WO2024004184A1 - Generation device, generation method, and program - Google Patents

Generation device, generation method, and program Download PDF

Info

Publication number
WO2024004184A1
WO2024004184A1 PCT/JP2022/026407 JP2022026407W WO2024004184A1 WO 2024004184 A1 WO2024004184 A1 WO 2024004184A1 JP 2022026407 W JP2022026407 W JP 2022026407W WO 2024004184 A1 WO2024004184 A1 WO 2024004184A1
Authority
WO
WIPO (PCT)
Prior art keywords
constraint
unit
information
sequence
generation
Prior art date
Application number
PCT/JP2022/026407
Other languages
French (fr)
Japanese (ja)
Inventor
克己 帖佐
睦 森下
昌明 永田
Original Assignee
日本電信電話株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 日本電信電話株式会社 filed Critical 日本電信電話株式会社
Priority to PCT/JP2022/026407 priority Critical patent/WO2024004184A1/en
Publication of WO2024004184A1 publication Critical patent/WO2024004184A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/40—Processing or translation of natural language
    • G06F40/42—Data-driven translation

Definitions

  • the present invention relates to the technical field of machine translation.
  • Vocabulary-constrained machine translation is when a sentence in one domain is translated into another domain (e.g., another language), with constraints imposed to ensure that all specified words (constraint words) are included. Since machine translation with vocabulary constraints can unify the translation of specific words, machine translation with vocabulary constraints is a particularly important technology in the translation of patents, legal documents, technical documents, etc. that require consistency.
  • the present invention has been made in view of the above points, and it is an object of the present invention to provide a technique for accurately performing sequence conversion using constraint information.
  • a generation device for generating a second series that is another information series from constraint information and a first series that is an information series, an input generation unit that inputs a constraint information list and outputs each element of a subset of one or more constraint information included in the constraint information list as a vocabulary constraint; a sequence generation unit that generates one or more candidates for the second sequence using the first sequence and the vocabulary constraint; and a reranking unit that calculates a score indicating suitability as the second series for each of the one or more candidates.
  • FIG. 2 is a diagram showing an example of machine translation with vocabulary constraints.
  • 1 is a diagram showing a configuration example of a generation device 100.
  • FIG. 3 is a flowchart for explaining the operation of the generation device 100.
  • 3 is a diagram showing an example configuration of an extraction unit 120.
  • FIG. 3 is a diagram showing an example configuration of an extraction unit 120.
  • FIG. 1 is a diagram showing a configuration example of a generation device 100.
  • FIG. 3 is a diagram showing an example of the configuration of a sequence generation unit 140.
  • FIG. It is a diagram showing an example of the configuration of a machine translation model.
  • 3 is a diagram showing an example of the configuration of a sequence generation unit 140.
  • FIG. 5 is a diagram showing a display image on a display unit 500.
  • FIG. 1 is a diagram showing a configuration example of a generation device 100.
  • FIG. 3 is a flowchart for explaining the operation of the generation device 100.
  • 3 is a diagram showing an example configuration
  • FIG. 1 is a diagram showing a configuration example of a generation device 100.
  • FIG. FIG. 3 is a diagram showing detailed settings and hyperparameters that serve as a base for each setting used in the experiment. It is a figure showing an evaluation result. It is a diagram showing an example of the hardware configuration of the device.
  • the present invention can be applied to machine translation, but the present invention can be applied to sequence conversion in any field as long as it uses constraint information. It is.
  • the present invention can be used for summarization tasks, utterance generation tasks, tasks for adding explanatory text to images, and the like.
  • the unit of translation is a sentence, but the unit of translation may be any unit.
  • the generation device 100 described below provides certain improvements over prior art techniques such as performing constrained sequence transformations, and represents an improvement in the technical field of constrained sequence transformations. be. Additionally, the extraction apparatus described below provides certain improvements over the prior art in extracting constraint information and represents an improvement in the field of technology related to extracting constraint information.
  • FIG. 1 shows an example of input and output in machine translation with vocabulary constraints.
