Translation device and translation method

Abstract

A translation device (1) is configured so as to comprise: an original text acquisition unit (11) that acquires an original text to be translated; a first translation unit (12) that causes the original text acquired by the original text acquisition unit (11) to be translated into a first language; a back translation unit (13) that causes the translated text by the first translation unit (12) to be translated back; a display processing unit (16) that causes a display device (3) to display the translated-back text by the back translation unit (13); and a correction reception unit (14) that receives correction of the original text after the translated-back text is displayed by the display processing unit (16). The translation device (1) also comprises a second translation unit (15) that, when the correction reception unit (14) receives the correction of the original text, causes the corrected original text received by the correction reception unit (14) to be translated into a second language.

Description

Translation device and translation method

[0001] The present disclosure relates to a translation device and a translation method.

[0002] There is a translation device that translates an original text to be translated. For example, Patent Document 1 discloses a translation device that includes a processing control unit that translates an original text in a source language into a first language, displays the translation into the first language on a screen, and translates the translation into the first language back into the source language and displays the back-translation into the source language on a screen.

[0003] Japanese Patent Application Laid-Open No. 2001-318918

[0004] The translation device disclosed in Patent Document 1 is only intended to translate an original text written in a source language into a first language. Therefore, in order to translate an original text written in a source language into multiple languages, it is necessary to use multiple translation devices as disclosed in Patent Document 1. When using multiple translation devices as disclosed in Patent Document 1 to translate an original text written in a source language into multiple languages, if the back-translated text differs from the user's intention, the user must correct the original text in each of the multiple languages ​​to be translated. This means that the user has to perform the same correction work multiple times, which takes a long time to translate into multiple languages.

[0005] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a translation device that can translate an original text to be translated into multiple languages ​​and reduce the time required to translate into multiple languages.

[0006] A translation device according to the present disclosure includes an original text acquisition unit that acquires an original text to be translated, a first translation unit that translates the original text acquired by the original text acquisition unit into a first language, a back translation unit that back-translates the translation by the first translation unit, a display processing unit that displays the back-translated text by the back translation unit on a display device, and a correction accepting unit that accepts corrections to the original text after the back-translation is displayed by the display processing unit. The translation device also includes a second translation unit that, when the corrections to the original text are accepted by the correction accepting unit, translates the corrected original text accepted by the correction accepting unit into a second language.

[0007] According to the present disclosure, an original text to be translated can be translated into multiple languages, and the time required for translation into multiple languages ​​can be reduced.

[0008] 1 is a configuration diagram showing a translation system including a translation device 1 according to a first embodiment. FIG. 2 is a hardware configuration diagram showing the hardware of the translation device 1 according to the first embodiment. FIG. 3 is a hardware configuration diagram of a computer in which the translation device 1 is realized by software, firmware, or the like. FIG. 4 is a flowchart showing a translation method, which is a processing procedure of the translation device 1. FIG. 5 is a configuration diagram showing a translation system including a translation device 1 according to a second embodiment. FIG. 6 is a hardware configuration diagram showing the hardware of the translation device 1 according to the second embodiment. FIG. 7 is a configuration diagram showing a translation system including a translation device 1 according to a third embodiment. FIG. 8 is a hardware configuration diagram showing the hardware of the translation device 1 according to the third embodiment. FIG. 9 is an explanatory diagram showing an example of technical terms, definitions of technical terms, how to read technical terms, and translations of technical terms in a first language, which are recorded in a technical terminology dictionary. FIG. 10 is a configuration diagram showing a translation system including a translation device 1 according to a fourth embodiment. FIG. 11 is a hardware configuration diagram showing the hardware of the translation device 1 according to the fourth embodiment. FIG. 12 is an explanatory diagram showing an example of an image in which a target person is shown. FIG. 13 is a configuration diagram showing a translation system including a translation device 1 according to a fifth embodiment. FIG. 14 is a hardware configuration diagram showing the hardware of the translation device 1 according to the fifth embodiment. FIG. 15 is a flowchart showing a translation method, which is a processing procedure of the translation device 1.

[0009] In order to explain the present disclosure in more detail, embodiments of the present disclosure will be described below with reference to the accompanying drawings.

[0010] Embodiment 1. Fig. 1 is a configuration diagram showing a translation system including a translation device 1 according to embodiment 1. Fig. 2 is a hardware configuration diagram showing the hardware of translation device 1 according to embodiment 1. The translation system shown in Fig. 1 includes translation device 1, translation engine 2, and display device 3. Translation device 1 provides an original text to be translated to translation engine 2, obtains translations in multiple languages ​​from translation engine 2, and displays each translation on display device 3. When the original text to be translated is provided from translation device 1, translation engine 2 translates the original text into each language and outputs the translations in each language to translation device 1. Display device 3 displays the translations in each language, etc.

[0011] The translation device 1 includes an original text acquisition unit 11, a first translation unit 12, a back translation unit 13, a correction acceptance unit 14, a second translation unit 15, and a display processing unit 16. The original text acquisition unit 11 is realized, for example, by an original text acquisition circuit 21 shown in FIG. 2. The original text acquisition unit 11 acquires an original text to be translated. The original text acquisition unit 11 outputs the original text to be translated to the first translation unit 12, the correction acceptance unit 14, the second translation unit 15, and the display processing unit 16, respectively.

[0012] The first translation unit 12 is realized, for example, by the first translation circuit 22 shown in FIG. 2 . The first translation unit 12 acquires an original text to be translated from the original text acquisition unit 11. The first translation unit 12 translates the original text acquired by the original text acquisition unit 11 into a first language. Specifically, the first translation unit 12 provides the original text to be translated to the translation engine 2 and acquires a translation in the first language from the translation engine 2. In the translation system shown in FIG. 1 , the translation engine 2 is provided outside the translation device 1. However, this is merely an example, and the translation engine 2 may be provided inside the translation device 1. For example, if the first translation unit 12 includes a translation engine, the first translation unit 12 translates the original text to be translated into the first language. The first translation unit 12 outputs the translation in the first language to the back translation unit 13 and the display processing unit 16.

[0013] The back translation unit 13 is realized, for example, by the back translation circuit 23 shown in FIG. 2 . The back translation unit 13 obtains a translation in the first language from the first translation unit 12. The back translation unit 13 back-translates the translation by the first translation unit 12. Back translation is a process of translating the translation by the first translation unit 12 into the language used in the original text to be translated. Specifically, the back translation unit 13 provides the translation in the first language to the translation engine 2 and obtains the back-translated text from the translation engine 2. In the translation system shown in FIG. 1 , the translation engine 2 is provided outside the translation device 1. However, this is merely an example, and the translation engine 2 may be provided inside the translation device 1. For example, if the back translation unit 13 includes a translation engine, the back translation unit 13 back-translates the translation in the first language. The back translation unit 13 outputs the back-translated text to the display processing unit 16.

[0014] The correction accepting unit 14 is realized by, for example, the correction accepting circuit 24 shown in Fig. 2. The correction accepting unit 14 acquires the original text to be translated from the original text acquiring unit 11. The correction accepting unit 14 accepts corrections to the original text after the back-translated text is displayed by the display processing unit 16. The correction accepting unit 14 outputs the accepted corrected original text to each of the first translation unit 12 and the second translation unit 15.

[0015] The second translation unit 15 is realized by, for example, the second translation circuit 25 shown in FIG. 2 . If the correction receiving unit 14 has accepted a correction, the second translation unit 15 acquires the corrected original text from the correction receiving unit 14. If the correction receiving unit 14 has not accepted a correction, the second translation unit 15 acquires the Japanese original text from the original text acquisition unit 11. If the correction receiving unit 14 has accepted a correction, the second translation unit 15 translates the corrected original text accepted by the correction receiving unit 14 into the second language. If the correction receiving unit 14 has not accepted a correction, the second translation unit 15 translates the original text acquired by the original text acquisition unit 11 into the second language. Specifically, the second translation unit 15 provides either the corrected original text accepted by the correction receiving unit 14 or the original text acquired by the original text acquisition unit 11 to the translation engine 2, and acquires a translation in the second language from the translation engine 2. 1 , the translation engine 2 is provided outside the translation device 1. However, this is merely an example, and the translation engine 2 may be provided inside the translation device 1. For example, if the second translation unit 15 is provided with a translation engine, the second translation unit 15 translates either the original text corrected by the correction receiving unit 14 or the original text acquired by the original text acquisition unit 11 into the second language. The second translation unit 15 outputs the translation in the second language to the display processing unit 16.

