached and the number of words surrounded by the start tag and end tag of the markup symbol are stored in the buffer in association with the markup symbol. As a result, the text body is automatically separated from the markup symbol, and a structure analysis tree of the first language is created based on the separated text body. The structural analysis tree of the first language is converted into the structural analysis tree of the second language. Then, the text of the second language is generated from the structural analysis tree of the second language.

Here, each markup symbol included in the original text is buffered in association with the position of the word with the markup symbol and the number of words surrounded by the start tag and end tag of the markup symbol. Is remembered. Therefore, in the above prior art, when the text in the second language is generated, the markup symbol corresponding to the translation word corresponding to the word surrounded by the markup symbol start tag and end tag in the original text is generated. The start and end tags of markup symbols equivalent to are added at the appropriate positions.
Japanese Patent No. 3220560 (paragraphs 0011 and 0012)

  With the current machine translation technology, complete translation is impossible when translating sentences that can be interpreted in multiple ways. In such a case, post-editing work on the translation result is required. In particular, when the word order changes in the translation result, or when one word is composed of a plurality of words, depending on the position of the markup symbol, the markup symbol equivalent to the markup symbol is appropriately positioned. May be difficult to add to. Even in such a case, post-editing work on the translation result is required.

  In order to edit a translation result file (translation file) reflecting a markup symbol, it is common to use a text editor or an HTML editor. However, for example, when an editing operation is performed using a text editor, it is difficult for the user to specify the editing target portion because markup symbols are mixed in the edited character string.

  In an editing operation for reflecting the markup symbol at an appropriate position, the user needs to select the entire markup symbol and perform cut and paste. Such an editing operation has a risk that a part of the markup symbol is accidentally deleted.

  On the other hand, if the HTML editor is used, the above-mentioned danger when editing with a text editor is reduced. However, in order for a user to edit a translation using an HTML editor, it is often necessary to refer to a corresponding original sentence or a dictionary. Here, it is common that two files of the original text and the translated text are opened simultaneously. In this state, the user is forced to perform complicated operations such as switching the screen and scrolling to refer to the original text corresponding to the translation to be edited, or switching the screen to a dictionary reference tool activated separately to draw a dictionary. It is done.

  The present invention has been made in consideration of the above circumstances, and its purpose is to divide markup symbols included in a text original into markup symbols used in a sentence and markup symbols used outside a sentence, A machine translation editing device, a machine translation editing program, and a machine translation editing program capable of efficiently editing document data in which markup symbols and text parts are mixed by limiting markup symbols reflected in the translation to markup symbols in the sentence, and It is to provide a machine translation editing method.

  According to one aspect of the present invention, when a text source text in a first language including a markup symbol is translated into a second language to generate a translation, a word to which a markup symbol is added in the text source text A machine translation editing apparatus is provided that adds the markup symbol added to the word to the translated word corresponding to. The machine translation editing apparatus includes: a storage unit; a separation unit that separates the text source into a word as a markup symbol and a word as a text part that does not include the markup symbol; and the word separated by the separation unit In the case of a markup symbol, a markup symbol type for determining whether the markup symbol is an in-sentence markup symbol used in a sentence or an out-of-sentence markup symbol used outside a sentence Each of the words separated by the judging means and the separating means indicates whether the word is a markup symbol or a text part, and in the case of a markup symbol, the markup symbol type judging means A function for sequentially storing in the storage means in association with type information indicating the determined markup symbol type. A markup symbol that is reflected in the translation at the time of generation of the translation of the original text, and the out-of-sentence mark in the group of words stored in the storage means A source word list including a group of words excluding a word associated with the type information indicating that it is an up symbol, and type information indicating the text portion of the group of words stored in the storage unit; A translation of all the associated words, a type information indicating that the translation is a markup symbol in the sentence, and a markup symbol added to the translation of the word to which the associated markup symbol is added Source / translation word list generation means for generating a translation word list and storing it in the storage means; and based on the source word list and the translation word list Each and a display control means for editable displaying the original text and translation.

  According to the present invention, a markup symbol included in a text original is divided into a markup symbol used in the sentence and a markup symbol used outside the sentence, and the translated word to which the markup symbol is added is When editing document data that includes both markup symbols and text parts, the user can select markup symbols used outside the sentence. Since the markup symbol used in the sentence related to the word to be translated or the translation to be translated can be edited while being visually recognized, efficient editing is possible.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing a hardware configuration of a machine translation editing apparatus according to an embodiment of the present invention. In FIG. 1, a computer such as a personal computer (PC) 11 is connected to an external storage device 12 such as a hard disk drive. The external storage device 12 stores a machine translation editing program 120 for machine translation and translation result editing. The personal computer 11 includes a main memory 110. The computer 11 reads the machine translation editing program 120 stored in the external storage device 12 into the main memory 110 and executes it, thereby realizing machine translation and editing functions necessary for the machine translation editing apparatus. The machine translation editing program 120 may be read into the main memory 110 via a network.

  FIG. 2 is a block diagram showing a functional configuration of the machine translation editing apparatus of FIG. The machine translation editing apparatus includes a machine translation editing unit 20, an input unit 31, a display unit 32, a translation dictionary unit 33, and a buffer unit 34. The machine translation editing unit 20 is realized when the personal computer 11 in FIG. 1 reads the machine translation editing program 120 stored in the external storage device 12 into the main memory 110 and executes it. The machine translation editing unit 20 includes a system control unit 21, a translation unit 22, and an editing unit 23.

  The input unit 31 and the display unit 32 are provided in the personal computer 11 in FIG. The input unit 31 includes a keyboard (KB) 311 and a mouse 312. The input unit 31 inputs data of an original text file to be translated, that is, original text data (original text data of the first language) and various instructions (commands) through a communication means such as the Internet, a keyboard 311 and the like. It is also possible to input the original text data from the external storage device 12 of FIG. The display unit 32 displays the original text input by the input unit 31, that is, the first language text to be translated (original text), the second language text translated by the translation section 22 (translated text), and the like.

  The translation dictionary unit 33 is secured in the external storage device 12 in FIG. The translation dictionary unit 33 stores various dictionaries (dictionary data) (not shown) used in the translation process. The translation dictionary unit 33 also stores a tag type table 330.

