JPS63300361A - Machine translation device - Google Patents

Abstract

PURPOSE:To evade unnecessary translation processing at the time of retranslating operation after original document correction or when only part of a long original document is translated partially by specifying a desired part to be translated in the input original document and displaying a partial translation result obtained by translating only the part. CONSTITUTION:When an operator specifies a desired part to be translated in the input original document, a translation part 4 translates the specified desired part to be translated and the translation result is displayed on a display part 7. For example, when part of the input original document after correction is translated after a 1st translation, the part corresponding to the last translation result in the input original document is displayed as a substitute for the partial translation result. In another way, the partial translation result is displayed on a display position specified for the last translation result with respect to the input original document as a substitute for the last translation result or displayed while added to the head or tail of the last translation result as to the input original document. Consequently, the time required for the translation is shortened as compared with the translation of the whole input original document.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a machine translation device, and particularly to a machine translation device having a partial translation function.

(Prior Art) Machine translation devices are being developed that utilize computer technology to automatically convert an original input text in a first language input from a keyboard or a file into a translated text in a second language. This machine translation device basically divides the input original text into predetermined processing units consisting of words or phrases by morphological analysis, syntactic analysis, etc., and searches a translation dictionary for each processing unit to find the corresponding translation or A translated text is obtained by finding translated words and phrases and combining them according to predetermined translation rules.

When using a machine translation device in this manner, after translating the entire input original text, the operator often corrects spelling mistakes and typographical errors in the input original text before retranslating.

For example, when the suitability of the initially obtained translation is judged and a mistranslation or an inappropriate translation is discovered, a part of the input original text is revised to obtain a more appropriate translation.

Conventionally, when an input original text is modified in this way and then translated again, the entire original text is translated. However, when a part of the original text is modified, it is possible to translate the original text without necessarily referring to the entire original text, so retranslating the entire original text is a big waste of time.

In addition, with conventional machine translation devices, even when only a part of the input original text is to be translated, unnecessary parts are deleted after translating the entire original text.

(Problem to be solved by the invention) In this way, with conventional machine translation devices, when retranslating the original text after correcting it, the entire original text is translated, which is inefficient, and it is difficult to obtain the final translated text. There is a problem in that it takes a long time to complete the process, and even when it is desired to translate a part of the input original text, the whole process is translated and unnecessary parts are deleted, which is a wasteful process.

SUMMARY OF THE INVENTION An object of the present invention is to provide a machine translation device that enables efficient translation by selectively translating only necessary parts when retranslating an original text after it has been corrected.

[Structure of the Invention] (Means for Solving the Problems) The present invention provides an input unit, an original text storage unit that stores an input original text to be translated from the input unit or an original text file to be translated, and a translation process. It is equipped with a translation dictionary section that stores information on various dictionaries including the word dictionary to be used, and a translation section that uses the information in the translation dictionary section to translate the human-powered original text read from the original text storage section and output the translated text. In a machine translation device, the input section allows the operator to specify the desired part of the input original text to be translated, and the translation section, when the input section specifies the desired part for translation, translates only this part and translates the resulting part. The present invention is characterized in that it has a function of obtaining a translation result, and that the display unit further has a function of displaying this partial translation result.

(Operation) When the operator specifies the part of the input original text that he or she wishes to translate after correcting the original text after the first translation is complete, or after the first manual translation of the original text, the translation department will translate only the specified part of the original text that he or she wishes to translate. Then, the translation result is displayed on the display section. For example, if a partial translation is performed on the corrected input original text after the first translation, the corresponding part of the previous translation result for the input original text will be replaced with the partial translation result, or the input original text will be displayed as a partial translation result. The partial translation result is displayed in place of the previous translation result at the display position specified for the previous translation result, or it is added and displayed at the beginning or end of the previous translation result for the input original text.

In this way, since it is possible for the operator to selectively translate only the desired part of the input original text, the time required for translation is reduced compared to the case where the entire original text is translated.

(Embodiment) FIG. 1 is a block diagram showing the overall configuration of an English-Japanese machine translation device according to an embodiment of the present invention.