[0016] The display processing unit 16 is realized by, for example, the display processing circuit 26 shown in Fig. 2. The display processing unit 16 displays the original text to be translated output from the original text acquisition unit 11 on the display device 3, and displays the translation in the first language output from the first translation unit 12 on the display device 3. The display processing unit 16 also displays the back-translation output from the back translation unit 13 on the display device 3, and displays the translation in the second language output from the second translation unit 15 on the display device 3.

[0017] 1, it is assumed that each of the components of the translation device 1, namely, the source text acquisition unit 11, the first translation unit 12, the back translation unit 13, the correction acceptance unit 14, the second translation unit 15, and the display processing unit 16, is realized by dedicated hardware as shown in Fig. 2. That is, it is assumed that the translation device 1 is realized by a source text acquisition circuit 21, a first translation circuit 22, a back translation circuit 23, a correction acceptance circuit 24, a second translation circuit 25, and a display processing circuit 26. Each of the original text acquisition circuit 21, the first translation circuit 22, the back translation circuit 23, the correction reception circuit 24, the second translation circuit 25, and the display processing circuit 26 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof.

[0018] The components of translation device 1 are not limited to those realized by dedicated hardware, and translation device 1 may be realized by software, firmware, or a combination of software and firmware. Software or firmware is stored as a program in the memory of a computer. A computer refers to hardware that executes a program, such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer, processor, or DSP (Digital Signal Processor).

[0019] 3 is a hardware configuration diagram of a computer when translation device 1 is realized by software, firmware, etc. When translation device 1 is realized by software, firmware, etc., programs for causing a computer to execute the respective processing procedures of original text acquisition unit 11, first translation unit 12, back translation unit 13, correction acceptance unit 14, second translation unit 15, and display processing unit 16 are stored in memory 31. Then, a processor 32 of the computer executes the programs stored in memory 31.

[0020] 2 shows an example in which each of the components of translation device 1 is realized by dedicated hardware, while Fig. 3 shows an example in which translation device 1 is realized by software, firmware, etc. However, this is merely an example, and some of the components in translation device 1 may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0021] Next, the operation of the translation system shown in Fig. 1 will be described. Fig. 4 is a flowchart showing a translation method, which is a processing procedure of the translation device 1. For the sake of convenience, in the translation system shown in Fig. 1, it is assumed that the original text to be translated is Japanese, the first language is English, and the second language is Portuguese.

[0022] The original text acquisition unit 11 acquires a Japanese original text as the original text to be translated (step ST1 in FIG. 4 ). For ease of explanation, it is assumed here that the acquired Japanese original text is "After fastening the fastener again, please turn it to check." The original text acquisition unit 11 outputs the original text to be translated to each of the first translation unit 12, the correction acceptance unit 14, the second translation unit 15, and the display processing unit 16. The display processing unit 16 displays the original text to be translated output from the original text acquisition unit 11 on the display device 3.

[0023] The first translation unit 12 acquires the original Japanese text from the original text acquisition unit 11. The first translation unit 12 translates the original Japanese text acquired by the original text acquisition unit 11 into English, which is the first language (step ST2 in FIG. 4 ). Specifically, the first translation unit 12 provides the original Japanese text to the translation engine 2 and acquires an English translation from the translation engine 2. The first translation unit 12 outputs the English translation to both the back translation unit 13 and the display processing unit 16. The display processing unit 16 displays the English translation output from the first translation unit 12 on the display device 3.

[0024] The back translation unit 13 obtains the English translation from the first translation unit 12. The back translation unit 13 back-translates the translation by the first translation unit 12 (step ST3 in FIG. 4 ). Specifically, the back translation unit 13 provides the English translation to the translation engine 2 and obtains the back-translated text from the translation engine 2. The back translation unit 13 outputs the back-translated text to the display processing unit 16. When the English translation of the original text is provided to the translation engine 2, the translation engine 2 may output, for example, "Please reattach the fastener and turn it to check." This back-translated text may be interpreted as implying that the fastener is being reattached, which differs from the nuance of the original text. In other words, the original text needs to be corrected and translated again. The display processing unit 16 displays the back-translated text output from the correction receiving unit 14 on the display device 3 (step ST4 in FIG. 4 ).

[0025] The correction receiving unit 14 acquires the original Japanese text from the original text acquisition unit 11. The correction receiving unit 14 is realized by, for example, a man-machine interface, and if a user who has checked the back-translated text recognizes a difference in nuance between the original text and the back-translated text, the user can correct the original Japanese text by operating the correction receiving unit 14.

[0026] When the user performs an operation to correct the original Japanese text (step ST5 in FIG. 4 : YES), the correction receiving unit 14 accepts the correction of the original text (step ST6 in FIG. 4 ) and outputs the corrected original text to the first translation unit 12. Here, for the sake of convenience, it is assumed that the original text has been corrected to "Please make sure to turn the fastener again after installing it and check it thoroughly." If the user does not perform an operation to correct the original Japanese text (step ST5 in FIG. 4 : NO), the correction receiving unit 14 does not accept the correction of the original text.

[0027] When the corrected original text is output from the correction receiving unit 14, the first translation unit 12 acquires the corrected original text. The first translation unit 12 provides the corrected original text to the translation engine 2 and acquires an English translation from the translation engine 2. The first translation unit 12 outputs the English translation to both the back translation unit 13 and the display processing unit 16.

[0028] The back translation unit 13 obtains the English translation from the first translation unit 12. The back translation unit 13 provides the English translation to the translation engine 2 and obtains the back-translated text from the translation engine 2. The back translation unit 13 outputs the back-translated text to the display processing unit 16. When the English translation of the corrected original text is provided to the translation engine 2, the translation engine 2 may output a back-translated text such as, "After re-installing, be sure to turn the fastener again and check thoroughly." The nuance of this back-translated text matches the nuance of the original text, so there is no need to re-correct the original text. The display processing unit 16 displays the back-translated text output from the correction receiving unit 14 on the display device 3 (step ST4 in FIG. 4 ).

[0029] If the user, after checking the back-translated text, recognizes the difference in nuance between the corrected original text and the back-translated text, the user can correct the Japanese original text by operating the correction accepting unit 14. If the user performs an operation to further correct the corrected original text (step ST5 in FIG. 4: YES), the correction accepting unit 14 accepts the correction of the original text (step ST6 in FIG. 4) and outputs the corrected original text to the first translation unit 12. If the user does not perform an operation to further correct the corrected original text (step ST5 in FIG. 4: NO), the correction accepting unit 14 does not accept the correction of the original text. If the original text has been corrected one or more times, the correction accepting unit 14 outputs the most recently corrected original text to the second translation unit 15.

[0030] The second translation unit 15 acquires either the original text last corrected by the correction receiving unit 14 or the original Japanese text acquired by the original text acquisition unit 11. The second translation unit 15 translates either of the acquired original texts into Portuguese, which is the second language (step ST7 in FIG. 4 ). Specifically, the second translation unit 15 provides either of the acquired original texts to the translation engine 2 and acquires a translation into Portuguese from the translation engine 2. The second translation unit 15 outputs the translation in the second language to the display processing unit 16.

[0031] When the Japanese original text acquired by the original text acquisition unit 11 is given to the translation engine 2, the translation engine 2 may output, for example, a Portuguese translation such as "I repeat, after installation, please be sure to undo the zipper and check." The nuance of this Portuguese translation matches the nuance of the original text, so there is no need to re-edit the original text.

[0032] The display processing unit 16 causes the Portuguese translation output from the second translation unit 15 to be displayed on the display device 3 .

[0033] In the first embodiment described above, translation device 1 is configured to include original text acquisition unit 11 that acquires an original text to be translated, first translation unit 12 that translates the original text acquired by original text acquisition unit 11 into a first language, back translation unit 13 that back-translates the translation by first translation unit 12, display processing unit 16 that displays the back-translation by back translation unit 13 on display device 3, and correction receiving unit 14 that accepts corrections to the original text after the back-translation is displayed by display processing unit 16. Translation device 1 also includes second translation unit 15 that, when corrections to the original text are accepted by correction accepting unit 14, translates the corrected original text accepted by correction accepting unit 14 into a second language. Thus, translation device 1 can translate an original text to be translated into multiple languages ​​and can reduce the time required for translation into multiple languages. The reason why the time required for translation can be reduced in this way is that after the display processing unit 16 displays the translated text, the correction receiving unit 14 receives corrections to the original text, and the second translation unit 15 translates the corrected original text received by the correction receiving unit 14 into the second language, thereby enabling the translation into the second language to reflect the results of the translation into the first language. For example, if the original text is written in the user's native language, the original text may contain a mixture of colloquial and written expressions, resulting in a sentence that is understandable to a native speaker but has an ambiguous grammatical structure. For this reason, it is generally considered difficult to translate the original text into a correct translation in one translation run, and the original text may need to be corrected. Since translation device 1 translates an original text into a first language, displays the back-translated text, allows the user to confirm and correct the original text, and then translates the text into a second language using the corrected original text, translation into the second language can be performed using the corrected original text that is already easy to translate, making it possible to obtain a translation with higher translation accuracy in a shorter time than before. Also, even if the similarity between the grammar in the first language and the grammar in the second language is higher than the similarity between the grammar in the language used in the original text and the grammar in the first language, translation device 1 translates the original text into the second language using the back-translated text that the user has confirmed and corrected, making it possible to obtain a translation with higher translation accuracy in a shorter time than before.