  • Non-Patent Document 1 As a conventional technology for machine translation with vocabulary constraints, see Non-Patent Document 1 "Chen, G., Chen, Y., and Li, V. O. (2021). "Lexically Constrained Neural Machine Translation with Explicit Alignment Guidance.” Proceedings of The ⁇ AAAI Conference on Artificial Intelligence'' discloses a machine translation method with vocabulary constraints for manually created constraint phrases. The method disclosed in Non-Patent Document 1 is also called a soft method. The method disclosed in Non-Patent Document 1 does not guarantee that the constraint phrase will always be included in the translated sentence.
  • Non-patent document 2 “Matt Post and David Vilar. 2018. Fast Lexically Constrained Decoding with Dynamic Beam Allocation for Neural Machine Translation. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Vol. ume 1 (Long Papers), pages 1314-1324, New La, Louisiana. Association for Computational Linguistics” and Reference 1 “Chousa, K. and Morishita, M. (2021). “Input Augmentation Improves Constrained Beam Search for Neural Machine Translation : NTT at WAT 2021.” In Proceedings of the 8th Workshop on Asian Translation (WAT), pp. 53--61, Online. Association for Computational Linguistics.” A translation method is disclosed. This method guarantees that the constraint phrase will always be included in the translated sentence. This method is also called the hard method.
  • the extracted constraint phrases include words that become noise. Further, even when constraint words and phrases are extracted manually, noise may be included.
  • a source language sentence is input using the input unit 110.
  • the extraction unit 120 automatically extracts constraint phrases based on the source language sentence (input sentence) input by the input unit 110 and the bilingual dictionary read from the bilingual dictionary DB 200.
  • FIG. 4 is a block diagram of the extraction unit 120.
  • the extraction unit 120 includes a filtering unit 121, a division unit 122, and a constraint phrase extraction unit 123.
  • the extraction unit 120 also refers to the bilingual dictionary 200. Note that the extraction section 120 may be configured without the filtering section 121.
  • the bilingual dictionary DB 200 stores a set of pairs of two words that are made to correspond when converting sequences. Specifically, in this embodiment, which targets translation, the bilingual dictionary DB 200 stores a set of ⁇ source language phrase, target language phrase> pairs.
  • the source language phrase and the target language phrase may each consist of multiple words. In this embodiment, one ⁇ source language word/phrase, target language word/phrase> pair is referred to as a "bilingual translation".
  • the source language phrase and the target language phrase may be called a source language translation word and a target language translation word, respectively.
  • ⁇ Extraction unit 120 Filtering unit 121>
  • the filtering unit 121 deletes bilingual translations that fall under (A) to (C) below, or words included in the bilingual translations, from the bilingual dictionary.
  • the filtering unit 121 does not necessarily need to implement all of (A) to (C), but may implement at least one of (A) to (C). Further, filtering other than (A) to (C) may be performed. In particular, in Modifications 1 and 2, which will be described later, the process (C) below may be skipped.
  • the parallel translation of "source language: computer, target language: computer, computer” corresponds to (C). so that there is a one-to-one relationship between the source language phrase and the target language phrase.
  • the source language sentence after processing by the dividing unit 122 will be "As long as it is, it is not.”
  • Constraint phrase extraction unit 123 extracts bilingual translations corresponding to the phrases included in the source language sentence, and creates a constraint phrase list using the extracted bilingual translations.
  • a specific example of the constraint phrase extraction method will be described below. Note that the format of the dictionary, the search method, etc. are not limited to the method described below, and other methods may be used as long as the method can extract the constraint phrases corresponding to the words included in the source language sentence.
  • prefix match longest match
  • word division as described here are examples of means for realizing constraint phrase extraction with less noise and reduced ambiguity. Other means of disambiguation may be used.
  • the segmentation unit 122 when performing morphological analysis in the segmentation unit 122, information necessary for disambiguation, such as part of speech, prototype, stem, conjugation, and reading (pronunciation), is attached to the divided words and the attached information is also used.
  • Perform matching In other words, by using not only the character string but also attached information such as its part of speech during matching, for example, the character "in" in the source language sentence can be matched with the preposition in, which is the source language translation, and the noun inn (inn).