This device includes a human input section 1 consisting of a keyboard, an original text storage section 2 that stores the English text as the first language input input through the input section 1, and a translation section 2 that stores information from various dictionaries used in translation processing. A dictionary unit 3, a translation unit 4 that uses information in the translation dictionary unit 3 to translate an input original English sentence into a second language Shiraki sentence, and a Japanese sentence as a translated sentence obtained by the translation unit 4. a display control section 6 and a display section 7 for displaying the original text and the translated text, a printing section 8 for transcribing the original text and the translated text and outputting them as hard copies, and an editing control section for controlling the entire system. It consists of part 9.

The translation dictionary section 3 is a conjugation/inflection dictionary 3a that converts words with changes in their endings etc. into their original forms (basic forms).

A word/idiom dictionary 3b in which translations in the second language corresponding to words/idioms in the first language (English) are stored together with part-of-speech information.
, an analysis grammar dictionary 3c that stores correct sequences of parts of speech, and a conversion grammar dictionary 3 that stores conversion information from English sentences to Japanese sentences.
d. Generative grammar dictionary 3 that determines the structure of converted Japanese
e, and a morpheme generating grammar dictionary 3f that completes a translated sentence by changing Japanese word forms such as endings.

FIG. 2 shows an example of the key arrangement of the input unit 1, in which character keys 11. Translation instruction key 129 Edit key 13-1
89 function keys 19-26. It is provided with cursor keys 27 to 30 and other keys. The character key 11 is for inputting an English sentence as an input original text, and the translation instruction key 12 is a key for giving an instruction to start translation processing after completing the English text.

Editing keys include the insertion key 13, which inserts a character before the cursor position; Delete key 14. Delete the character string in the range indicated by the cursor. Movement keys for moving the character string in the range indicated by the cursor 15. Cancel key 16 to nullify the effects of keys 13-15. A key 17 is provided for modifying other modifications of the word indicated by the cursor using auxiliary information.

As a function key, there is a translated word display key 19 which gives an instruction to display the translated word for a word in the translated sentence. Dictionary display key 20 for giving dictionary display instructions for words in the input original text. Dictionary registration key 21 for registering new words and phrases. Dictionary deletion key 22 for deleting words and phrases registered through dictionary registration. Partial translation key 2 to display the partial translation when translation fails
In addition to 3, functions related to the gist of the present invention are controlled by a translation part specification key 249 for specifying a desired part of the input original text to be translated, a partial translation result display position specification key 25, a partial translation result display request key 26, etc. It is equipped with

Cursor keys include cursor movement keys 27 for moving the cursor in up, down, left and right directions. A unit switching key 28 for switching the unit in which the cursor moves; an inter-area movement key 29 for moving the cursor between display areas of the display section 7; It is provided with an enlargement/reduction key 30 for reducing the size of the cursor on a character-by-character basis or enlarging it on a word-by-word basis.

FIG. 3 shows an example of the layout on the screen of the display section 7. As shown in FIG. The input original text is displayed in the original text display area on the left side of the screen, and the translated text that is the translation result is displayed in the translated text display area on the right side of the screen so as to correspond to the input original text. Furthermore, the editing area at the top of the screen is used to display information necessary for various edits.

FIG. 4 shows the editing control section 9 using the input section 1 and display section 7.
This figure shows the flow of processing when performing dialogue translation processing. In interactive translation processing, an operator can appropriately edit a translated text corresponding to an input original text while interacting with the device. In steps 31 to 35, the editing control unit 9 monitors whether there is any key input from the input unit 1 or whether a translation completion signal is received from the translation unit 4, and if there is any key input, it takes appropriate action. Perform the following processing.

When the operator operates any of the function keys 19 to 26, the editing control unit 9 detects this in step 35, and performs the process corresponding to that function key in step 42.

When the operator operates any of the editing keys 13 to 18, the editing control unit 9 detects this in step 34, and performs processing corresponding to the editing key in step 41.

When the operator operates the cursor keys 27 to 30 or other control keys, the editing control section 9 performs processing for moving the cursor corresponding to the key operation and other processing in step 33.

When the operator inputs the original text by operating the character keys 11, the character keys are detected in step 33, and then the corresponding character codes are sequentially set in an input buffer (not shown) in the editing control unit 9 (step 38). This input original text is displayed in the original text display area of the display section 7 via the display control section 6 (step 39). After the operator enters the original text,
When the translation instruction key 12 is pressed, this is detected in step 32, and the editing control unit 9 supplies the input original text in the input buffer to the translation unit 4, and instructs the translation unit 4 to start translation processing (step 40). ). If input editing such as correction, insertion, deletion, etc. is required while inputting the original text, the operator moves the cursor to the desired editing location using the cursor keys, and then presses the insert key 13. Editing operations can be performed using editing keys such as the delete key 14.