  The buffer unit 34 is secured in the main memory 110 in FIG. The buffer unit 34 stores the first language original text data and the second language data as a translation result thereof in association with each other, and stores the attributes of the words constituting the first language and the second language. Used as The buffer unit 34 includes an original sentence buffer 341, a word list buffer 342, a sentence list buffer 343, an original sentence word list buffer 344, and a translated word list buffer 345.

  The system control unit 21 controls the entire machine translation editing apparatus shown in FIG. Here, the system control unit 21 sends, for example, original text data to be translated input by the input unit 31 to the translation unit 22. Further, the system control unit 21 sends the input instruction to the translation unit 22, the editing unit 23, the display unit 32, or the like that should execute the instruction.

  The translation unit 22 performs translation processing on the original text data (sentence data) in the first language sent by the system control unit 21 using the dictionary group stored in the translation dictionary unit 33. The editing unit 23 performs editing processing according to an editing instruction sent from the input unit 31 via the system control unit 21. An instruction for editing is given by, for example, a character deletion command (representing a key code) sent from the input unit 31 in response to pressing of the BackSpace key. The editing unit 23 identifies the character position where the cursor exists, and edits the corresponding character based on the character position and an instruction for editing. The editing unit 23 performs control such as copying, cutting, and moving of the selected text.

  FIG. 3 is a block diagram showing a functional configuration of the translation unit 22. The translation unit 22 includes a translated sentence generation unit 220, a separation unit 221, a tag type determination unit 222, a word list generation unit 223, a sentence list generation unit 224, an original / translated sentence word list generation unit 225, and a tag type designation unit 226.

  The translated sentence generation unit 220 generates a translated sentence by translating the original text of the first language including the tag (markup symbol) into the second language. When the translated sentence is generated, the translated sentence generation unit 220 adds the tag added to the word to the translated word corresponding to the word to which the tag is added in the original text. In the present embodiment, the tag reflected in the translation in this way is limited to a tag in a sentence to be described later.

  The separation unit 221 separates the text original sentence into a word as a tag and a word as a text part not including the tag. When the separated word is a tag, the tag type determination unit 222 determines the tag type of whether the tag is an in-sentence tag used in a sentence or an out-sentence tag used outside a sentence.

  The word list generation unit 223 includes, for each separated word, data type information indicating a data type indicating whether the word is a tag or a text part, and tag type information indicating a tag type in the case of a tag, It functions as an association means for associating and storing in the word list buffer 342. The sentence list generation unit 224 generates a list of sentences (sentence list) based on the contents (word list) of the word list buffer 342 in the buffer unit 34. Sentences included in this sentence list consist of a sequence of words as out-of-sentence tags, consisting of a sentence whose data type is treated as a tag (tag sentence) and a sequence of words other than out-of-sentence tags, and the data type as text It is classified as a sentence (text sentence) to be handled. The words other than the out-of-sentence tags include a word as a text part and a word as an in-sentence tag. In the present embodiment, only text sentences are processed as translation target sentences.

  The original / translated sentence word list generation unit 225 restricts the tag type reflected in the translated sentence to the intra-sentence tag when the translated sentence of the original text is generated. The original / translated word list generating unit 225 generates the original word list and the translated word list, and stores them in the original word list buffer 344 and the translated word list buffer 345 in the buffer unit 34, respectively. The original word list is a group of words constituting a text sentence (that is, an out-of-sentence tag) associated with data identification information indicating that it is text among sentences stored in the sentence list buffer 343. A group of words excluding words associated with the tag type information shown). The translated word list indicates the translated words of all the words associated with the data type information indicating the text portion in the group of words constituting the text sentence, and the out-of-sentence tag among the translated words. A tag added to the translation of the word to which the tag associated with the tag type information is added. The system control unit 21 is a display control unit that displays the original sentence and the translated sentence on the display unit 32 based on the original word list and the translated word list generated by the original / translated word list generation unit 225 and provides them to the user's editing operation. Function.

  FIG. 4 is a block diagram showing a functional configuration of the editing unit 23. The editing unit 23 includes a tag display designation unit 231, a detection unit 232, a tag determination unit 233, and a prohibition unit 234 in addition to the editing processing unit 230 that performs editing processing.

  The tag display designation unit 231 allows the user to designate whether or not to display markup symbols included in the original word list and the translated word list. The detection unit 232 detects that any character in the original sentence or the translated sentence is selected by a user operation while the original sentence and the translated sentence are displayed. The tag determination unit 233 determines whether a word including the detected character is a tag. When the word including the detected character is a tag, the prohibition unit 234 ignores the editing operation performed by the user on the character in the tag. In other words, editing the characters in the tag is prohibited.

  Next, the translation processing according to the present embodiment will be described with reference to FIGS. 5A, 5B, and 6, taking as an example the case of translating an original text in English (first language) into a translated text in Japanese (second language). To do. 5A and 5B are flowcharts showing the procedure of the translation process, and FIG. 6 is a flowchart showing the procedure of the display process for displaying the original sentence and the translated sentence.

  First, it is assumed that an instruction for translation of original English text data described in HTML is input by the input unit 31. This instruction is passed from the input unit 31 to the system control unit 21. Further, the original text data to be translated according to this instruction is input by the input unit 31. An example of the original English text data is shown in FIG.

  When receiving a translation instruction from the input unit 31, the system control unit 21 passes the original text data to be translated input by the input unit 31 to the translation unit 22. The translation unit 22 stores the original text data in the original text buffer 341 in the buffer unit 34 (step S100).

  Next, the translation unit 22 sets the character position pointer to the initial value 0 (step S110). This character position pointer indicates the position of the character to be referred to (to be extracted) next to the original text data stored in the original text buffer 341. Here, the character position pointer indicates the 0th (first) character of the original text data, that is, “<”. In step S110, the translation unit 22 sets both the word number N and the sentence number M to the initial value 1. Word number N and sentence number M indicate the Nth and Mth entries in word list buffer 342 and sentence list buffer 343, respectively.