  • the ambiguity can be resolved by placing the situation in which both of the above match. Resolving ambiguity during matching is an important element in improving translation accuracy.
  • the generation device 100 may not include the extraction unit 120.
  • the configuration of the generation device 100 in this case is shown in FIG.
  • the constraint phrase list generated by the extraction device is input to the generation device 100.
  • a constraint phrase list that is not the constraint phrase list generated by the extraction device eg, a constraint phrase list that includes a lot of noise
  • the operations of the input generation section 130, sequence generation section 140, and reranking section 150 in FIG. 6 are the same as the operations of the input generation section 130, sequence generation section 140, and reranking section 150 in FIG.
  • the input generation unit 130 receives the constraint phrase list as input, and sets all elements of a subset of the words included in the constraint phrase list as vocabulary constraints. However, some elements among all the elements may be subjected to vocabulary constraints.
  • ⁇ A, B, C ⁇ is input to the input generation unit 130 as a constraint phrase list.
  • A, B, and C are each constraint phrases.
  • the input generation unit 130 generates ⁇ , ⁇ A ⁇ , ⁇ B ⁇ , ⁇ C ⁇ , ⁇ A, B ⁇ , ⁇ A, C ⁇ , ⁇ B, C ⁇ , ⁇ A, B, C ⁇ and output each of them as a vocabulary constraint.
  • ⁇ , ⁇ A ⁇ , ⁇ B ⁇ , ⁇ C ⁇ , ⁇ A, B ⁇ , ⁇ A, C ⁇ , ⁇ B, C ⁇ , ⁇ A, B, C ⁇ is a constraint vocabulary set.
  • one ⁇ ... ⁇ is one vocabulary constraint.
  • sequence generation unit 140 Next, the sequence generation unit 140 will be explained. It is assumed that the series generation unit 140 holds a trained machine translation model read from the model DB 300. Furthermore, the sequence generation unit 140 repeats the following process by the number of lexical constraints (the number of elements in the lexical constraint set). For example, the vocabulary constraint set is ⁇ , ⁇ A ⁇ , ⁇ B ⁇ , ⁇ C ⁇ , ⁇ A, B ⁇ , ⁇ A, C ⁇ , ⁇ B, C ⁇ , ⁇ A, B, C ⁇ Then, repeat 8 times.
  • the sequence generation unit 140 receives an input sentence (source language sentence) and vocabulary constraints as input.
  • the sequence generation unit 140 generates a translated sentence (target language sentence) using a machine translation model by applying an existing method of machine translation with vocabulary constraints.
  • a plurality of translated sentences are generated as translated sentence candidates (target language sentence candidates).
  • a translation sentence candidate is given a score as a translation sentence.
  • LeCA is disclosed in Non-Patent Document 1 and is also called the soft method.
  • LeCA+LCD is disclosed in Reference 1 mentioned above, and is also called the hard method.
  • the series generation unit 140 outputs a plurality of generated translation sentence candidates.
  • the sequence generation unit 140 outputs a predetermined number of translation sentence candidates in descending order of scores.
  • the "predetermined number" may be one. In other words, only the translated sentence with the highest score may be output.
  • 30 translation candidates are output for each vocabulary constraint.
  • FIG. 7 shows a configuration example of the sequence generation unit 140.
  • the sequence generation section 140 includes a sequence conversion section 141 and a search section 142.
  • the sequence conversion unit 141 uses lexical constraint information, and when using the hard method, the sequence conversion unit 141 uses the lexical constraints depending on the type of hard method. Sometimes it's used, sometimes it's not. Arrows for inputting vocabulary constraints to the series conversion unit 142 are indicated by dotted lines.
  • the hard methods in the aforementioned LeCA+LCD, information on vocabulary constraints is used in the sequence conversion unit 141. The configuration/operation assuming LeCA+LCD will be described below.
  • the sequence conversion unit 141 uses a general encoder-decoder model (for example, Transformer) as a machine translation model, which has an encoder and a decoder, as shown in FIG. model can be used.