FIG. 5 is a flowchart showing the flow of translation processing in the translation section 4. First, when the translation unit 4 is given an input English sentence, it converts words with changes in endings, etc. into original forms (basic forms) using the inflection/inflection dictionary 3a, and performs morphological analysis (step 51).

Next, the translation of the original word broken down into each word by morphological analysis is added to the word/idiom dictionary 3 along with its part-of-speech information.
Search from b (step 52).

When the translation is searched, control is performed in step 53 to subject the translation candidates to syntactic analysis one by one, and then the translation candidates are syntactically analyzed using the parsing grammar dictionary fi3c to generate an English sentence structure. (step 54). Next, the success or failure of the parsing is determined in step 55,
If the analysis fails, return to analysis control step 53,
If the analysis is successful, the process proceeds to step 56, where the English sentence structure is converted into the Japanese sentence structure using the conversion grammar dictionary 3d. It is determined in step 57 whether or not the structure of the Japanese sentence has been rejected in this structure conversion step 56. If rejected, the process returns to the parsing step 54. Furthermore, if it is determined that the structure of the Japanese sentence is correct, the Japanese word order is determined from the Japanese structure based on the generative grammar dictionary 3e, and a syntax generation process is performed to convert it into a word string.

Next, in step 59, the morpheme generation grammar dictionary 3f is used to change the word forms, etc. to complete the translated sentence of the Japanese sentence.

The translation section 4 sends a signal to the editing control section 9 each time it starts processing steps 51 to 59 in FIG. .

Returning to the flowchart of FIG. 4, when the editing control unit 9 receives the translation completion signal from the editing unit 4 in step 31, the process proceeds to step 37, where the translation candidates and auxiliary information obtained from the translation unit 4 are displayed on the display control unit. 6 and display the translation.

The display control unit 6 displays the translation candidate in the translation display area in FIG. 3 at a position corresponding to the input original text, and
If there is auxiliary information, the translated word is displayed with high brightness or reversed so that the operator can identify it.

When the editing keys 13 to 18 are pressed in this state, the editing control unit 9 detects this in step 34 and performs editing processing corresponding to each key on the word or phrase indicated by the cursor at that time. (Step 41).

Next, a processing procedure will be described in which the entire input original text is translated once, the human original text is partially corrected, and then the translation is performed again. 6 to 8 are flowcharts showing different specific examples of this processing procedure.

In FIG. 6, when translation completion is detected in step 31 of FIG. 4, it is determined in step 61 whether the translation is the first translation of the entire input original text.

When it is determined that the first translation has been completed, the process moves to step 62 and step 33.3 in FIG.
4, it is determined whether the input original text has been modified. When it is determined that the input original text has been corrected, it is then determined in step 63 whether or not a part of the input original text has been designated by the translation portion designation key 24. If the result of this determination is negative, the process proceeds to step 67 by pressing the translation instruction key 12 and the entire original text is translated; however, if a part of the input original text is specified, the process proceeds to step 64 by pressing the translation instruction key 12. The process proceeds and the specified portion is translated.

FIG. 9 shows an example of an input original text and a translated text obtained by the first translation. For example, the operator judges that the translation of "process the machine" in the translated text is inappropriate, and as shown in FIG.
processing-ma
For example, modify this part to “proc
If a desired part is specified for translation by an underline, such as ``esslng machines'', the resulting partial translation result will be ``processing machines''.

When the partial translation result is obtained in this way, the display position of the partial translation result is detected in step 65. To detect the display position of this partial translation result, for example, an original/translation word correspondence table as shown in FIG. 11 is used. This table shows the correspondence of each word between the original text and the translated text, and by using this table, the previous translation result of the input original text, that is, the translation specified in step 63 in the translated text shown in FIG. The portion corresponding to the desired portion can be determined as the partial translation display position. For example, the part of the translated text corresponding to the desired translation part "processing machines" is "processing machines", and this part is detected as the partial translation display position.

Next, by pressing the partial translation result display request key 26, the process proceeds to step 66, where the corresponding part of the first translation result is displayed at the partial translation result display position detected as described above.
The partial translation result ``processing machine'' is replaced with ``processing machine'' as shown in FIG. 10.