  FIGS. 8 and 11 show how the contents of the word list buffer 342 and the sentence list buffer 343 change during the translation process, respectively. As is apparent from FIG. 8, each entry in the word list buffer 342 includes a field in which a word number is set, a data field in which a word as data is set, and the data (word) is a tag (markup symbol). Or a data type field in which data type information indicating a data type indicating a data type other than a tag (that is, a text part) is set, and the tag (markup symbol) when the data (word) is a tag The tag type (mark) that is a tag in a sentence (markup symbol in a sentence) used in a sentence or a tag that is not used in a sentence, that is, a tag outside a sentence (markout symbol outside a sentence) Tag type field in which tag type information indicating up symbol type) is set. Here, if the data type field and the tag type field are combined into one field, whether the data (word) set in the data field is a text part, an in-sentence tag, or an out-sentence tag It may be a field in which type information indicating is set.

  On the other hand, as is apparent from FIG. 11, each entry in the sentence list buffer 343 includes a field in which a sentence number is set, a data field in which a sentence (word string) as data is set, and the data (sentence). And a data type field in which data type information indicating a data type indicating whether it is a tag (markup symbol) or text is set. As described above, the sentences stored in the sentence list buffer 343 are classified into a tag sentence and a text sentence.

  After step S110, the translation unit 22 determines whether the character pointed to by the character position pointer indicates the start of a tag (markup symbol) (step S120). Here, if “<”, it is determined that the start of the tag is indicated, and the process proceeds to step S130. If it is not “<”, the process proceeds to step S140.

  In step S <b> 130, the translation unit 22 stores the character type “<” in the data field of the entry whose word number N is 1 in the word list buffer 342 and the data type information “tag” in the data type field. FIG. 8A shows the state of the word list buffer 342 at this time. When data is already stored in the entry having the word number N (= 1), the word number N is incremented by 1, and the character “<” is stored in the next word number entry.

  On the other hand, in step S140, the translation unit 22 determines whether the character indicated by the character position pointer indicates the end of the tag. If “>” at this time, the translation unit 22 determines the end of the tag and proceeds to the process of step S220. If it is not “>”, the translation unit 22 proceeds to S150.

  In step S150, the separation unit 221 of the translation unit 22 determines whether or not the character indicated by the character position pointer is a character used at the end of the sentence. In English, “.”, “!”, “?”, Etc. correspond to this. Here, in consideration of the case where “.” Is used as part of a word, such as “Mr.”, it is determined whether or not the separation unit 221 of the translation unit 22 is the end of the sentence by referring to the dictionary. Is done. The translation unit 22 proceeds to step S290 when it is determined that the sentence ends, and proceeds to step S160 when it is determined that the sentence does not end.

  In step S160, the separation unit 221 of the translation unit 22 determines whether the character indicated by the character position pointer is a character (first character) to be stored at the beginning of the character string field of the entry having the word number N in the word list buffer 342. Determine. If it is not the first character, the translation unit 22 proceeds to step S190. In the case of the first character, the tag type determination unit 222 of the translation unit 22 determines whether or not all the data of entries with word numbers 1 to N−1 in the word list buffer 342 are out-of-sentence tags (step S170). ).

  In the case of all out-of-sentence tags, the sentence list generation unit 224 of the translation unit 22 adds all the data in the word list buffer 342 (entries with the word numbers 1 to N−1) to the entry with the sentence number M in the sentence list buffer 343. Data) is stored (step S180). At this time, the translation unit 22 clears the contents of the word list buffer 342, returns the word number N to the initial value 1, and increments the sentence number M by 1. Next, the word list generation unit 223 of the translation unit 22 stores the character indicated by the character position pointer as the first character in the entry having the word number N of 1 in the word list buffer 342 (step S190). At this time, the word list generation unit 223 stores the attribute that the data type is text in the entry having the word number N of 1. On the other hand, if it is determined in step S160 that the character is not the first character, the word list generation unit 223 adds the character indicated by the character position pointer to the entry having the word number N in the word list buffer 342 (step S190).

  Next, the separation unit 221 of the translation unit 22 checks whether or not unprocessed data remains in the original text buffer 341 (step S200), and ends the process if there is no remaining data. If it remains, the separating unit 221 increments the character position pointer by one (step S210), and evaluates the next character indicated by the character position pointer after the increment (step S120).

  In this manner, the separation unit 221 repeats the process on the character string that constitutes the original text data stored in the original text buffer 341, thereby converting the character string into a tag (markup symbol) and a character (text) other than the tag. And to separate.

  Next, the case where it is determined in step S140 that the tag has ended will be described. In this case, the word list generation unit 223 of the translation unit 22 adds “>” to the entry whose word number is N in the word list buffer 342 (step S220). As a result, the word list buffer 342 is in a state where the tag <HTML> is stored as shown in FIG. Next, the word list generation unit 223 increments the word number N by 1 (step S221).

  Subsequently, the tag type determination unit 222 of the translation unit 22 determines whether or not the tag type stored in the entry having the word number N−1 in the word list buffer 342 is an out-of-sentence tag (step S230). . In HTML, usable tags and their roles are defined. For this reason, it is possible to easily determine whether or not the tag is an in-sentence tag used in a sentence (that is, an out-of-sentence tag) based on information on the tag and its role. In this embodiment, the tag type table 330 is used for this determination. An example of the data structure of the tag type table 330 is shown in FIG. As is clear from FIG. 10, the tag type table 330 includes, for each tag that can be used in HTML, information indicating the tag type, that is, an in-sentence tag or an out-of-sentence tag (tag type information). ) Is registered. The tag type table 330 is prepared in advance in the translation dictionary unit 33 as described above.

  On the other hand, in the case of XML or SGML, it is difficult to specify the role (type) of the tag. Therefore, when translating the original text described in XML or SGML, it is preferable to let the user specify the tag type as described below.