[0034] 1, the back translation unit 13 back-translates the English translation by the first translation unit 12. The back translation unit 13 may also back-translate the Portuguese translation by the second translation unit 15. In this case, the display processing unit 16 may display the back-translated Portuguese translation on the display device 3, and then the correction receiving unit 14 may receive corrections to the original text.

[0035] Second Embodiment In a second embodiment, a translation device 1 will be described in which a second translation unit 17 translates either the corrected original text received by the correction receiving unit 14 or the original text acquired by the original text acquisition unit 11 into one or more languages ​​different from each of the first language and the second language.

[0036] FIG. 5 is a configuration diagram showing a translation system including a translation device 1 according to embodiment 2. In FIG. 5, the same reference numerals as those in FIG. 1 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted. FIG. 6 is a hardware configuration diagram showing the hardware of the translation device 1 according to embodiment 2. In FIG. 6, the same reference numerals as those in FIG. 2 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted. The translation system shown in FIG. 5 includes a translation device 1, a translation engine 2, and a display device 3. The translation device 1 shown in FIG. 5 includes an original text acquisition unit 11, a first translation unit 12, a back translation unit 13, a correction acceptance unit 14, a second translation unit 17, and a display processing unit 16.

[0037] The second translation unit 17 is realized by, for example, the second translation circuit 27 shown in FIG. 6 . If the correction receiving unit 14 has accepted a correction, the second translation unit 17 acquires the corrected original text from the correction receiving unit 14. If the correction receiving unit 14 has not accepted a correction, the second translation unit 17 acquires the Japanese original text from the original text acquisition unit 11. If the correction receiving unit 14 has accepted a correction, the second translation unit 17 translates the corrected original text accepted by the correction receiving unit 14 into the second language and also translates the corrected original text into one or more languages ​​different from each of the first language and the second language. If the correction receiving unit 14 has not accepted a correction, the second translation unit 17 translates the original text acquired by the original text acquisition unit 11 into the second language and also translates the original text into one or more languages ​​different from each of the first language and the second language. Specifically, the second translation unit 17 provides the translation engine 2 with either the corrected original text received by the correction receiving unit 14 or the original text acquired by the original text acquisition unit 11, and acquires a translation in the second language and translations in one or more languages ​​different from each of the first language and the second language from the translation engine 2. The second translation unit 17 outputs the translation in the second language and the translations in one or more languages ​​different from each of the first language and the second language to the display processing unit 16.

[0038] 5, it is assumed that each of the components of the translation device 1, namely, the original text acquisition unit 11, the first translation unit 12, the back translation unit 13, the correction acceptance unit 14, the second translation unit 17, and the display processing unit 16, is realized by dedicated hardware as shown in FIG. 6. That is, it is assumed that the translation device 1 is realized by an original text acquisition circuit 21, a first translation circuit 22, a back translation circuit 23, a correction acceptance circuit 24, a second translation circuit 27, and a display processing circuit 26. Each of the original text acquisition circuit 21, the first translation circuit 22, the back translation circuit 23, the correction acceptance circuit 24, the second translation circuit 27, and the display processing circuit 26 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

[0039] The components of translation device 1 are not limited to those realized by dedicated hardware, and translation device 1 may be realized by software, firmware, or a combination of software and firmware. When translation device 1 is realized by software, firmware, or the like, programs for causing a computer to execute the respective processing procedures of original text acquisition unit 11, first translation unit 12, back translation unit 13, correction acceptance unit 14, second translation unit 17, and display processing unit 16 are stored in memory 31. Then, a processor 32 of the computer executes the programs stored in memory 31.

[0040] 6 shows an example in which each of the components of translation device 1 is realized by dedicated hardware, while Fig. 3 shows an example in which translation device 1 is realized by software, firmware, etc. However, this is merely an example, and some of the components in translation device 1 may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0041] Next, the operation of the translation system shown in FIG. 5 will be described. If the correction receiving unit 14 has accepted a correction, the second translation unit 17 acquires the corrected original text from the correction receiving unit 14. If the correction receiving unit 14 has corrected the original text one or more times, the second translation unit 17 acquires the original text most recently corrected by the correction receiving unit 14 as the corrected original text. If the correction receiving unit 14 has not accepted a correction, the second translation unit 17 acquires the Japanese original text from the original text acquisition unit 11. If the correction receiving unit 14 has accepted a correction, the second translation unit 17 translates the corrected original text accepted by the correction receiving unit 14 into a second language, similar to the second translation unit 15 shown in FIG. 1. Furthermore, the second translation unit 17 translates the corrected original text accepted by the correction receiving unit 14 into one or more languages ​​different from each of the first language and the second language. The one or more languages ​​different from each of the first language and the second language may be, for example, Spanish, German, Italian, Chinese, Korean, or Tagalog.

[0042] If no correction has been accepted by the correction accepting unit 14, the second translation unit 17 translates the original text acquired by the original text acquiring unit 11 into the second language, similar to the second translation unit 15 shown in Fig. 1. The second translation unit 17 also translates the original text acquired by the original text acquiring unit 11 into one or more languages ​​different from each of the first language and the second language. The second translation unit 17 outputs the translation in the second language and the translation in one or more languages ​​different from each of the first language and the second language to the display processing unit 16.

[0043] The display processing unit 16 displays the original text to be translated, which is output from the original text acquisition unit 11, on the display device 3. The display processing unit 16 displays the English translation, which is output from the first translation unit 12, on the display device 3. The display processing unit 16 displays the back-translation, which is output from the back translation unit 13, on the display device 3. The display processing unit 16 displays the Portuguese translation, which is output from the second translation unit 15, on the display device 3. The display processing unit 16 also displays the translations, in one or more languages ​​different from each of the first language and the second language, which are output from the second translation unit 15, on the display device 3.

[0044] In the second embodiment described above, the translation device 1 is configured so that the second translation unit 17 translates either the corrected original text received by the correction receiving unit 14 or the original text acquired by the original text acquisition unit 11 into the second language, and also translates either the corrected original text received by the correction receiving unit 14 or the original text acquired by the original text acquisition unit 11 into one or more languages ​​different from the first language and the second language. Therefore, the translation device 1 can reduce mistranslations when translating an original text to be translated into multiple languages.

[0045] Embodiment 3 In embodiment 3, a translation device 1 will be described that includes a first translation unit 41 that, if a term included in an original text acquired by an original text acquisition unit 11 is a technical term recorded in a technical term dictionary, acquires translation information corresponding to the term included in the original text from the technical term dictionary and uses the translation information to translate the term included in the original text into a first language.

[0046] FIG. 7 is a configuration diagram showing a translation system including a translation device 1 according to embodiment 3. In FIG. 7, the same reference numerals as those in FIGS. 1 and 5 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted. FIG. 8 is a hardware configuration diagram showing the hardware of the translation device 1 according to embodiment 3. In FIG. 8, the same reference numerals as those in FIGS. 2 and 6 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted. The translation system shown in FIG. 7 includes a translation device 1, a translation engine 4, and a display device 3. The translation device 1 shown in FIG. 7 includes an original text acquisition unit 11, a first translation unit 41, a back translation unit 42, a correction acceptance unit 14, a second translation unit 43, and a display processing unit 16.

[0047] 1, when an original text to be translated is provided from the translation device 1, the translation engine 4 translates the original text into each language and outputs the translations in each language to the translation device 1. Unlike the translation engine 2 shown in Fig. 1, the translation engine 4 includes a technical term dictionary in which correspondences between technical terms and translation information are recorded. When translating an original text into, for example, a first language, the translation engine 4 refers to the technical term dictionary, and if a term contained in the original text is a technical term recorded in the technical term dictionary, it obtains translation information corresponding to the term contained in the original text from the technical term dictionary and uses the translation information to translate the term contained in the original text into the first language.