  • a general encoder-decoder model for example, Transformer
  • the invention can be implemented using models other than the encoder-decoder model.
  • the sequence conversion unit 141 receives the source language sentence and vocabulary constraints as input, first expands the source language sentence using the vocabulary constraints to create an input sequence with information on the vocabulary constraints added, and then machine translates it. Use as input to the model.
  • the sequence conversion unit 141 generates a sentence by using the expanded input sequence as input to a machine translation model. More specifically, the probability of each word in a set of words that can constitute an output sequence is output.
  • the search unit 142 uses the output probability of the decoder in the machine translation model to search for (an approximate solution of) the output sequence that maximizes the generation probability when the input sequence is given.
  • the search unit 142 uses a grid beam search method based on beam search to ensure that the output sequence satisfies all of the constraint vocabulary.
  • searching unit 142 searching using grid beam search is an example. Any processing method may be used as long as it performs a lexically constrained search so as to include the constraint word/phrase.
  • the reranking unit 150 receives as input one or more translated sentence candidates generated by the series generation unit 140. For example, if the series generation unit 140 generates 30 translation candidates per lexical constraint and there are 8 lexical constraints, the reranking unit 150 generates 30 translation candidates per lexical constraint. Each time, 30 translation candidates are received as input.
  • the reranking unit 150 calculates a score for each translated sentence candidate using the input sentence (source language sentence), and outputs the translated sentence candidate with the highest score as the final translated sentence.
  • the output unit 160 can present the translated sentences to the user in a ranking format using the scores.
  • any method may be used as long as it can calculate the score of the translated sentence, but for example, the methods in Example 1 and Example 2 below may be used. be able to.
  • Example 1 The reranking unit 150 uses, as a score, the likelihood of translated sentence candidates output by the machine translation model used for translation in the sequence generation unit 140.
  • Example 2 The reranking unit 150 uses a machine translation model learned by Transformer, which is an encoder-decoder model, for a Right-to-Left translation task that generates a translated sentence from the end of a sentence to the beginning of a sentence.
  • the likelihood when translation candidates are forced to be output is used as a score.
  • Forcibly outputting translated sentence candidates may be rephrased as forced decoding using translated sentence candidates.
  • the source language sentence is input to the encoder of the reranking model, and the words of the translation sentence candidate whose score (likelihood) is to be evaluated are sequentially input to the decoder of the reranking model.
  • the likelihood output by the machine translation model may be any value as long as it indicates plausibility.
  • the likelihood output by the machine translation model may be a probability or a value other than probability.
  • the reranking unit 150 may calculate the reranking score using both the likelihood of Example 1 and the likelihood of Example 2. For example, the average of the likelihood of Example 1 and the likelihood of Example 2 may be used as the reranking score.
  • the constraint phrase list generated by the extraction unit 120 can be one in which a plurality of target language phrases correspond to one source language phrase.
  • a constraint phrase list may be called a constraint phrase list that allows multiple translations. For example, if the filtering unit of the extraction unit 120 does not perform the step (C), such a constraint phrase list may be generated.
  • a and A' as a plurality of target language phrases for a certain source language phrase, and "A, A', B, C" including these and B and C are added to the constraint phrase list by the extraction unit 120.
  • it is generated as multiple elements. For example, if the word in the source language sentence is "computer” and the word in the target language sentence is "computer”, A and A' correspond to "computer” and "computer".
  • the input generation unit 130 that receives ⁇ A, A′ ⁇ , ⁇ B ⁇ , ⁇ C ⁇ from the extraction unit 120 generates ⁇ , ⁇ A ⁇ , ⁇ B ⁇ , ⁇ C ⁇ , ⁇ A, B ⁇ , ⁇ A, C ⁇ , ⁇ B, C ⁇ , ⁇ A, B, C ⁇ , as well as ⁇ A' ⁇ , ⁇ A', B ⁇ , ⁇ A', C ⁇ , ⁇ A', B, C ⁇ is also generated as a vocabulary constraint.
  • the input generation unit 130 inputs each of the plurality of generated vocabulary constraints to the sequence generation unit 140.