  First, the tag type designation unit 226 of the translation unit 22 extracts tags existing in the original text data from the original text data (original text file data) stored in the original text buffer 341. The tag type designation unit 226 displays the extracted tag list (tag list) on the display screen of the display unit 32 under the control of the system control unit 21. An example of the display screen at this time is shown in FIG. In addition to the tag list 71, an in-sentence tag designation button 72 and an out-sentence tag designation button 73 are displayed on the display screen shown in FIG. In addition, an in-sentence tag display area 74 and an out-of-sentence tag display area 75 are secured on the display screen. For example, the user operates the mouse 312 of the input unit 31 to select an arbitrary tag from the tag list 71. The user selects either the in-sentence tag designation button 72 or the out-of-sentence tag designation button 73 in order to designate whether the selected tag is an in-sentence tag or an out-sentence tag. Then, the system control unit 21 displays the selected tag in the in-sentence tag display area 74 if the in-sentence tag is specified, and displays the selected tag in the out-of-sentence tag display area if the out-of-sentence tag is specified. 75. The tag type designation unit 226 is specific to the original text data to be translated in the same format as the tag type table 330 shown in FIG. Generate a tag type table. The generated table is used only for translation processing of the original text data.

  Now, in this embodiment in which the original text to be translated is described in HTML as shown in FIG. 7, a tag type table 330 (see FIG. 10) prepared in advance is used to determine the tag type. . Here, information indicating the type (attribute) of the tag that matches the tag in the original sentence buffer 341 is stored in the word list buffer 342 in association with the tag. In the example of the tag <HTML> stored in the original text buffer 341, the tag matches <HTML> in the tag type table 330 shown in FIG. In this case, the tag type determination unit 222 of the translation unit 22 acquires information “out of sentence” from the tag type table 330 as an attribute of <HTML>. That is, the tag type determination unit 222 determines that <HTML> is an out-of-sentence tag (step S230). Then, as shown in FIG. 8C, the word list generation unit 223 of the translation unit 22 adds “HTML” to the entry of word number 1 (word number N−1) in the word list buffer 342 as the attribute of <HTML>. The tag type information indicating “” is stored.

  As described above, when the tag is determined to be an out-of-sentence tag in step S230, the word list generation unit 223 uses the tag type indicating “out-of-sentence” in the entry of the word number N−1 in the word list buffer 342 in which the tag is stored. Store information. In this case, the tag type determination unit 222 determines whether or not all data of entries having word numbers 1 to N-2 in the word list buffer 342 are out-of-sentence tags (step S240). If all the tags are out-of-text tags, the separation unit 221 proceeds to step S210 through step S200, and acquires the next data from the original text buffer 341.

  On the other hand, when it is not determined that all the data of entries having the word numbers 1 to N-2 are out-of-sentence tags, that is, when the data of entries having word numbers 1 to N-2 are other than out-of-sentence tags In step S240, the tag type determination unit 222 determines that the text or in-sentence tag is stored in the entry having the word number 1 to N-2. In this case, the translated sentence generation unit 220 uses the data in the buffer 342 excluding the extra-sentence tag stored in the word list buffer 342 at the end (that is, data of entries having word numbers 1 to N-2) as the translation target sentence. Is translated (step S250). Since this translation procedure is well known from the past as described in Patent Document 1 and the like, a description thereof will be omitted. The sentence list generation unit 224 stores the translation target sentence used for the translation in step S250 in an entry whose sentence number is M in the sentence list buffer 343, and increments the sentence number M by 1 (step S260). At this time, the word list generation unit 223 moves the data of the entry whose word number is N−1 in the word list buffer 342 to the entry of word number 1 in the buffer 342 and sets the word number N to 2.

  On the other hand, if it is determined in step S230 that the tag is an in-sentence tag, the tag type determination unit 222 determines that all the data of the entries having the word numbers 1 to N-2 in the word list buffer 342 are the out-of-sentence tags, as in step S240. It is determined whether or not (step S270). In the case of all out-of-sentence tags, the sentence list generation unit 224 takes the entry data (that is, out-of-sentence tag group) with the word numbers 1 to N-2 as one sentence, and the entry with the sentence number M in the sentence list buffer 343. And the sentence number M is incremented by 1 (step S280). At this time, the word list generation unit 223 moves the data stored last in the word list buffer 342, that is, the data of the entry having the word number N-1 to the entry of the word number 1 in the buffer 342, and the word number Set N to 2. . On the other hand, when all the data of the entries having the word numbers 1 to N-2 are other than the out-of-sentence tags, that is, in the case of text or in-sentence tags (step S270), the separation unit 221 goes to step S210 through step S200. Then, the next data is acquired from the original text buffer 341.

  Next, the case where it is determined that the sentence ends in step S150 will be described. In this case, the word list generation unit 223 stores the character determined to be the end of the sentence (end of sentence character) in an entry whose word number is N in the word list buffer 342 (step S290). Then, the translated sentence generation unit 220 performs translation using all data currently stored in the word list buffer 342 (that is, data stored in entries having word numbers 1 to N) as translation target sentences (step S300). . The sentence list generation unit 224 stores the sentence to be translated in an entry whose sentence number is M in the sentence list buffer 343, and increments the sentence number M by 1 (step S310).

  Next, the above-described translation processing in the case where translation for the original English text data shown in FIG. 7 is instructed will be described more specifically. First, it is determined that the first character “<” of the character string (tag) “<HTML>” in the original text shown in FIG. 7 is the start of the tag (step S120). As shown in FIG. 8A, the character “<” is stored in the entry of word number 1 in the word list buffer 342 (step S130).

  Subsequently, “H”, “T”, “M”, “L” in the character string “<HTML>” are sequentially stored in the entry of the word number 1 (step S190). Next, “>” in the character string “<HTML>” is determined to be the end of the tag (step S140), and the “>” is stored in the entry of the word number 1 as shown in FIG. 8B. (Step S220). Next, the tag type table 330 shown in FIG. 10 is referred to, and it is determined that the character string “<HTML>” stored in the entry of the word number 1 is an out-of-sentence tag (step S230). As a result, the entry of the word number 1 stores information indicating that the tag type of “<HTML>” is a non-sentence tag, as shown in FIG. 8C. At this time, the word number N is incremented from 1 to 2.

  Next, the processing of the character string (tag) “<BODY>” portion in the original text shown in FIG. 7 is similarly performed character by character. Similarly to “<HTML>”, “<BODY>” is also determined to be an out-of-sentence tag, so the word list buffer 342 is in the state shown in FIG. At this time, the word number N is incremented from 2 to 3.

