JPS62271055A - Editing system for translated sentence - Google Patents

Abstract

PURPOSE:To wuickly produce the accurate and natural translated sentences by providing a free editing mode and an after-editing mode and selecting optionally a free editing mode or an after-editing mode for editing jobs and therefore correcting freely the translation results with high efficiency or using the translation information as necessary. CONSTITUTION:When the translation processing is through with an interpreting device main body 6, the translation results are stored in a translated sentence file 25 together with the data on translated sentences and the translation information on each sentence stored in a translation information file 26 respectively. A free editing part 19A reads the results of translated sentences out of the file 25 to display them on a CRT 4 for each sentence and then carries out a free editing mode to edit freely the translated sentences by the commands and characters inputted through a keyboard 1. While an after-editing part 19B reads the translation information out of the file 26 after the after-editing mode is selected via the keyboard 1 in a free editing mode and displays it on the CRT 4 to perform the after-editing mode process where the translated sentences can be corrected based on said translation information. Thus an operator can correct easily the translation results with high efficiency and with use of the translation information as necessary. Then the natural and accurate translated sentences are quickly obtained.

Description

[Detailed Description of the Invention] 3. Detailed Description of the Invention Technical Field The present invention relates to a translation editing method for creating a more accurate and natural translation by correcting the translation result by a translation device using a computer system. .

BACKGROUND OF THE INVENTION In recent years, various automatic or semi-automatic translation devices using computer systems have been developed.

In this kind of translation 'Wffi', the original text is generally input using an input device such as a keyboard or an optical character reader, and it is morphologically analyzed and syntactically analyzed using pre-prepared word dictionaries and analysis rules. analysis, structural transformation,
Translation processing is performed by executing translation processes such as translation generation, and the translated text that is the translation result is displayed on a display or printed out using a printer.

However, the algorithm for performing the above-mentioned translation processing is complex, and it is extremely difficult to completely perform translation according to the customs of natural language.

Therefore, it is necessary for the operator to check the correspondence between the original text and the translated text and correct any mistranslations or inappropriate translations.

For this reason, such translation devices generally use "post-editing" in which an operator corrects the translated text by using the information used by the device when translating the translation result (this is referred to as "translation information" in this specification). and.

The operator can freely modify or change the translation results.
"Free Collection 1111" is now available.

However, in an editing method in which post-editing is always performed before free editing, there is a problem in that the degree of freedom of the user is restricted. In addition, editing methods that lose translation information for sentences that have been post-edited have the problem of not being able to perform post-editing again if you want to do it again. There was a lack of consideration.

The invention of objective 2 was made in view of this background, and it allows an operator to efficiently modify the translation results of a translation device freely or as needed using translation information.
The purpose is to quickly obtain more accurate and natural translations.

Structure In order to achieve the above-mentioned object, when the translation process is completed by the translation device, the translation result is displayed on a sentence-by-sentence basis and the translated sentence is set in a free editing mode in which the translated sentence can be freely modified.
When the latter g collection is selected in this free editing mode, translation information regarding each specified sentence is also displayed,
A post-editing mode is entered in which the translated sentence can be corrected using the translation information, and when the 8g page is finished, the post-editing result of the sentence specified at that time is memorized and the mode returns to the free editing mode. When post-editing is selected again for the sentence,
The memorized post-editing result corresponding to the sentence is displayed and the post-editing mode is entered. When the initial state is selected in the post-gt collection mode, the translation result corresponding to the specified sentence is displayed. A translated text editing method is provided, which is characterized by making it possible to create a tttgI collection for the translated text.

Hereinafter, a detailed explanation will be given based on one embodiment of the present invention.

FIG. 2 is an external perspective view showing an example of the system configuration of an English-Japanese automatic translation device embodying the present invention.

This system has character keys such as alphanumeric keys and function keys such as cursor movement keys as input devices, a keyboard for inputting operator instructions, and an image image that scans the document optically to generate image information including characters. It is equipped with an image scanner 2 for inputting data, and an optical character recognition bag [3] for converting the image data read by the image scanner 2 into character code data.

The image scanner 2 and the optical character recognition device 3 constitute an optical character reading bag r11 (OCR).

Additionally, there is a CRT display device (hereinafter simply referred to as "display J") 4 which is a display means for displaying various information including guidance for the operator, and an output device which prints out the translated text translated by the main body 6 of this translation device. A printer 5 such as a laser printer is provided.