  • the sequence generation unit 140 generates ⁇ , ⁇ A ⁇ , ⁇ B ⁇ , ⁇ C ⁇ , ⁇ A, B ⁇ , ⁇ A, C ⁇ , ⁇ B, C ⁇ , ⁇ A, B, C ⁇ , ⁇ A'
  • machine translation with vocabulary constraints is performed 12 times. and obtain translation candidates. For example, if one translated sentence candidate is generated for one vocabulary constraint, 12 translated sentence candidates will be obtained.
  • the reranking unit 150 After performing the machine translation with vocabulary constraints, the reranking unit 150 performs the reranking process using the method described above, and outputs, for example, the translated sentence candidate with the highest score as the final translated sentence. .
  • Modification 2 Next, modification 2 will be explained. Also in the second modification, the constraint phrase list generated by the extraction unit 120 can be one in which a plurality of target language phrases correspond to one source language phrase.
  • the search in the translation search process in the search unit 142 of the sequence generation unit 140, the search may be performed by allowing the existence of a plurality of expression types of one constraint phrase. In other words, the search may be performed such that one element is satisfied from each constraint word candidate. Specifically, the details are as follows.
  • ⁇ A, B, C ⁇ is generated as the constraint word list, and information indicating that A may be A' or A'' is generated. It is assumed that the input is input from the extraction unit 120 to the input generation unit 130. Alternatively, ⁇ A, A', A'', B, C ⁇ is generated as a constraint word list, and information indicating that any of A, A', and A'' is acceptable is sent from the extraction unit 120 to the input generation unit. 130 may be input.
  • the input generation unit 130 For the constraint word list ⁇ A, B, C ⁇ , if A may be A', the input generation unit 130 generates ⁇ , ⁇ A, A' ⁇ , ⁇ B ⁇ , ⁇ C ⁇ , ⁇ A, A' ⁇ , ⁇ B ⁇ , ⁇ A, A' ⁇ , ⁇ C ⁇ , ⁇ A, A' ⁇ , ⁇ B ⁇ , ⁇ C ⁇ seven vocabulary candidate constraints generate.
  • Modification 2 there is ambiguity in the translated word because there are cases where multiple target language words (e.g., A, A') correspond to a certain source language word, and the vocabulary used as a constraint is determined. Therefore, we call it ⁇ vocabulary candidate constraint'' instead of lexical constraint.
  • the "vocabulary candidate constraint” is a vocabulary constraint that maintains ambiguity.
  • the expression format of the vocabulary candidate constraints described above is an example. As long as it can be expressed that either A or A' is acceptable, expression formats other than those described above may be used as the expression format.
  • the sequence generation unit 140 receives the source language sentence and the vocabulary candidate constraints as input.
  • the sequence generation unit 140 generates ⁇ , ⁇ A, A′ ⁇ , ⁇ B ⁇ , ⁇ C ⁇ , ⁇ A, A′ ⁇ , ⁇ B ⁇ , ⁇ A, A′ ⁇ , ⁇ C ⁇ , ⁇ A, A' ⁇ , ⁇ B ⁇ , ⁇ C ⁇ , each of the seven vocabulary candidate constraints was used as a vocabulary candidate constraint, and machine translation with vocabulary constraints was performed seven times. , obtain translation candidates. For example, if one translated sentence candidate is generated for one vocabulary candidate constraint, seven translated sentence candidates will be obtained.
  • the reranking unit 150 After performing the machine translation with vocabulary constraints, the reranking unit 150 performs the reranking process using the method described above, and outputs, for example, the translated sentence candidate with the highest score as the final translated sentence. .
  • the search unit 142 of the sequence generation unit 140 When using a vocabulary candidate constraint including ⁇ A, A' ⁇ , the search unit 142 of the sequence generation unit 140 performs a search assuming that the word A may be A'. In other words, a search is performed that takes ambiguity into account.
  • a search is performed that takes ambiguity into account.
  • reference 2 “Peter Anderson, Basura Fernando, Mark Johnson, and Stephen Gould. 2017. Guided Open Vocabulary Image Captioning with Constrained Beam Search In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 936-945, Copenhagen, Denmark. Association for Computational Linguistics can be used. This method is an example of a "search considering ambiguity" method.