  Subsequently, the processing of the portion of the character string “Click” in the original text shown in FIG. 7 is performed as follows. First, when the character “C” is acquired, it is determined that the “C” is the first character in the entry of the word number N (that is, the word number 3) in the word list buffer 342 (step S160). At this time, as apparent from FIG. 8D, the data “<HTML>” and “<BODY” stored in the entries of the word numbers 1 to N−1 (that is, the word numbers 1 and 2) in the word list buffer 342. > "Are all out-of-sentence tags (step S170). In this case, “<HTML>” and “<BODY>” are stored as one data in the entry having the sentence number M of 1 in the sentence list buffer 343 (step S180). The state of the sentence list buffer 343 at this time is shown in FIG. In step S180, the contents of the word list buffer 342 are cleared and the word number N (= 3) is initialized to 1.

  Next, the character “C” is stored in the entry of word number 1 in the word list buffer 342 and the attribute “text” is stored as shown in FIG. 8E (step S190). The remaining “lick” part is added to the position following “C” in the entry of word number 1 (step S190).

  Subsequently, the character string “<A href =” http: // www. aaa. The process of the part “com>” is performed as follows: First, when “<” is acquired, the “<” is stored in the entry of the word number 2 in the word list buffer 342 (step S130). Next, of the remaining “A href =” http://www.aaa.com> ”, a character string excluding“> ”is sequentially added to a position subsequent to“ <”in the entry of the word number 2. (Step S190). When the last “>” is stored in the entry of the word number 2 (step S220), the character string “<A href =” http: // www. aaa. The tag type of “com>” is performed with reference to the tag type table 330 shown in FIG. 10 (step S230).

  The character string “<A href =” http: // www. aaa. com> ”is recognized as a tag“ <A> ”together with the last character“> ”when the character string“ <A ”is followed by a space. In this case, the tag type table 330 shown in FIG. As is apparent, the tag “<A>” is determined to be an in-sentence tag (step S230), and the data of the entry of the word number smaller than the word number 2 of the word list buffer 342, that is, the entry of the word number 1 is stored. It is determined whether or not the tag is an out-of-sentence tag (step S270) At this time, the character string “Click” is stored in the entry of word number 1, and the character string “Click” is included in the out-of-sentence tag. It is determined that it is not.

  Subsequently, the same processing is performed on the character strings “Here” and “</A>” in the original text shown in FIG. 7, and the “Here” and “</A>” are respectively stored in the word list buffer 342. It is stored in the entries of word numbers 3 and 4. As a result, the word list buffer 342 is in the state shown in FIG.

  Further, “.” Following “</A>” is determined as the end-of-sentence character (step S150), and as shown in FIG. 8G, the “.” Is added to the entry of word number 5 in the word list buffer 342. Stored (step S290). Then, all the data stored in the word list buffer 342 at that time, that is, the data stored in the entries having the word numbers 1 to 5) are translated as translation target sentences (step S300). In this translation, as described in Patent Document 1, the tag included in the original sentence (translation target original sentence) is reflected in the translated sentence. It should be noted here that the translation target text does not include extra-text tags. At this time, the original sentence / translated word list generating section 225 of the translating section 22 generates an original sentence word list corresponding to the contents of the word list buffer 342 and a list of words constituting the translated sentence reflecting the tag information (translated word list). . The original word list and the translated word list include information indicating the correspondence between words (text) between the original sentence and the translated sentence, and the correspondence between the tag attached to the original sentence and the tag reflected in the translated sentence.

  The sentence list generation unit 224 stores the original text to be translated stored in the word list buffer 342 in the entry of sentence number 2 in the sentence list buffer 343 as shown in FIG. 11B (step S310). At this time, the original / translated word list generating unit 225 stores the original word list and the translated word list corresponding to the original sentence to be translated in the original word list buffer 344 and the translated word list buffer 345, respectively. FIGS. 12 and 13 show the contents (original word list and translated word list) of the original word list buffer 344 and the translated word list buffer 345 at this time point, respectively. Each entry in the word list buffers 344 and 345 includes a correspondence field instead of the tag type field in each entry in the word list buffer 342.

  12 and 13, the same unique numerical value (correspondence information) indicates that the word (text or tag) of the entry set in the correspondence field corresponds. For example, the same numerical value 1 is set in the correspondence field of the entry storing the text “Click” in the source word list buffer 344 and the entry storing the text “click” in the translated word list buffer 345. Has been. This indicates that “Click” corresponds to “Click”. Note that a special numerical value indicating this, for example, -1, is set in the correspondence field of the entry in which words having no correspondence between the original word list and the translation list are stored.

  Subsequently, the same processing is performed on the character strings “</ BODY>” and “</ HTML>” in the original text shown in FIG. 7, and the “</ BODY>” and “</ HTML>” Are also determined as out-of-sentence tags. As a result, the sentence list buffer 343 finally becomes in the state shown in FIG.

  When the translation process by the translation unit 22 is completed, the system control unit 21 executes a display process using the display unit 32. Hereinafter, this display process will be described with reference to the flowchart of FIG. 6 and the display example of the original text and the translated text of FIG.

  First, the system control unit 21 sets the sentence number M designating the Mth entry in the sentence list buffer 343 to an initial value 1 (step S500). The system control unit 21 refers to the entry of the sentence number M (= 1) in the sentence list buffer 343 (step S501), and determines whether or not a sentence exists in the entry (step S510). If there is no sentence in the entry of sentence number M, the system control unit 21 ends the display process. If there is a sentence in the entry with the sentence number M, the system control unit 21 determines whether or not the sentence is a tag based on the data type field of the entry (step S520).

  When the type of sentence existing in the entry of sentence number M is a tag (step S520), the system control unit 21 increments the sentence number M by 1 (step S521), and the sentence list specified by the sentence number M after this increment The entry in the buffer 343 is referred to (step S501).

  On the other hand, when the type of the sentence existing in the entry with the sentence number M (that is, the original sentence with the sentence number M) is text (step S520), the system control unit 21 selects the original sentence word list associated with the original sentence with the sentence number M. And the translated word list are obtained from the original word list buffer 344 and the translated word list buffer 345, respectively (step S530). The system control unit 21 repeats the following processing for the words included in each of the original word list and the translated word list.