[0048] The first translation unit 41 is realized, for example, by the first translation circuit 51 shown in FIG. 8 . The first translation unit 41 acquires an original text to be translated from the original text acquisition unit 11. Similar to the first translation unit 12 shown in FIG. 1 , the first translation unit 41 translates the original text acquired by the original text acquisition unit 11 into a first language. When translating the original text acquired by the original text acquisition unit 11 into the first language, the first translation unit 41 refers to a technical term dictionary. If a term contained in the original text is a technical term recorded in the technical term dictionary, the first translation unit 41 acquires translation information corresponding to the term contained in the original text from the technical term dictionary and uses the translation information to translate the term contained in the original text into the first language. In the translation system shown in FIG. 7 , the translation engine 4 is provided external to the translation device 1. However, this is merely an example, and the translation engine 4 may be provided inside the translation device 1. For example, if the first translation unit 41 includes the translation engine 4, the first translation unit 41 translates the original text acquired by the original text acquisition unit 11 into the first language. The first translation unit 41 outputs the translation of the first language to both the back translation unit 42 and the display processing unit 16.

[0049] The back translation unit 42 is realized by, for example, the back translation circuit 52 shown in FIG. 8 . The back translation unit 42 acquires a translation in the first language from the first translation unit 41. The back translation unit 42 back-translates the translation by the first translation unit 41. Specifically, the back translation unit 42 provides the translation in the first language to the translation engine 4 and acquires the back-translated text from the translation engine 4. If the first translation unit 41 has translated a term contained in the original text acquired by the original text acquisition unit 11 using translation information acquired from a technical terminology dictionary, the back translation unit 42 back-translates the translation by the first translation unit 41 by referring to the correspondence between the term contained in the original text and the translation information. The back translation unit 42 outputs the back-translated text to the display processing unit 16.

[0050] The second translation unit 43 is realized by, for example, the second translation circuit 53 shown in FIG. 8 . If the correction accepting unit 14 has accepted a correction, the second translation unit 43 acquires the corrected original text from the correction accepting unit 14. If the correction accepting unit 14 has not accepted a correction, the second translation unit 43 acquires the Japanese original text from the original text acquisition unit 11. If the correction accepting unit 14 has accepted a correction, the second translation unit 43 translates the corrected original text accepted by the correction accepting unit 14 into the second language. If the correction accepting unit 14 has not accepted a correction, the second translation unit 43 translates the original text acquired by the original text acquisition unit 11 into the second language. Specifically, the second translation unit 43 provides either the corrected original text accepted by the correction accepting unit 14 or the original text acquired by the original text acquisition unit 11 to the translation engine 4, and acquires a translation in the second language from the translation engine 4. When translating either the corrected original text received by the correction receiving unit 14 or the original text acquired by the original text acquisition unit 11 into the second language, if the first translation unit 41 has translated terms included in the original text acquired by the original text acquisition unit 11 using translation information acquired from a technical terminology dictionary, the second translation unit 43 uses the translation information to translate the terms included in the corrected original text received by the correction receiving unit 14 or the terms included in the original text acquired by the original text acquisition unit 11 into the second language. The second translation unit 43 outputs the translation in the second language to the display processing unit 16.

[0051] In the translation system shown in Fig. 7, the first translation unit 41, the back translation unit 42, and the second translation unit 43 are each applied to the translation device 1 shown in Fig. 1. However, this is merely an example, and the first translation unit 41, the back translation unit 42, and the second translation unit 43 may each be applied to the translation device 1 shown in Fig. 5.

[0052] 7, it is assumed that each of the components of the translation device 1, namely, the original text acquisition unit 11, the first translation unit 41, the back translation unit 42, the correction acceptance unit 14, the second translation unit 43, and the display processing unit 16, is realized by dedicated hardware as shown in FIG. 8. That is, it is assumed that the translation device 1 is realized by the original text acquisition circuit 21, the first translation circuit 51, the back translation circuit 52, the correction acceptance circuit 24, the second translation circuit 53, and the display processing circuit 26. Each of the original text acquisition circuit 21, the first translation circuit 51, the back translation circuit 52, the correction acceptance circuit 24, the second translation circuit 53, and the display processing circuit 26 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

[0053] The components of translation device 1 are not limited to those realized by dedicated hardware, and translation device 1 may be realized by software, firmware, or a combination of software and firmware. When translation device 1 is realized by software, firmware, or the like, programs for causing a computer to execute the respective processing procedures of original text acquisition unit 11, first translation unit 41, back translation unit 42, correction acceptance unit 14, second translation unit 43, and display processing unit 16 are stored in memory 31 shown in Fig. 3. Then, processor 32 shown in Fig. 3 executes the programs stored in memory 31.

[0054] 8 shows an example in which each of the components of translation device 1 is realized by dedicated hardware, while Fig. 3 shows an example in which translation device 1 is realized by software, firmware, etc. However, this is merely an example, and some of the components in translation device 1 may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0055] Next, the operation of the translation system shown in Fig. 7 will be described. Except for the translation engine 4, the first translation unit 41, the back translation unit 42, and the second translation unit 43, the system is the same as the translation system shown in Fig. 1. Therefore, the operation of the translation engine 4, the first translation unit 41, the back translation unit 42, and the second translation unit 43 will be mainly described here.

[0056] As shown in FIG. 9 , the translation engine 4 includes a technical term dictionary in which correspondences between technical terms and translation information are recorded. Examples of translation information include information indicating how to read the technical terms and information indicating the translation of the technical terms in a first language. FIG. 9 is an explanatory diagram showing examples of technical terms, their definitions, how to read the technical terms, and their translations in a first language recorded in the technical term dictionary. The technical terms recorded in the technical term dictionary include not only so-called technical terms but also, for example, in-house terminology. Technical terms are, for example, terms used by people engaged in a specific occupation or terms used only in a specific academic field or industry. In the example of FIG. 9 , the technical terms recorded include "Advanced Research Institute," "Information Research Institute," "IS Building," "M Seminar," and "Deadline." The definitions of the technical terms are displayed on the display device 3 by the display processing unit 16, for example, as annotations of the original text when the original text to be translated is displayed on the display device 3. For example, the definition of "Advanced Research Institute" is "Advanced Technology Research Institute," and the definition of "M Seminar" is "MELCO Seminar." For example, the reading of "Joho-Soken" is "Joho-Soken" and the reading of "IS-tou" is "IS-Tou." For example, the translation of "M Seminar" in the first language is "MELCO Seminar," and the translation of "Regular Closing Day" in the first language is "Regular Closing Day."

[0057] The first translation unit 41 acquires the original text to be translated from the original text acquisition unit 11. The first translation unit 41 provides the original text acquired by the original text acquisition unit 11 to the translation engine 4 and causes the translation engine 4 to translate the original text into the first language. When translating the original text acquired by the original text acquisition unit 11 into the first language, if a term contained in the original text is a technical term recorded in a technical terminology dictionary, the first translation unit 41 uses translation information corresponding to the term contained in the original text to translate the term contained in the original text into the first language.

[0058] When the translation engine 4 is given the original text from the first translation unit 41, it extracts the terms included in the original text and checks whether the extracted terms are specialized terms recorded in the specialized term dictionary. If the terms included in the original text do not include the specialized terms recorded in the specialized term dictionary, the translation engine 4 translates the original text into the first language and outputs the translated text in the first language to the first translation unit 41, similar to the translation engine 2 shown in FIG. 1. If the terms included in the original text include the specialized terms recorded in the specialized term dictionary, the translation engine 4 obtains the translation information corresponding to the terms included in the original text from the specialized term dictionary and uses the translation information to translate the terms included in the original text into the first language. Specifically, if the term included in the original text is, for example, "fixed retirement date", the term included in the original text is translated into either "Teitaibi", which is the reading of "fixed retirement date", or "Regular Closing Day", which is the translation of "fixed retirement date" into the first language. Note that the first translation unit 41 may be provided with a man-machine interface that accepts either the selection of information indicating the reading of the specialized term or the selection of information indicating the translation of the specialized term into the first language as the translation information used when translating the terms included in the original text into the first language, so that the user can select either the information indicating the reading of the specialized term or the information indicating the translation into the first language as the translation information. If the translation information selected by the user indicates the reading of the specialized term, the translation engine 4 translates the term "fixed retirement date", which is included in the original text, like "Teitaibi". If the translation information selected by the user indicates the translation into the first language, the translation engine 4 translates the term "fixed retirement date", which is included in the original text, like "Regular Closing Day".

[0059] The first translation unit 41 outputs the translation in the first language to each of the back translation unit 42 and the display processing unit 16. Any method may be used to check whether the terms included in the original text are technical terms recorded in the technical dictionary, but for example, the translation engine 4 may provide the original text to a generation AI (Artificial Intelligence), which may determine whether the terms included in the original text are technical terms recorded in the technical dictionary.