In FIG. 7, steps 71-74.76°77 are the same as steps 61-64, 66.67 in FIG.
When the partial translation result is obtained in step 74, step 7
5, it is determined whether the partial translation result display position has been designated by the partial translation result display position designation key 25. When the display position is specified, the partial translation result is displayed at the specified display position in step 76 in a replaced manner.

FIG. 12 shows an example in which the part "processing machine" is designated as the display position of the partial translation result, and the partial translation result "processing machine" is displayed at this position.

In FIG. 8, steps 81-84, 86°87 are the same as steps 61-64, 66, 67 in FIG.
When the partial translation result is obtained in step 84, step 8
Proceeding to step 5, the partial translation result ``processing machine'' is displayed as being added to the beginning or end of the sentence of the first translation result, as shown in FIG. By moving the display position of "processing machine" with the cursor, the operator will display the part "processing machine" replaced with "processing machine" in step 86.

Note that the present invention is not limited to the above-described embodiments; for example, although an example of translation from an English sentence to a Japanese sentence has been described in the embodiment, the language pair of the original text and the translated text is not limited to this. Further, in the embodiment, a case has been described in which the original text is corrected after the first translation and then partial translation is performed, but for example, after reading the entire input original text from the original text file, only the necessary parts may be translated. The present invention is also effective when completing the entire translated text by combining or moving partial translations.

Furthermore, the present invention can be applied to a case where only a designated portion is summarized in a summarization system. Furthermore, when a partial translation result is obtained, the entire previous translation result for the input original text may be replaced with the partial translation result and displayed.

[Effects of the Invention] According to the present invention, by having a function of specifying a desired part of an input original text to be translated and displaying a partial translation result obtained by translating only this part, it is possible to easily perform re-translation after correcting the original text. When it is desired to partially translate only a part of a long original text, unnecessary translation processing can be avoided, and an efficient machine translation device can be provided.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of a machine translation device according to an embodiment of the present invention, FIG. 2 is a diagram showing the key arrangement of the input section in the same embodiment, and FIG. 3 is a diagram showing the layout of the display section in the same embodiment. FIG. 4 is a flowchart showing the flow of processing by the editing control section in the same embodiment; FIG. 5 is a flowchart showing the flow of translation processing by the translation section in the embodiment; FIGS. Figure 8 is a flowchart 1. showing the flow of partial translation processing in the same embodiment. Figure 9 is a diagram showing a general translation example, Figure 10 is a diagram showing a partial translation example after original text correction based on the process shown in Figure 6, and Figure 11 is a diagram showing original and translated words used in the same example. Diagram showing the correspondence table, 1st
2 is a diagram showing a display example of a partial translation result based on the process of FIG. 7, and FIG. 13 is a diagram showing a display example of a partial translation result based on the process of FIG. 8. DESCRIPTION OF SYMBOLS 1... Input part, 2... Original text storage part, 3... Translation dictionary part, 4... Translation part, 5... Translation text storage part, 6...
Display control section. 7...Display section, 8...Print section, 9...Edit control section. Applicant's Representative Patent Attorney Shikihiko Suzushi Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11

Claims (5)

[Claims] (1) An input section, an original text storage section that stores the input source text to be translated from this input section or the source file for translation, and a translation dictionary section that stores information on various dictionaries including word dictionaries used for translation processing. and a translation unit that uses information in the translation dictionary unit to translate the input original text read from the original text storage unit and output a translated text, wherein the input unit is configured to translate the input original text. The translation section has the function of specifying the desired part, and when the desired part is specified by the input section, the translation section has the function of translating only this section and obtaining a partial translation result, and the display section has the function of obtaining a partial translation result obtained by translating only this section. A machine translation device characterized by having a function of displaying translation results. (2) The display unit has a function of, when a partial translation result is obtained, replacing the corresponding part of the previous translation result of the input original text with the partial translation result and displaying the result. The machine translation device according to item 1. (3) The display unit shall have a function that, when a partial translation result is obtained, displays the partial translation result in place of the previous translation result at the display position specified for the previous translation result for the input original text. A machine translation device according to claim 1, characterized in that: (4) The display unit has a function of displaying a partial translation result added to the beginning or end of a previous translation result of the input original text when a partial translation result is obtained. machine translation device. (5) The display unit has a function of, when a partial translation result is obtained, replacing the entire previous translation result of the input original text with the partial translation result and displaying the result. machine translation device.