  First, the system control unit 21 performs processing for extracting unprocessed words from the original word list or the translated word list (step S540). If there is an unprocessed word and the unprocessed word can be extracted (step S550), the system control unit 21 determines the data type associated with the extracted word. (Step S560).

  If the data type of the extracted word is text, that is, if it is not a tag, the system control unit 21 displays the word using the display unit 32 (step S570). On the other hand, when the data type of the extracted word is a tag, the system control unit 21 performs a process of extracting the next unprocessed word from the original word list or the translated word list (step S540). Here, it is assumed that the processing for the words included in the original word list is executed first, but this may be reversed. In addition, for example, the word numbers may be alternately extracted from the source word list and the translated word list in ascending order.

  When all the words in the original word list and the translated word list are processed (step S550), the system control unit 21 increments the sentence number M by 1 (step S521), and the sentence number M after the increment is used. The entry of the designated sentence list buffer 343 is referred to (step S501). In this way, the system control unit 21 performs processing up to the last sentence. As a result, in the case of the sentence list shown in FIG. 11C, the original word list and the translated word list shown in FIGS. 12 and 13 respectively corresponding to the sentence with sentence number 2 in the sentence list, the original sentence shown in FIG. The translation is displayed. Since the original sentence and the translated sentence do not include a tag, the user can edit the original sentence and the translated sentence without being aware of the presence of the tag. In FIG. 14, the original sentence and the translated sentence are arranged at the upper limit for convenience of the speed diagram, but they may be arranged on the left and right.

  In the above embodiment, it is assumed that the original sentence buffer 341, the word list buffer 342, the sentence list buffer 343, the original sentence word list buffer 344, and the translated word list buffer 345 are secured in the buffer unit 34 in advance. However, these buffers may be secured as necessary. Further, for example, each entry of the word list and sentence list may be generated sequentially.

[Modification]
Next, a modification of the above embodiment will be described. The first feature of this modification is that, unlike the above embodiment, tags are also displayed. The second feature of this modification is that the tag and text are displayed on the display screen so that the user can easily identify them. Hereinafter, the display processing in the modification of the above embodiment will be described with reference to the display example of the original text and the translation in FIG. 15 and the flowchart in FIG. In FIG. 16, the same parts as those in FIG. 6 are denoted by the same reference numerals.

  First, it is assumed that the data type of the word extracted from the original word list or the translated word list is text (step S560). In this case, the system control unit 21 cancels, for example, highlighting as the display attribute of the display font of the extracted word (step S561), and displays the word (step S570). On the other hand, when the data type of the extracted word is a tag (step S560), the system control unit 21 sets highlighting as a display attribute of the display font of the word (step S562) and displays the word ( Step S570).

  As described above, in this modification, in the case of the sentence list shown in FIG. 11C, the original word list and the translated word list shown in FIG. 12 and FIG. 13 corresponding to the sentence with sentence number 2 in the sentence list, The original sentence and the translated sentence shown in FIG. 15 are displayed. In FIG. 15, text and tags are mixedly displayed in the original sentence and the translated sentence. However, the tag portion is highlighted as shown by hatching in FIG. For this reason, the user can identify the tag portion (that is, the markup symbol) and the text portion at a glance, and can confirm the position of the tag in the original sentence and the translated sentence. As a result, the user can efficiently edit, for example, the translated sentence without erroneously correcting the tag portion. In FIG. 15, the original sentence and the translated sentence are arranged at the upper limit for convenience of the speed diagram, but they may be arranged on the left and right.

  It should be noted that whether the tag is not displayed as in the above embodiment or whether the tag is displayed as in this modification can be designated by the user by the tag display designation unit 231 of the editing unit 23. The tag display designation unit 231 may be included in the system control unit 21.

  Hereinafter, editing processing according to a user's editing operation, which is applied in the present modification, will be described with reference to the flowchart of FIG. In this editing process, editing of the original sentence or the translated sentence displayed on the display unit 32 is controlled according to the type of data.

  First, the character string “<A href =” http: // www. aaa. It is assumed that the BackSpace key on the keyboard 311 included in the input unit 31 is pressed in a state where the cursor is present in “com>”. Then, a key code unique to the BackSpace key is input by the input unit 31. The system control unit 21 sends this input key code to the editing unit 23.

  Upon receiving the key code sent by the system control unit 21, the editing processing unit 230 of the editing unit 23 determines whether or not an editing key has been pressed based on the key code (step S700). Here, the edit key refers to a BackSpace key, a Delete key, a character key, a numeric key, and the like. Here, the key code of the BackSpace key is sent by the system control unit 21. In this case, the edit processing unit 230 determines that the edit key has been pressed (step S700).

  Then, the detection unit 232 of the editing unit 23 acquires the current cursor position, and detects that the character in the sentence existing at the cursor position has been selected by the user (step S710). The detecting unit 232 acquires a word list associated with the sentence existing at the cursor position (step S720). Here, the translated word list (see FIG. 13) is acquired from the original word list stored in the original word list buffer 344 and the translated word list stored in the translated word list buffer 345.

  Next, the detection unit 232 sets the word number N to an initial value 1 (step S721), and refers to the word with the word number N in the word list acquired in step S720 (step S730). The detection unit 232 acquires the length (character string length) LN of the character string from the beginning of the word list to the word with the word number N (step S740). When N = 1, the character string length of the word with the word number 1, that is, the first word is acquired.

  Next, the detection unit 232 determines whether or not the character string length LC from the top of the word list to the current cursor position is longer than the character string length LN from the top of the word list to the word with the word number N. (Step S750). If LC> LN (step S750) and the word number N is not the last word in the word list (step S780), the detection unit 232 increments the word number N by 1 (step S761). Then, the detecting unit 232 refers to the word with the incremented word number N (step S730). The detecting unit 232 acquires the character string length LN from the top of the word list to the word with the incremented word number N (step S740). This character string length LN is acquired by adding the character string length of the word with the word number N to the character string length from the top of the word list acquired at the previous step S740 to the word with the word number N-1. Is done. That is, the detection unit 232 acquires the character string length LN by sequentially adding the character string lengths of the words from the first word (word with word number 1) to the word with word number N in the word list.