  • Reference 2 is a language generation method, but is not a translation technology. There is no prior art that applies this method to search during translation decoding.
  • multiple target language phrases e.g. A, A'
  • synonyms such as "calculator” and "computer”.
  • ⁇ Trunk'' it may be a word or phrase other than a synonym, such as ⁇ car trunk'', ⁇ elephant's trunk'', trunk, trunk line, etc. Since the meaning of words is not taken into account when searching in the search unit 142, A and A' may be completely unrelated words.
  • arbitrary criteria can be used to determine which words in the converted series correspond to words in the original series.
  • the bilingual dictionary is English-Japanese, and there is a dictionary entry for "corn".
  • the source language sentence "We roasted corns over the charcoal.”
  • the extraction unit 120 performs matching on a morpheme basis, corns is matched in the bilingual dictionary because corn is included as a morpheme.
  • the bilingual dictionary entry is feet and the morpheme in the input sentence is foot, there will be no match. Therefore, by returning the foot to its original form and performing matching, this problem can be resolved.
  • the display unit 500 displays a plurality of constraint phrases (constraint phrase list) for the input source language sentence.
  • the words and phrases displayed here as constraint words are the words and phrases that have been filtered by the filtering unit 121.
  • the filtered constraint phrases are displayed in the form of "Do you want to add?"
  • FIG. 11 shows a configuration example of a generation device 100 for realizing the above display.
  • the generation device 100 of this embodiment includes an extraction section 120, a display information generation section 170, a modification section 180, a generation section 190, a bilingual dictionary DB 200, and a constraint phrase list DB 400.
  • the modification unit 180 may be included in the display information generation unit 170.
  • the bilingual dictionary DB 200 and the constraint phrase list DB 400 may be provided outside the generation device 100.
  • the generation unit 190 may also be provided outside the generation device 100 (eg, another server).
  • the generation device 100 may be used for the purpose of displaying a list of constraint words on the display unit 500.
  • the generation device 100 may include only the extraction unit 120 and the display information generation unit 170 among the functional units shown in FIG.
  • the generation device 100 may also be called an extraction device.
  • the functions of each part are as follows.
  • the extraction unit 120 is the extraction unit 120 shown in FIG. 4 or 5. It takes the source language sentence as input and outputs a list of constraint words.
  • the output constraint phrase list is stored in the constraint phrase list DB 400 and is input to the display information generation section 170. Further, the extraction unit 120 may output the filtered constraint words as a filter word list. The output filter word list is input to the display information generation section 170.
  • the display information generation unit 170 generates information for displaying the constraint phrase list on the display unit 500 (referred to as constraint phrase list presentation information).
  • the constraint word/phrase list presentation information includes a constraint word/phrase list. Further, the information for presenting the constraint word/phrase list may include information on the filter word/phrase list as deleted information, filter candidate words, or additional candidates.
  • the constraint word list presentation information is transmitted from the display information generation section 170 to the display section 500 and input to the display section 500. Further, the display information generation unit 170 may generate display information to be displayed together with the target language sentence (translated sentence) generated using the constraint phrase in a format in which the constraint phrase can be modified.
  • the display information generation unit 170 when the generation device 100 receives the added or modified constraint phrase from the display unit 500, the display information generation unit 170 generates a target language sentence (translated sentence) based on the received constraint phrase. Display information for displaying the target language sentence (translated sentence) may be generated.
  • the display information generation unit 170 may generate “correction support information” for the user to check the constraint phrase list, and transmit it to the display unit 500.
  • the modification support information includes at least one of a source language sentence input by the user, an extracted constraint phrase list, and a target language sentence generated based on the extracted constraint phrase list.
  • the modification unit 180 receives from the display unit 500 at least one of the additional constraint phrases and the modified constraint phrases as information that the user has modified the presented constraint phrase list.
  • the modification unit 180 modifies the information stored in the constraint phrase list DB 400 based on the received information.