The main body 6 has a computer function and is equipped with a hard disk device (HDD) 7 and a floating disk device (FDD) 8 as storage devices.

FIG. 3 shows an example of the system configuration of this English-Japanese automatic translation device in the form of functional blocks as "... sections".

The control unit 10 is a mini-computer that centrally controls the operation of the entire system, and includes a central processing unit (CPU), a program memory (ROM), a data memory (RAM), and the like.

The input device 11 is a device for inputting an English manuscript 12 to be translated into Japanese, and is a device for inputting an English manuscript 12 to be translated into Japanese. This is an optical character reading device (OCR) consisting of an image scanner 2 and an optical character recognition device 3, and is inputted either manually by an operator or by reading English text recorded on paper.

Note that the HDD 7 provided in the main body 6 serves as this input device 11.
Alternatively, the FDD 9 can be used to directly read English data recorded on a storage medium such as a magnetic disk.

The English manuscript (Jilif text) inputted by the input device 11 is read into the pre-editing section 13 and pre-processed for translation. Here, we mainly perform sentence recognition and processing of unknown words. This functions as part of morphological analysis.

The pre-edited English text data is transferred to the morphological analysis unit 14 together with the information obtained in the pre-editing. Morphological analysis section 14
Then, the word dictionary 22 stored in the dictionary file is indexed, the input English sentence data is divided into sentences, and the morphemes of each English sentence are analyzed.

In other words, it identifies the morphological and semantic features of each sentence, determines syntactic clusters, processes unknown words, and identifies proper nouns.
We perform various groupings such as ``unit'', which is a collection of words that are the smallest unit of the translation process, including expressions of time and number, and ``block J'', which is a collection of larger words, and create additional questions and appositives. It also processes entire sentences such as certification.

Morphological analysis rules used at that time are stored in the analysis rule file 23.

The morphologically analyzed English text data is transferred to the syntactic analysis unit 15 together with the dictionary information obtained through the morphological analysis.

This part 15 of the text is 1, for example, a context-free grammar (con
text-free gratrImar: cf
This is a functional unit that applies the rules of G-grammar to English data to analyze the surface structure of a single sentence and discover all syntactic possibilities. The English text data parsed here is sent to the elliptical sentence analysis unit 16 together with the analysis information.

Here, from the superficial analysis results by the cotton sentence analysis section 15, #! Apply the structural description to select a solution.

As a result, a reliable parse tree for English sentences is created.

Create that structure. The parsing rules are also stored in the parsing rule file 23.

The English data parsed by the parser 1 section 16 is transferred to the structure conversion section 17 as parse tree data.

Here, a Japanese syntax tree corresponding to a syntax tree, which is an intermediate structure of an English sentence, is created and a Japanese sentence is converted into a Japanese basic structure that is easy to translate.

The syntax tree data indicating the Japanese syntax tree whose structure has been converted in this way is sent to the translation generation unit 18, where a translation is generated. This is a function that generates Japanese sentences from the tree structure of Japanese syntax trees.

First, to match the word order to that of Japanese.

Syntax generation is performed to change the main structure by permuting the order, and then morpheme generation is performed to generate a translated sentence.

The translated Japanese text data, that is, the translated text data, is sent to the free editing/post-editing department on the 1st. Here, the translated text that is the result of translation by this device is freely modified or the translated text data is modified using the information used in the translation process to complete a more accurate and natural Japanese sentence.

This Japanese text data is transferred to the output device 20.

The translated Japanese sentence 21 is output.

This output device 20 includes, for example, the printer 5, display 4, etc. shown in FIG. and HDD7. Includes a file storage device such as FDD9.

The flow of these series of translation processes is controlled by a control unit 10 that controls the entire bag opening.

In this embodiment, the word dictionary 22 stores dictionary data for English and Japanese words, and the analysis rule file 23 stores rule data for morphological analysis and syntactic analysis.

Note that in FIG. 3, an operation display section 24 is connected to the control section 10. This operation display section 24 is.

The operator gives various instructions to this device.

For example, the operation unit has various operation keys such as translation instruction keys and cursor movement keys, and the display unit displays input English sentences, translated Japanese sentences, intermediate data such as dictionary information, and various instructions for the operator. It has a display to display.

However, as shown in FIG. 2, when the input device 11 is equipped with a keyboard, the operation display section 24 displays the keyboard 1, and when the output device 20 includes the display 4, the display 4 is displayed. can be used.

FIG. 1 is a block diagram showing only the portions related to the present invention in the above-mentioned English-Japanese automatic translation device.