  Eventually, it is assumed that the character string length LN acquired in step S740 becomes equal to or longer than the character string length LC (step S750). That is, it is assumed that the character string length LN from the beginning of the word list to the word with the word number N is equal to or longer than the character string length LC from the beginning of the word list to the cursor position. In this case, the detection unit 232 specifies the word with the current word number N as the word where the current cursor is located.

  Then, the tag determination unit 233 of the editing unit 23 determines whether the data type of the word at the specified cursor position, that is, the word with the word number N is a tag (step S770). The cursor is “<A href =” http: // www. aaa. In the present modification example that exists in “com>”, the word at the cursor position is determined as a tag. In this way, when the word at the cursor position is determined as a tag (step S770), the prohibition unit 234 is an editing processing unit. 230 is controlled so as to ignore the key code of the BackSpace key (edit key) sent from the system control unit 21. As a result, the editing operation by the user on the tag attached to the translation is prohibited. Similarly, user editing operations on tags included in the original text are also prohibited.

  When the word including the original text or the text in the translated text (here, the text in the translated text) determined by the cursor is determined as a tag, the system control unit 21 selects the determined tag (word). For example, if the determined (selected) tag is associated with the tag in the original sentence list or the translated word list by the correspondence information in the original sentence list or the translated sentence word list, the user has a corresponding relationship with the determined tag. Both tags are displayed with special display attributes so that they can be seen.

  On the other hand, when it is determined that the word at the cursor position, that is, the word with the word number N is text (step S770), the editing processing unit 230 applies the character in the word with the word number N from the system control unit 21. Editing specific to the sent key code is performed (step S780). For example, if the key code of the BackSpace key is sent, the character immediately before the cursor position in the word of the word number N (edit key) is deleted, and the word list (here, the translated word list) is updated. .

  On the other hand, if the key code sent by the system control unit 21 is not the key code of the editing key (step S700) but the key code of the cursor movement key (step S790), the editing processing unit 230 will execute the cursor movement key. The cursor is moved based on the key code (step S800).

  In addition, this invention is not limited to the said embodiment or its modification example as it is, A component can be deform | transformed and embodied in the range which does not deviate from the summary in an implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment or the modification thereof. For example, you may delete a some component from all the components shown by embodiment or its modification.

The block diagram which shows the hardware constitutions of the machine translation editing apparatus which concerns on one Embodiment of this invention. The block diagram which shows the function structure of the machine translation editing apparatus of FIG. The block diagram which shows the structure of the translation part 22 in FIG. The block diagram which shows the structure of the edit part 23 in FIG. The flowchart which shows the procedure of the translation process in the embodiment. The flowchart which shows the procedure of the translation process in the embodiment. The flowchart which shows the procedure of the display process in the embodiment. The figure which shows the example of the original text data which should be translated in the same embodiment. The figure which shows a mode that the content of the word list buffer 342 changes in the translation process of the embodiment. The example of a display screen for making a user specify the classification of a tag in the embodiment is shown. The figure which shows the example of a data structure of the tag classification table 330 applied in the embodiment. The figure which shows a mode that the content of the sentence list buffer 343 changes in the translation process of the embodiment. The source word list generated in the translation process of the embodiment is stored in the source word list buffer 344. A translated word list generated in the translation process of the embodiment is stored in the translated word list buffer 345. The figure which shows the example of a display screen containing the original sentence and translation in the display process of the embodiment. The figure which shows the example of a display screen containing the original sentence and translation in the display process of the modification of the embodiment. The flowchart which shows the procedure of the display process in the said modification. The flowchart which shows the procedure of the edit process in the said modification.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 11 ... Personal computer, 12 ... External storage device, 20 ... Machine translation editing part, 21 ... System control part, 22 ... Translation part, 23 ... Editing part, 31 ... Input part, 32 ... Display part, 33 ... Translation dictionary part, 34 ... Buffer unit, 110 ... Main memory, 120 ... Machine translation editing program, 220 ... Translation sentence generation unit, 221 ... Separation unit, 222 ... Tag type determination unit, 223 ... Word list generation unit, 224 ... Sentence list generation unit, 225 ... Original / translated word list generation unit, 226 ... Tag type designation unit, 230 ... Tag display designation unit, 231 ... Tag display designation unit, 232 ... Detection unit, 233 ... Tag determination unit, 234 ... Inhibition unit, 341 ... Original text Buffer, 342 ... Word list buffer, 343 ... Sentence list buffer, 344 ... Original word list buffer, 345 ... Translated word list buffer.

Claims (11)