[0060] The back translation unit 42 obtains a translation in the first language from the first translation unit 41. The back translation unit 42 back translates the translation by the first translation unit 41. Specifically, the back translation unit 42 provides the translation in the first language to the translation engine 4 and obtains the back-translated text from the translation engine 4. When translating terms contained in the original text into the first language, the translation engine 4 obtains translation information corresponding to the terms contained in the original text from a technical terminology dictionary, and if the terms contained in the original text have been translated into the first language using the translation information, back translates the translation by the first translation unit 41 by referring to the correspondence between the terms contained in the original text and the translation information. Specifically, if the term included in the original sentence is, for example, "regular closing day" and "regular closing day" is translated as either "Teitaibi" or "Regular Closing Day," the translation engine 4 refers to the correspondence between the term included in the original sentence and the translation information and back-translates either "Teitaibi" or "Regular Closing Day" into "regular closing day." The back-translation unit 42 outputs the back-translated sentence to the display processing unit 16.

[0061] If the correction accepting unit 14 has accepted a correction, the second translation unit 43 acquires the corrected original text from the correction accepting unit 14. If the correction accepting unit 14 has not accepted a correction, the second translation unit 43 acquires the original text in Japanese from the original text acquisition unit 11. If the correction accepting unit 14 has accepted a correction, the second translation unit 43 translates the corrected original text accepted by the correction accepting unit 14 into the second language, and if the correction accepting unit 14 has not accepted a correction, the second translation unit 43 translates the original text acquired by the original text acquisition unit 11 into the second language.

[0062] Specifically, the second translation unit 43 provides the translation engine 4 with either the corrected original text received by the correction receiving unit 14 or the original text acquired by the original text acquisition unit 11, and acquires a translation into the second language from the translation engine 4. When translating terms contained in the original text into the first language, the translation engine 4 acquires translation information corresponding to the terms contained in the original text from a technical terminology dictionary, and, if the terms contained in the original text have been translated into the first language using the translation information, uses the translation information to translate either the terms contained in the corrected original text received by the correction receiving unit 14 or the terms contained in the original text acquired by the original text acquisition unit 11 into the second language. For example, if the term included in the original text is "Tei-teibi" (regular closing day), and "Tei-teibi" is translated as either "Tei-teibi" or "Regular Closing Day," the translation engine 4 translates either "Tei-teibi" or "Regular Closing Day" into the second language. When translating into the second language using information indicating the pronunciation of a technical term, such as "Tei-teibi," the translation into the second language may be performed by simply using "Tei-teibi." Alternatively, the translation into the second language may be performed by converting "Tei-teibi" into a phonetic character corresponding to the second language that has the same pronunciation as "Tei-teibi." As described above, by using information indicating the pronunciation of a technical term or a first language translation of the technical term to translate into the second language, accurate multilingual translation can be performed even if the original text contains technical terms. The translation engine 4 outputs the translation in the second language to the second translation unit 43. Upon receiving the translation in the second language from the translation engine 4, the second translation unit 43 outputs the translation in the second language to the display processing unit 16.

[0063] In the third embodiment described above, the translation device 1 shown in FIG. 7 is configured such that, when the first translation unit 41 translates the original text acquired by the original text acquisition unit 11 into the first language, the first translation unit 41 refers to a technical term dictionary that stores correspondences between technical terms and translation information. If a term contained in the original text is a technical term stored in the technical term dictionary, the first translation unit 41 acquires translation information corresponding to the term contained in the original text from the technical term dictionary and uses the translation information to translate the term contained in the original text into the first language. Therefore, similar to the translation device 1 shown in FIG. 1 , the translation device 1 shown in FIG. 7 can translate an original text to be translated into multiple languages ​​and can reduce the time required for translation into multiple languages. The translation device 1 shown in FIG. 7 can improve the translation accuracy of the first language when the original text contains technical terms compared to the translation device 1 shown in FIG. 1 .

[0064] In the third embodiment, the translation device 1 shown in Fig. 7 is configured so that, if a term contained in an original text acquired by the original text acquisition unit 11 by the first translation unit 41 is translated using translation information acquired from a technical terminology dictionary, the back translation unit 42 refers to the correspondence between the term contained in the original text and the translation information and back translates the translation by the first translation unit 41. Therefore, the translation device 1 shown in Fig. 7 can obtain a correct back translation even if the original text includes technical terms.

[0065] In the third embodiment, the translation device 1 shown in Fig. 7 is configured so that, when the second translation unit 43 translates the corrected original text received by the correction receiving unit 14 into the second language, if the terms included in the original text acquired by the original text acquiring unit 11 have been translated using translation information acquired from a technical terminology dictionary by the first translation unit 41, the terms included in the original text are translated into the second language using the translation information. Therefore, the translation device 1 shown in Fig. 7 can improve the translation accuracy into the second language when the original text includes technical terms compared to the translation device 1 shown in Fig. 1.

[0066] Fourth Embodiment In a fourth embodiment, a translation device 1 will be described that includes a corresponding language identification unit 44 that identifies, as a second language, a language supported by a target person to whom a translation by a second translation unit 45 is provided.

[0067] FIG. 10 is a configuration diagram showing a translation system including a translation device 1 according to embodiment 4. In FIG. 10, the same reference numerals as those in FIGS. 1, 5, and 7 indicate the same or corresponding parts, and detailed descriptions thereof will be omitted. FIG. 11 is a hardware configuration diagram showing the hardware of the translation device 1 according to embodiment 4. In FIG. 11, the same reference numerals as those in FIGS. 2, 6, and 8 indicate the same or corresponding parts, and detailed descriptions thereof will be omitted. The translation system shown in FIG. 10 includes a translation device 1, a translation engine 2, and a display device 3. The translation device 1 shown in FIG. 10 includes an original text acquisition unit 11, a first translation unit 12, a back translation unit 13, a correction acceptance unit 14, a corresponding language identification unit 44, a second translation unit 45, and a display processing unit 16.

[0068] The supported language identification unit 44 is realized, for example, by the supported language identification circuit 54 shown in FIG. 11 . The supported language identification unit 44 identifies the supported language of the recipient as the second language. Specifically, the supported language identification unit 44 identifies the supported language of the recipient based on image information indicating an image in which the recipient is shown. Alternatively, the supported language identification unit 44 identifies the supported language of the recipient based on audio information indicating the voice of the recipient. The supported language identification unit 44 outputs the supported language identification result to the second translation unit 45.

[0069] The second translation unit 45 is realized by, for example, the second translation circuit 55 shown in FIG. 11 . The second translation unit 45 acquires the identification result of the corresponding language from the corresponding language identification unit 44. If the correction accepting unit 14 has accepted a correction, the second translation unit 45 acquires the corrected original text from the correction accepting unit 14. If the correction accepting unit 14 has not accepted a correction, the second translation unit 45 acquires the original text in Japanese from the original text acquisition unit 11. If the correction accepting unit 14 has accepted a correction, the second translation unit 45 translates the corrected original text accepted by the correction accepting unit 14 into the language accepted by the target user. If the correction accepting unit 14 has not accepted a correction, the second translation unit 45 translates the original text acquired by the original text acquisition unit 11 into the language accepted by the target user. The second translation unit 45 outputs the translation in the language accepted by the target user, which is the second language, to the display processing unit 16.

[0070] In the translation system shown in Fig. 10, the corresponding language identification unit 44 and the second translation unit 45 are each applied to the translation device 1 shown in Fig. 1. However, this is merely an example, and the corresponding language identification unit 44 and the second translation unit 45 may each be applied to either the translation device 1 shown in Fig. 5 or the translation device 1 shown in Fig. 7.

[0071] 10 assumes that each of the components of the translation device 1, namely, the original text acquisition unit 11, the first translation unit 12, the back translation unit 13, the correction acceptance unit 14, the corresponding language identification unit 44, the second translation unit 45, and the display processing unit 16, is realized by dedicated hardware as shown in FIG. 11. That is, it assumes that the translation device 1 is realized by an original text acquisition circuit 21, the first translation circuit 22, the back translation circuit 23, the correction acceptance circuit 24, the corresponding language identification circuit 54, the second translation circuit 55, and the display processing circuit 26. Each of the original text acquisition circuit 21, the first translation circuit 22, the back translation circuit 23, the correction acceptance circuit 24, the corresponding language identification circuit 54, the second translation circuit 55, and the display processing circuit 26 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

[0072] The components of translation device 1 are not limited to those realized by dedicated hardware, and translation device 1 may be realized by software, firmware, or a combination of software and firmware. When translation device 1 is realized by software, firmware, or the like, programs for causing a computer to execute the respective processing procedures of original text acquisition unit 11, first translation unit 12, back translation unit 13, correction acceptance unit 14, corresponding language identification unit 44, second translation unit 45, and display processing unit 16 are stored in memory 31 shown in Fig. 3. Then, processor 32 shown in Fig. 3 executes the programs stored in memory 31.