To explain this, when the translation process is completed by the translation device main body 6, the data of the translated sentence that is the first translation result is stored in the translation file (memory) 25, and the data of the translated sentence that is the same translation result and the translation information for each sentence. and are respectively stored in the translation information file (memory) 26.

The free editing section 19A and the post-editing section 19B correspond to the free editing/post-editing section 1S in FIG.

The free m eastern part 19A is in a free editing mode in which the translation data, which is the translation result, is read out from the translation file 30, displayed in units of sentences on the display 4, and the translated text can be freely modified by inputting commands and character data from the keyboard 1. Process.

When post-editing is selected using the keyboard 1 in the free editing mode, the post-editing section 19B reads translation information regarding each sentence specified by the cursor from the translation information file 2S and displays it on the display 4, A post-editing mode process is performed in which the translated text can be corrected using the translation information.

The processing operations of the free editing section 19A and the post-editing section 19B will be further explained with reference to the flowchart of FIG. 4 and the display examples of FIGS. 5 and 6.

When the translation process by the translation device main body 6 is completed.

The routine of FIG. 4 starts, and in step 1 it is determined whether or not the post-editing call key has been pressed. If it is not pressed, the routine proceeds to step 2, and the translated text data as the translation result is obtained from the translated text file 25 of FIG. 1. After reading, the original text and translated text are displayed side by side on the display 4, sentence by sentence, as shown in FIG. 5, for example, and a free editing mode is entered.

In this state, it has functions similar to a word processor,
By operating the cursor movement keys on the keyboard 1, the operator can move the cursor 30 (Fig. 5) to any position of the characters displayed on the display 4, and freely correct, delete, or add characters using the keys manually. This allows you to freely modify the translation.6 Although only 4 sentences are displayed in Figure 5, in reality 1
Once you have finished editing the text displayed on the screen.

By scrolling or turning pages, you can display the next sentence in sequence and edit it freely.

Then, in step 3, it is determined whether or not the editing end key has been pressed, and if the editing is not completed, the process returns to step 1.If the gi collection is completed, the editing result is transferred to the translation file 25 in step 4.
(rewrites the translation result) and ends this routine.

Note that in the main routine, which is not shown.

The editing results stored in the translation file 25 are printed out using the printer 5 shown in FIG.

The HDD 7 or FDD performs processing such as registering the sentence f on the hard disk or floppy disk.

When the post-editing call key is pressed, the process proceeds from step 1 to step 5, where the number of the sentence specified by the cursor 30 is memorized, and in step 6, the immediately preceding post-editing result corresponding to that sentence is read from the translation file 25 and is displayed. Display on bracelet 4.

If no post-editing has been performed on the specified sentence, the translated text data read from the translated text file 25 remains immediately after machine translation.

For example, as shown in Figure 6, the translated text on display 4 is
This is the same as when starting free editing in the figure, but in order to clearly indicate which sentences can be post-edited, the sentence specified by the cursor is surrounded by a frame 31 as shown in the figure, or is displayed in reverse video.

Then, in step 7, it is determined whether the post-editing end key has been pressed, and in step 8, it is determined whether the key to return to the state at the beginning of post-editing has been pressed. If neither has been pressed, step 9 is performed. Enters post-editing mode where you can use the translation information to correct the translated text.

In the post-edit mode, translation information regarding a specified sentence (number stored) can be read from the translation information file 26 and displayed.

In FIG. 6, words in English and corresponding words in Japanese are underlined. Then, by instructing the display of another translation from keyboard 1.

Alternative words selected during the translation process (synonyms with different expressions 1, e.g. spring, spring for r spring J.

The user can display a selection number in a working area (not shown) of the display 4, and replace another translated word selected using the numeric keypad with the underlined translated word in the current translated sentence.

Furthermore, this specified translation can be replaced with any of the other translations stored in the translation information file.

In this way, post-editing is performed using the translation information, but when the post-editing end key is pressed, the process proceeds from step 7 to step 10, and the post-editing result of the sentence specified at that time is edited as shown in Figure 1. The translated text is stored in the translated text file 25 (the translated text of the translation result is rewritten) and the process returns to step 1.

Also, when the key to return to the initial state of post-editing is pressed, the process advances from step 8 to step 11, and the translation data of the first translation result corresponding to the specified sentence is read from the translation information file 26 in FIG. Then, the translated text is displayed in the frame 31 of the display 4, and the post-editing mode is continued.