When a text source text in a first language including a markup symbol is translated into a second language to generate a translation, it is added to the translation word corresponding to the word to which the markup symbol is added in the text source text. In the machine translation editing device that adds the markup symbols that have been
Separating means for separating the storage means and the original text into a word as a markup symbol and a word as a text part not including the markup symbol;
When the word separated by the separating means is a markup symbol, the markup symbol indicates whether the markup symbol is an in-sentence markup symbol used in a sentence or an out-of-sentence markup symbol used outside a sentence. Markup symbol type determination means for determining a symbol type;
Each of the words separated by the separating means indicates whether the word is a markup symbol or a text part, and in the case of a markup symbol, the mark determined by the markup symbol type determining means Association means for sequentially storing in the storage means in association with type information indicating an up symbol type;
When generating the translation of the original text, the markup symbol reflected in the translation is limited to the in-sentence markup symbol, and is the extrasentence markup symbol of the group of words stored in the storage means All of the original word list including a group of words excluding the word associated with the type information indicating that, and all of the group information associated with the type information indicating the text portion of the group of words stored in the storage unit A translated word list including a translation of the word and a markup symbol added to the translation of the word to which the markup symbol associated with the type information indicating that it is the in-sentence markup symbol of the translation is added Source / translation word list generation means for generating and storing in the storage means;
A machine translation editing apparatus comprising: display control means for displaying an original sentence and a translated sentence in an editable manner based on the original sentence word list and the translated word list, respectively. A markup symbol type table in which type information indicating a markup symbol type is registered in association with each of a plurality of different markup symbols;
The markup symbol type determination means, when the word separated by the separation means is a markup symbol, based on the type information registered in the markup symbol type table in association with the markup symbol. The machine translation editing apparatus according to claim 1, wherein a markup symbol type of the up symbol is determined. A markup symbol type specifying means for allowing a user to specify a markup symbol type of a markup symbol included in the original text,
The machine translation editing apparatus according to claim 1, wherein the markup symbol type determining unit determines the markup symbol type based on a user designation according to the markup symbol type specifying unit. The markup symbol type designation means displays a list of markup symbols included in the text source text, and allows the user to specify the markup symbol type of each markup symbol in the list, thereby allowing the text source text to be specified. Generate a markup symbol type table in which type information indicating the markup symbol type is registered in association with each of the included markup symbols,
The markup symbol type determining means, when the word separated by the separating means is a markup symbol, based on the type information registered in the markup symbol type table in association with the markup symbol, The machine translation editing apparatus according to claim 3, wherein a markup symbol type of the markup symbol is determined. The original / translated word list generating means sets data type information indicating whether the word is a markup symbol or a text part for each word in the original word list, and the translated word in the translated word list And for each markup symbol, data type information indicating that it is a text part and data type information indicating that it is a markup symbol are set, and the display control means includes the source word list and the translated word list. 2. The machine translation editing apparatus according to claim 1, wherein the markup symbol and other than the markup symbol are displayed with different attributes based on the included data type information. Markup symbol display designating means for allowing the user to designate whether or not to display the markup symbols included in the original word list and the translated word list,
When it is specified that the markup symbol is not displayed, the display control means, based on the data type information included in the original word list and the translated word list, only the text part as the original sentence and the translated sentence. The machine translation editing apparatus according to claim 5, wherein the machine translation editing apparatus is displayed. Detecting means for detecting that an original character or an arbitrary character in the translated text is selected by a user operation while the original text and the translated text are displayed by the display control means;
Markup symbol determination means for determining whether a word including the character detected by the detection means is a markup symbol;
2. The machine translation according to claim 1, further comprising: a prohibiting unit that prohibits editing of a character in the markup symbol when a word including the character detected by the detecting unit is a markup symbol. Editing device. Detecting means for detecting that an original character or an arbitrary character in the translated text is selected by a user operation while the original text and the translated text are displayed by the display control means;
Markup symbol determination means for determining whether a word including the character detected by the detection means is a markup symbol;
The original sentence / translation word list generating means sets correspondence information for associating each word in the original word list with a translation or a markup symbol in the corresponding translation word list corresponding to the word, and in the original word list For each translated word and markup symbol, set correspondence information for associating the translated word and markup symbol with the corresponding word in the original word list,
When it is determined that a word including an arbitrary character in the original sentence or the translated sentence is the markup symbol, the display control means determines the determined markup symbol and the original word list or the translated word. The markup symbol in the translation or the original sentence associated with the correspondence information in the list is displayed so as to be visible when there is a correspondence with the determined markup symbol. The machine translation editing apparatus according to claim 1. When a text source text in a first language including a markup symbol is translated into a second language to generate a translation, it is added to the translation word corresponding to the word to which the markup symbol is added in the text source text. A machine translation editing program for adding a markup symbol,
In a computer comprising input means, storage means and display means ,
Storing the original text input by the input means in the storage means;
Separating the original text stored in the storage means into a word as a markup symbol and a word as a text part not including the markup symbol in order from the top;
When the separated word is a markup symbol, a markup symbol type indicating whether the markup symbol is an in-sentence markup symbol used in a sentence or an out-of-sentence markup symbol used outside a sentence A determining step;
For each word separated from the original text stored in the storage means, the word indicates whether the word is a markup symbol or a text part, and in the case of a markup symbol, the determination and storing in the storage means in association with the type information indicating the mark-up symbols type is,
When generating the translation of the original text stored in the storage means, the markup symbols reflected in the translation are limited to the in-sentence markup symbols, and the group of words stored in the storage means A source word list including a group of words excluding a word associated with type information indicating an extra-sentence markup symbol, and a text part of the group of words stored in the storage means A translation of all words associated with the type information, and a markup symbol added to the translation of the word to which the markup symbol associated with the type information indicating the markup symbol in the sentence is added. Generating a translated word list including and storing the translated word list in the storage means;
Machine translation editing program for executing a step of causing the display by the display means editable original and translation, respectively, based on the original word list and the translation word list stored in the storage means. The computer further includes a storage device that stores a markup symbol type table in which type information indicating a markup symbol type is associated with each of a plurality of different markup symbols.
In the determining step , the markup symbol type of the separated markup symbol is determined based on the type information registered in the markup symbol type table in association with the separated markup symbol. The machine translation editing program according to claim 9. A markup symbol is added to the original text when the text source of the first language including the markup symbol is translated into the second language by the computer having the input means, the storage means, and the display means to generate the translation. A machine translation editing method for adding a markup symbol added to a word to a translation corresponding to a word,
The computer storing the original text input by the input means in the storage means;
A step wherein the computer, to separate the text original text stored in said storage means from the top and words as words and text parts without markup symbol as a mark-up symbols in order,
If the separated word is a markup symbol, the computer marks whether the markup symbol is an in-sentence markup symbol used in a sentence or an out-of-sentence markup symbol Determining the up symbol type;
Each time a word is separated from the original text stored in the storage means, the computer indicates whether the word is a markup symbol or a text part, and in the case of a markup symbol, the computer and storing in the storage means in association with the type information indicating the determination markup symbols type,
When generating the translation of the original text stored in the storage means, the computer restricts the markup symbols reflected in the translation to in-sentence markup symbols, and the words stored in the storage means A source word list including a group of words excluding a word associated with the type information indicating the extra-sentence markup symbol in the group, and the text portion of the group of words stored in the storage unit A translation of all the words associated with the type information indicating that it is present and a translation of the word to which the markup symbol associated with the type information indicating that it is a markup symbol in the sentence is added. Generating a translated word list including a markup symbol and storing it in the storage means;
The computer comprises the step of causing the display means to display the original sentence and the translated sentence based on the original word list and the translated word list stored in the storage means, respectively. Editing method.

Images