[0073] 11 shows an example in which each of the components of translation device 1 is realized by dedicated hardware, while Fig. 3 shows an example in which translation device 1 is realized by software, firmware, etc. However, this is merely an example, and some of the components in translation device 1 may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0074] Next, the operation of the translation system shown in Fig. 10 will be described. Except for the corresponding language identification unit 44 and the second translation unit 45, the system is the same as the translation system shown in Fig. 1. Therefore, only the operation of the corresponding language identification unit 44 and the second translation unit 45 will be described here.

[0075] FIG. 12 is an explanatory diagram showing an example of an image in which a recipient is shown. As shown in FIG. 12 , the supported language identification unit 44 acquires image information indicating an image in which the recipient is shown. The supported language identification unit 44 identifies the supported language of the recipient by performing face authentication processing on the recipient shown in the image indicated by the image information. Specifically, if a table is prepared in which facial images of multiple recipients and the supported languages ​​of each recipient are recorded, the supported language identification unit 44 compares the facial image of the recipient shown in the image indicated by the image information with the facial images of the multiple recipients recorded in the table. Then, the supported language identification unit 44 identifies, from the facial images of the multiple recipients recorded in the table, a facial image that matches the facial image of the recipient shown in the image indicated by the image information. The supported language identification unit 44 acquires the supported language of the recipient related to the identified facial image from the table. If multiple recipients are shown in the image, the supported language identification unit 44 may be configured to identify the supported language of each recipient. The corresponding language identification unit 44 outputs the identification result of the corresponding language of the recipient to the second translation unit 45 .

[0076] Here, the supported language identification unit 44 identifies the supported language of the recipient by performing facial authentication processing on the recipient. However, this is merely an example, and the supported language identification unit 44 may provide image information indicating an image of the recipient to the generation AI and obtain an identification result of the supported language of the recipient from the generation AI. In this case, the supported language identification unit 44 may perform, as an authentication process other than facial authentication, for example, fingerprint authentication processing, palm print authentication processing, vein authentication processing, iris authentication processing, or ear shape authentication processing. Furthermore, the supported language identification unit 44 may acquire audio information indicating the recipient's voice and identify the supported language of the recipient by analyzing the language of the voice indicated by the audio information. In this case, the supported language identification unit 44 may provide the audio information to the generation AI and obtain an identification result of the supported language of the recipient from the generation AI.

[0077] The second translation unit 45 acquires the identification result of the supported language from the supported language identification unit 44 and determines the supported language to be the second language. If the correction accepting unit 14 has accepted a correction, the second translation unit 45 acquires the corrected original text from the correction accepting unit 14. If the correction accepting unit 14 has not accepted a correction, the second translation unit 45 acquires the Japanese original text from the original text acquisition unit 11. If the correction accepting unit 14 has accepted a correction, the second translation unit 45 translates the corrected original text accepted by the correction accepting unit 14 into the second language, similar to the second translation unit 15 shown in FIG. 1. If the correction accepting unit 14 has not accepted a correction, the second translation unit 45 translates the original text acquired by the original text acquisition unit 11 into the second language, similar to the second translation unit 15 shown in FIG. 1. The second translation unit 45 outputs the translation in the second language to the display processing unit 16.

[0078] In the above-described fourth embodiment, the translation device 1 shown in Fig. 10 is configured to include a corresponding language identification unit 44 that identifies, as a second language, the corresponding language of a target person, who is a person to whom a translation by the second translation unit 45 is provided, and the second translation unit 45 translates the corrected original text received by the correction receiving unit 14 into the corresponding language identified by the corresponding language identification unit 44. Therefore, similar to the translation device 1 shown in Fig. 1, the translation device 1 shown in Fig. 10 can translate an original text to be translated into multiple languages, can reduce the time required for translation into multiple languages, and can automatically set the corresponding language of the target person as the second language.

[0079] In the translation device 1 shown in Fig. 10, the corresponding language identification unit 44 identifies the corresponding language of the recipient of the translated text as the second language. If there is an error in the identification result of the corresponding language of the recipient, the corresponding language identification unit 44 may be provided with a man-machine interface that accepts corrections to the identification result of the corresponding language of the recipient. In this case, if the user corrects the identification result of the corresponding language identified by the corresponding language identification unit 44, the correct corresponding language can be determined as the second language even if there is an error in the identification result of the corresponding language identified by the corresponding language identification unit 44.

[0080] Fifth Embodiment In the fifth embodiment, a translation device 1 will be described that includes a first modified sentence acquisition unit 46 that provides the generation AI with an original sentence acquired by the original sentence acquisition unit 11 and acquires a modified sentence of the original sentence from the generation AI, and a second modified sentence acquisition unit 47 that provides the generation AI with a back-translated sentence by the back translation unit 13 and acquires a modified sentence of the back-translated sentence from the generation AI.

[0081] FIG. 13 is a configuration diagram showing a translation system including a translation device 1 according to embodiment 5. In FIG. 13, the same reference numerals as those in FIGS. 1, 5, 7, and 10 indicate the same or corresponding parts, and detailed descriptions thereof will be omitted. FIG. 14 is a hardware configuration diagram showing the hardware of the translation device 1 according to embodiment 5. In FIG. 14, the same reference numerals as those in FIGS. 2, 6, 8, and 11 indicate the same or corresponding parts, and detailed descriptions thereof will be omitted. The translation system shown in FIG. 13 includes a translation device 1, a translation engine 2, and a display device 3. The translation device 1 shown in FIG. 13 includes an original text acquisition unit 11, a first modified text acquisition unit 46, a first translation unit 12, a back translation unit 13, a second modified text acquisition unit 47, a correction acceptance unit 14, a second translation unit 45, and a display processing unit 16.

[0082] The first modified sentence acquisition unit 46 is realized, for example, by the first modified sentence acquisition circuit 56 shown in Fig. 14. The first modified sentence acquisition unit 46 provides the original sentence acquired by the original sentence acquisition unit 11 to the generation AI and acquires a modified sentence of the original sentence from the generation AI. The first modified sentence acquisition unit 46 outputs the modified sentence of the original sentence to each of the first translation unit 12, the correction acceptance unit 14, the second translation unit 15, and the display processing unit 16.

[0083] The second revised sentence acquisition unit 47 is realized by, for example, the second revised sentence acquisition circuit 57 shown in Fig. 14. The second revised sentence acquisition unit 47 provides the back-translated sentence by the back translation unit 13 to the generation AI and acquires the revised sentence of the back-translated sentence from the generation AI. The second revised sentence acquisition unit 47 outputs the revised sentence of the back-translated sentence to the display processing unit 16.

[0084] 13 includes both a first modified sentence acquisition unit 46 and a second modified sentence acquisition unit 47. However, this is merely an example, and the translation device 1 may include either the first modified sentence acquisition unit 46 or the second modified sentence acquisition unit 47.

[0085] In the translation system shown in Fig. 13, each of the first modified sentence acquisition unit 46 and the second modified sentence acquisition unit 47 is applied to the translation apparatus 1 shown in Fig. 1. However, this is merely an example, and each of the first modified sentence acquisition unit 46 and the second modified sentence acquisition unit 47 may be applied to any of the translation apparatus 1 shown in Fig. 5, the translation apparatus 1 shown in Fig. 7, or the translation apparatus 1 shown in Fig. 10.

[0086] 13, it is assumed that each of the components of translation apparatus 1, namely, original text acquisition unit 11, first revised text acquisition unit 46, first translation unit 12, back translation unit 13, second revised text acquisition unit 47, correction acceptance unit 14, second translation unit 15, and display processing unit 16, is realized by dedicated hardware as shown in Fig. 14. In other words, it is assumed that translation apparatus 1 is realized by original text acquisition circuit 21, first revised text acquisition circuit 56, first translation circuit 22, back translation circuit 23, second revised text acquisition circuit 57, correction acceptance circuit 24, second translation circuit 25, and display processing circuit 26. Each of the original text acquisition circuit 21, the first modified text acquisition circuit 56, the first translation circuit 22, the back translation circuit 23, the second modified text acquisition circuit 57, the modification reception circuit 24, the second translation circuit 25, and the display processing circuit 26 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination of these.

[0087] The components of translation device 1 are not limited to those realized by dedicated hardware, and translation device 1 may be realized by software, firmware, or a combination of software and firmware. When translation device 1 is realized by software, firmware, or the like, programs for causing a computer to execute the respective processing procedures of original text acquisition unit 11, first revised text acquisition unit 46, first translation unit 12, back translation unit 13, second revised text acquisition unit 47, correction acceptance unit 14, second translation unit 15, and display processing unit 16 are stored in memory 31 shown in FIG. 3. Then, processor 32 shown in FIG. 3 executes the programs stored in memory 31.