Next, an example of correcting a translated text using this translated text editing method will be explained with reference to FIG.

First, when the translation process by this translation device is completed.

For example, as shown in Figure 7 (,), the translation of the English sentence rHe has one, J is ``He has one, J''.

” is displayed.

At this time, the operator enters free editing mode, so if the operator looks at this and determines that this translated sentence is incorrect, he or she can freely modify it as is, but if the operator specifies this sentence with the cursor and presses the post-edit key, , the immediately preceding post-editing result for this sentence (a first is the same as the translation result) is called, and the seventh
The screen is displayed as shown in Figure (b), and the post-editing mode is entered.

So, for example, if you want to correct the translated word ``1'', you can underline it, and as a result, the word ``roneJ'' in the original text that corresponds to that translated word will also be underlined, as shown in Figure 6). When you specify the display of ``roneJ'', other translations for ``roneJ'' will be displayed in the working area of the display, so if you select ``one'' from among them, the screen shown in Figure 7 (
As shown in c), "1" is corrected to "Hitotsu wo".

When the editing end key is pressed, the post-editing result shown in FIG. 7(c) is stored, and the free editing mode is entered with the translation displayed as shown in FIG. 7(d).

Therefore, the operator can further freely modify this translated text, for example, he can modify "Hitotsu wo" in the translated text shown in FIG. 7(d) to "Pen wo".

After that, if you wish to perform post-editing again, press the post-editing call key and the previous post-editing result, ``He has one,'' is displayed as shown in FIG. 7(f).

Therefore, this time, by underlining the word ``mochimasu'' and selecting another interpretation, it is possible to correct it to ``mochimasu'' as shown in figure (g).

Then, when the g collection end key is pressed, the editing results are stored and the display shown in FIG.

Therefore, we can freely modify this translation again and use the same figure (i).
It can be changed as shown.

In this way, you can arbitrarily select between free editing and post-editing to correct the translated text, and in post-editing mode, you can select between the previous post-editing result and the machine-translated translation result. can also be selected and used.

Furthermore, this invention is applicable to English-Japanese automatic translation devices.
Not limited to IQ collections (Japanese), but also translation devices that automatically or semi-automatically translate from other foreign languages to Japanese, or from Japanese to various foreign languages.

It can similarly be applied to editing a machine-translated text to make it more accurate and natural.

Example - Difficulty As explained above, by implementing the translated text editing method according to the present invention, it is possible to freely edit a translated text that has been machine translated by a translation device.

You can freely select post-editing to make corrections using the translation information used during translation, and even after free editing, you can call up the results of post-editing or return to the original machine-translated translation. This makes it easier to correct the translation,
You can efficiently edit the translation to make it accurate and natural.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the main parts of an embodiment of the present invention. FIG. 2 is a system configuration diagram of an English-Japanese automatic translation device implementing the present invention. FIG. 3 is a block diagram of the entire structure. Figure 4 shows the free editing section 19A and post-editing section 1911 in Figure 1.
Flowchart of processing operation by. FIG. 5 is an explanatory diagram showing an example of the display on the display in the free editing mode, and FIG. 6 is an explanatory diagram showing an example of the display on the display in the post-editing mode. FIG. 7 is an explanatory diagram showing an example of correcting a translated text according to the present invention. 1...Keyboard 2...Image scanner 3.
・・Optical character recognition device 4・・CRT display mount (display) 5・
...Printer 6...Main body 7...Hard disk unit @ (HDD) 8...Floppy disk device 7 (FDD) 11...Input device IS...Free editing/post-editing section 19A...Free Editorial Department 19B... Post-editing Department 2
5... Translation file 26... Translation information file Figure 1 M1'19 Figure 4 Figure 5 Figure 6 Figure 7

Claims (1)

[Scope of Claims] 1. When the translation process is completed by the translation device, the translation result is displayed in units of sentences and a free editing mode is set in which the translation can be freely modified. When post-editing is selected in this mode, the specified Ta1
The translation information for each sentence is also displayed, and the post-editing mode allows you to use that translation information to modify the translated sentence. When the post-editing is finished, the post-editing results for the sentence specified at that time are displayed. If the post-editing mode is memorized and the post-editing mode is selected again for the same sentence, the memorized post-editing result corresponding to that sentence is displayed and the mode is changed to the post-editing mode. A translated text editing method characterized in that when a starting state is selected, the translation result corresponding to the specified sentence is displayed to enable post-editing of the translated text.