[0088] 14 shows an example in which each of the components of translation device 1 is realized by dedicated hardware, while Fig. 3 shows an example in which translation device 1 is realized by software, firmware, etc. However, this is merely an example, and some of the components in translation device 1 may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0089] Next, the operation of the translation system shown in Fig. 13 will be described. Except for the first modified sentence acquisition unit 46 and the second modified sentence acquisition unit 47, the translation system is the same as that shown in Fig. 1. Therefore, here, the operation of the first modified sentence acquisition unit 46 and the second modified sentence acquisition unit 47 will be mainly described. Fig. 15 is a flowchart showing a translation method, which is a processing procedure of the translation device 1. In Fig. 15, steps that are the same as those in Fig. 4 indicate the same or equivalent parts, and detailed description thereof will be omitted.

[0090] The original text acquisition unit 11 acquires a Japanese original text as the original text to be translated (step ST1 in FIG. 15 ). For ease of explanation, it is assumed here that the acquired Japanese original text is, "Yesterday, on the production line for the ENG-5678 model, a part was installed incorrectly, which required rework. Please be careful to avoid work errors such as incorrect assembly." The original text acquisition unit 11 outputs the Japanese original text to both the first modified text acquisition unit 46 and the display processing unit 16. The display processing unit 16 displays the Japanese original text, "Yesterday, on the production line for the ENG-5678 model, a part was installed incorrectly, which required rework. Please be careful to avoid work errors such as incorrect assembly." on the display device 3.

[0091] The first modified sentence acquisition unit 46 acquires the original Japanese text from the original text acquisition unit 11. The first modified sentence acquisition unit 46 is equipped with a man-machine interface such as a microphone. When a user wants the original Japanese text to be converted into clear Japanese, the user provides a voice prompt for the generation AI indicating, for example, "Please convert into clear Japanese" to the first modified sentence acquisition unit 46. When the above prompt is provided, the first modified sentence acquisition unit 46 provides the original Japanese text together with the above prompt to the generation AI. When the above prompt is not provided, the first modified sentence acquisition unit 46 outputs the original Japanese text to each of the first translation unit 12, the correction acceptance unit 14, the second translation unit 15, and the display processing unit 16.

[0092] When the generation AI receives the prompt and the original Japanese text from the first modified text acquisition unit 46, it converts the original Japanese text into clean Japanese and outputs the converted original text to the first modified text acquisition unit 46 as a modified text of the original text. The process of correcting the original text by the generation AI is a well-known technique, so a detailed description is omitted. The first modified text acquisition unit 46 acquires a modified text of the original text from the generation AI (step ST11 in FIG. 15 ). Here, the generation AI outputs the following modified text to the first modified text acquisition unit 46: "Yesterday, on the ENG-5678 model production line, due to an incorrect installation of a part, a review was required. Please be careful to avoid work errors such as misassembly." The first modified text acquisition unit 46 outputs the modified text of the original text to each of the first translation unit 12, the correction acceptance unit 14, the second translation unit 15, and the display processing unit 16. The display processing unit 16 displays on the display device 3 the corrected text of the original text, "Yesterday, on the production line for the ENG-5678 model, a review was required due to an incorrect part being installed. Please be careful to avoid work errors such as misassembly."

[0093] 13 shows that the first modified sentence acquisition unit 46 provides a voice prompt to the generation AI indicating, for example, "Please convert this into beautiful Japanese." However, this is merely one example, and for example, a screen for setting the beautiful Japanese mode to either ON or OFF may be displayed on the display device 3, and if the user sets it to ON, the first modified sentence acquisition unit 46 may provide the original Japanese sentence to the generation AI, causing the generation AI to convert the original sentence into beautiful Japanese.

[0094] The first translation unit 12 acquires the original Japanese text or the modified Japanese text from the first modified text acquisition unit 46. The first translation unit 12 translates the original Japanese text or the modified Japanese text into the first language, English (step ST2 in FIG. 15 ). Specifically, the first translation unit 12 provides the original Japanese text or the modified Japanese text to the translation engine 2 and acquires the English translation from the translation engine 2. The first translation unit 12 outputs the English translation to the back translation unit 13 and the display processing unit 16. The display processing unit 16 displays the English translation output from the first translation unit 12 on the display device 3.

[0095] The back translation unit 13 obtains the English translation from the first translation unit 12. The back translation unit 13 back-translates the translation by the first translation unit 12 (step ST3 in FIG. 15 ). Specifically, the back translation unit 13 provides the English translation to the translation engine 2 and obtains the back-translated sentence from the translation engine 2. The back translation unit 13 outputs the back-translated sentence to the second modified sentence obtaining unit 47.

[0096] The second revised sentence acquisition unit 47 acquires the back-translated sentence from the back translation unit 13. The second revised sentence acquisition unit 47 is equipped with a man-machine interface such as a microphone. When the user wants the back-translated sentence to be converted into beautiful Japanese, the user provides a voice prompt for the generation AI indicating, for example, "Please convert into beautiful Japanese" to the second revised sentence acquisition unit 47. When the above prompt is provided, the second revised sentence acquisition unit 47 provides the back-translated sentence together with the above prompt to the generation AI. When the above prompt is not provided, the second revised sentence acquisition unit 47 outputs the back-translated sentence to the display processing unit 16.

[0097] When the generation AI receives the prompt and the back-translated sentence from the second revised sentence acquisition unit 47, it converts the back-translated sentence into clean Japanese and outputs the converted back-translated sentence as a revised sentence of the back-translated sentence to the second revised sentence acquisition unit 47. The second revised sentence acquisition unit 47 acquires the revised sentence of the back-translated sentence from the generation AI (step ST12 in FIG. 15 ). The second revised sentence acquisition unit 47 outputs the revised sentence of the back-translated sentence to the display processing unit 16.

[0098] The display processing unit 16 displays the back-translated sentence output from the second revised sentence acquisition unit 47 or the revised sentence of the back-translated sentence output from the second revised sentence acquisition unit 47 on the display device 3 (step ST4 in Figure 15).

[0099] If the user, after checking the back-translated text, recognizes the difference in nuance between the Japanese original text or the revised text of the original text and the back-translated text or the revised text of the back-translated text, the user can correct the Japanese original text by operating the correction accepting unit 14. When the user performs an operation to further correct the original text (step ST5 in FIG. 15 : YES), the correction accepting unit 14 accepts the corrections to the original text (step ST6 in FIG. 15 ) and outputs the corrected original text to the first translation unit 12. If the first corrected text acquisition unit 46 has acquired a revised text of the original text, the original text for which the correction accepting unit 14 accepts corrections is the revised text of the original text acquired by the first corrected text acquisition unit 46. If the first corrected text acquisition unit 46 has not acquired a revised text of the original text, the original text for which the correction accepting unit 14 accepts corrections is the original text acquired by the original text acquisition unit 11.

[0100] If the user does not perform an operation to further correct the corrected original text (step ST5 in FIG. 15: NO), the correction receiving unit 14 does not accept any corrections to the original text. If the original text has been corrected one or more times, the correction receiving unit 14 outputs the most recently corrected original text to the second translation unit 15.

[0101] The second translation unit 15 acquires either the original text last corrected by the correction receiving unit 14, the Japanese original text acquired by the original text acquisition unit 11, or the modified text of the original text acquired by the first modified text acquisition unit 46. The second translation unit 15 translates any of the acquired original texts into Portuguese, which is the second language (step ST7 in FIG. 15 ). Specifically, the second translation unit 15 provides any of the acquired original texts to the translation engine 2 and acquires a translation into Portuguese from the translation engine 2. The second translation unit 15 outputs the translation in the second language to the display processing unit 16.

[0102] In the above-described fifth embodiment, the translation device 1 shown in Fig. 13 is configured to include a first modified sentence acquisition unit 46 that provides the original sentence acquired by the original sentence acquisition unit 11 to the generation AI and acquires a modified sentence of the original sentence from the generation AI, and to have the first translation unit 12 translate the modified sentence acquired by the first modified sentence acquisition unit 46 into the first language instead of the original sentence acquired by the original sentence acquisition unit 11. Therefore, the translation device 1 shown in Fig. 13 can improve the translation accuracy of the first language compared to the translation device 1 shown in Fig. 1.

[0103] In the fifth embodiment, the translation device 1 shown in Fig. 13 is configured to include a second revised sentence acquisition unit 47 that provides the back-translated sentence by the back translation unit 13 to the generation AI and acquires a revised sentence of the back-translated sentence from the generation AI, the display processing unit 16 displays the revised sentence of the back-translated sentence by the second revised sentence acquisition unit 47 on the display device 3, and the correction accepting unit 14 accepts corrections to the original sentence acquired by the original sentence acquisition unit 11 or the revised sentence acquired by the first revised sentence acquisition unit 46 after the revised sentence of the back-translated sentence is displayed by the display processing unit 16. Therefore, the translation device 1 shown in Fig. 13 can improve the translation accuracy of the second language more than the translation device 1 shown in Fig. 1.

[0104] In the fifth embodiment, an example is shown in which the first revised sentence acquisition unit 46 or the second revised sentence acquisition unit 47 provides the generation AI with a prompt such as "Please convert into beautiful Japanese." However, this is merely an example, and the first revised sentence acquisition unit 46 or the like may provide the generation AI with a prompt such as "Please convert into Japanese that is easy for even a new employee to understand" or "Please convert into Japanese that is easy for even a non-specialist to understand," thereby obtaining a revised sentence from the generation AI that is tailored to the level of the recipient. If the recipients include, for example, a new employee or someone who is not familiar with the field, correcting the original sentence using the above prompt can obtain a revised sentence that is easy for the recipient to understand.

[0105] Furthermore, the first corrected sentence acquisition unit 46 or the like may acquire information about the recipient and change the prompt based on the information about the recipient. The first corrected sentence acquisition unit 46 or the like may identify the attributes of the recipient using, for example, image information or audio information acquired by the corresponding language identification unit 44 shown in FIG. 10 , and use a prompt corresponding to the identified attributes of the recipient, thereby making corrections tailored to the level of the recipient. Specifically, if a table of recipients is prepared that stores attribute information indicating the level of the recipient, such as year of employment or specialty, the first corrected sentence acquisition unit 46 or the like may use either image information or audio information to identify the attributes of the recipient and acquire a prompt corresponding to the attributes of the recipient from the table. For example, if the table of recipients includes attribute information such as "first year with the company" and "low level of expertise," this information may be used to generate prompts such as "Please translate into Japanese that is easy to understand even for new employees" or "Please translate into Japanese that is easy to understand even for non-specialists." This allows sentences tailored to the level of the recipient to be provided.

[0106] In addition, the present disclosure allows for free combination of the respective embodiments, modification of any of the components of the respective embodiments, or omission of any of the components of the respective embodiments.

[0107] The present disclosure is suitable for a translation device and a translation method.

[0108] REFERENCE SIGNS LIST 1 Translation device, 2 Translation engine, 3 Display device, 4 Translation engine, 11 Original text acquisition unit, 12 First translation unit, 13 Back translation unit, 14 Correction acceptance unit, 15, 17 Second translation unit, 16 Display processing unit, 21 Original text acquisition circuit, 22 First translation circuit, 23 Back translation circuit, 24 Correction acceptance circuit, 25, 27 Second translation circuit, 26 Display processing circuit, 31 Memory, 32 Processor, 41 First translation unit, 42 Back translation unit, 43, 45 Second translation unit, 44 Corresponding language identification unit, 46 First corrected sentence acquisition unit, 47 Second corrected sentence acquisition unit, 51 First translation circuit, 52 Back translation circuit, 53, 55 Second translation circuit, 54 Corresponding language identification circuit, 56 First corrected sentence acquisition circuit, 57 Second corrected sentence acquisition circuit.

Claims

1. A translation device comprising: an original text acquisition unit that acquires an original text to be translated; a first translation unit that translates the original text acquired by the original text acquisition unit into a first language; a back translation unit that back translates the text translated by the first translation unit; a display processing unit that displays the back translation by the back translation unit on a display device; a correction receiving unit that accepts corrections to the original text after the back translation is displayed by the display processing unit; and a second translation unit that, when the corrections to the original text are accepted by the correction accepting unit, translates the original text after the corrections accepted by the correction accepting unit into a second language.

2. The translation device according to claim 1, characterized in that when the correction receiving unit receives a correction to the original text, the first translation unit translates the original text after the correction received by the correction receiving unit into the first language.

3. The translation device according to claim 1 or 2, characterized in that the second translation unit translates either the original text after corrections received by the correction receiving unit or the original text acquired by the original text acquisition unit into the second language, and also translates either the original text after corrections received by the correction receiving unit or the original text acquired by the original text acquisition unit into one or more languages ​​different from each of the first language and the second language.

4. A translation device according to any one of claims 1 to 3, characterized in that the first language is English, and the first translation unit translates the original text acquired by the original text acquisition unit into English.

5. A translation device as described in any one of claims 1 to 4, characterized in that when translating the original text acquired by the original text acquisition unit into the first language, the first translation unit refers to a technical terminology dictionary in which the correspondence between technical terms and translation information is recorded, and if a term contained in the original text is a technical term recorded in the technical terminology dictionary, it acquires translation information corresponding to the term contained in the original text from the technical terminology dictionary and uses the translation information to translate the term contained in the original text into the first language.

6. The translation device of claim 5, wherein the translation information corresponding to the technical term is information indicating how to read the technical term or information indicating the translation of the technical term in the first language, and the first translation unit accepts a selection of either information indicating how to read the technical term or information indicating the translation of the technical term in the first language as translation information to be used when translating terms contained in the original text acquired by the original text acquisition unit into the first language.

7. A translation device as described in claim 5 or claim 6, characterized in that if the first translation unit translates a term contained in the original text acquired by the original text acquisition unit using translation information acquired from the technical terminology dictionary, the back translation unit refers to the correspondence between the term contained in the original text and the translation information and back translates the translation by the first translation unit.

8. A translation device as described in any one of claims 5 to 7, characterized in that when translating the original text after corrections received by the correction receiving unit into the second language, if a term contained in the original text acquired by the original text acquisition unit has been translated by the first translation unit using translation information acquired from the technical terminology dictionary, the second translation unit translates the term into the second language using the translation information.

9. A translation device as described in any one of claims 1 to 8, characterized in that it is provided with a corresponding language identification unit that identifies the corresponding language of the recipient, who is the person to whom the translation by the second translation unit is provided, as the second language, and the second translation unit translates the original text after corrections received by the correction receiving unit into the corresponding language identified by the corresponding language identification unit.

10. The translation device according to claim 9, characterized in that the corresponding language identification unit identifies the corresponding language of the target person based on image information indicating an image in which the target person is shown.

11. The translation device according to claim 9, wherein the corresponding language identification unit identifies the corresponding language of the recipient based on voice information indicating the voice of the recipient.

12. A translation device according to any one of claims 9 to 11, characterized in that the corresponding language identification unit accepts corrections to the identification result of the corresponding language of the target user.

13. A translation device as described in any one of claims 1 to 12, characterized in that it comprises a first modified sentence acquisition unit that provides the original sentence acquired by the original sentence acquisition unit to a generation AI (Artificial Intelligence) and acquires a modified sentence of the original sentence from the generation AI, and the first translation unit translates the modified sentence of the original sentence acquired by the first modified sentence acquisition unit into the first language instead of the original sentence acquired by the original sentence acquisition unit.

14. A translation device as described in any one of claims 1 to 12, characterized in that it comprises a second modified sentence acquisition unit that provides a back-translated sentence by the back translation unit to a generation AI and acquires a modified sentence of the back-translated sentence from the generation AI, wherein the display processing unit displays the modified sentence of the back-translated sentence by the second modified sentence acquisition unit on a display device, and the correction accepting unit accepts corrections to the original sentence acquired by the original sentence acquisition unit after the modified sentence of the back-translated sentence is displayed by the display processing unit.

15. A translation device as described in claim 13, further comprising a second modified sentence acquisition unit that provides a back-translated sentence by the back translation unit to a generation AI and acquires a modified sentence of the back-translated sentence from the generation AI, wherein the display processing unit displays the modified sentence of the back-translated sentence by the second modified sentence acquisition unit on a display device, and the correction accepting unit accepts corrections to the original sentence acquired by the original sentence acquisition unit or the modified sentence of the original sentence acquired by the first modified sentence acquisition unit after the modified sentence of the back-translated sentence is displayed by the display processing unit.

16. The translation device described in claim 13, characterized in that the first modified sentence acquisition unit further provides attribute information to the generation AI, which is information indicating the level of the recipient, who is the person to whom the translation by the second translation unit is provided, and acquires from the generation AI a modified sentence of the original sentence that corresponds to the level of the recipient.

17. A translation method in which an original text acquisition unit acquires an original text to be translated; a first translation unit translates the original text acquired by the original text acquisition unit into a first language; a display processing unit displays the back-translated text by the back translation unit on a display device; a correction receiving unit accepts corrections to the original text after the back-translated text is displayed by the display processing unit; and a second translation unit, when the corrections to the original text are accepted by the correction accepting unit, translates the original text after the corrections accepted by the correction accepting unit into a second language.

Images