  • a target language sentence is generated again by machine translation with vocabulary constraints based on the modified constraint phrase list, and the display information generation unit 170 generates a target language sentence.
  • the modification support information may be generated and displayed on the display section 500 by transmitting it to the display section 500.
  • the display unit 500 is, for example, a computer (terminal) having a display.
  • the display unit 500 is connected to the generation device 100 via a network.
  • the display unit 500 receives a source language sentence from the user and displays a list of constraint words and the like.
  • the display unit 500 also accepts instructions for adding and modifying constraint words and sentences in the source language.
  • the display unit 500 can also output a source language sentence, a final target language sentence, and a final constraint phrase list as a set.
  • the generation device 100 in the above embodiment generates a target language sentence (translated sentence) that is closer to the user's image by interactively repeating the process of modifying the restricted word list while checking the results of machine translation with vocabulary constraints. can do.
  • the score used for reranking translation candidates was the score calculated by the reranking model from the source language sentence and translation candidates using Reranker.
  • FIG. 13 shows the translation accuracy of each method when using vocabulary constraints automatically extracted by a bilingual dictionary. It can be seen that LeCA and LeCA+LCD are able to improve translation accuracy compared to the baseline (Transformer) in Reranker, which uses scores based on the reranking model. Moreover, from FIG. 13, it can be seen that the translation accuracy is high regardless of the type of dictionary.
  • Any of the devices (generating device 100, extracting device) described in this embodiment can be realized, for example, by causing a computer to execute a program.
  • This computer may be a physical computer or a virtual machine on the cloud.
  • the device can be realized by using hardware resources such as a CPU and memory built into a computer to execute a program corresponding to the processing performed by the device.
  • the above program can be recorded on a computer-readable recording medium (such as a portable memory) and can be stored or distributed. It is also possible to provide the above program through a network such as the Internet or e-mail.
  • FIG. 14 is a diagram showing an example of the hardware configuration of the computer.
  • the computer in FIG. 14 includes a drive device 1000, an auxiliary storage device 1002, a memory device 1003, a CPU 1004, an interface device 1005, a display device 1006, an input device 1007, an output device 1008, etc., which are interconnected by a bus BS.
  • the computer may further include a GPU.
  • a program that realizes processing on the computer is provided, for example, on a recording medium 1001 such as a CD-ROM or a memory card.
  • a recording medium 1001 such as a CD-ROM or a memory card.
  • the program is installed from the recording medium 1001 to the auxiliary storage device 1002 via the drive device 1000.
  • the program does not necessarily need to be installed from the recording medium 1001, and may be downloaded from another computer via a network.
  • the auxiliary storage device 1002 stores installed programs as well as necessary files, data, and the like.
  • the technology described in this embodiment makes it possible to appropriately automatically extract constraint phrases used in machine translation with vocabulary constraints with low noise. Furthermore, with the technology described in this embodiment, it is possible to perform translation with high accuracy in machine translation with vocabulary constraints.
  • Additional note 1 memory and at least one processor connected to the memory; including;
  • the processor includes: dividing each of the first information and the first series in a dictionary which is a set of pairs of first information and second information into unit information;
  • the second information corresponding to the first information that matches the unit information of the first series is extracted from the dictionary as constraint information used to generate a second series based on the first series.
  • Extraction device The extraction device according to supplementary note 1, wherein the processor deletes pairs that match a predetermined rule from the dictionary, and uses the dictionary that has been subjected to the deletion process.
  • Pairs that match the predetermined rules include words other than nouns, pairs containing words other than noun phrases, pairs consisting of words with a length of 1, and pairs that are unique in the correspondence between the first information and the second information.
  • the processor generates display information for transmitting the constraint information to the display unit, and receives constraint information added or modified to the constraint information displayed on the display unit. Additional notes 1 to 4.
  • the extraction device according to any one of these.
  • the processor includes: extracting constraint information based on the first series and a dictionary that is a set of pairs of first information and second information with the first series as input; generating a second sequence based on the constraint information and the first sequence;
  • a generation device that generates display information for displaying the constraint information together with the second series in a modifiable format.
  • the processor obtains a series generated based on the received constraint information, and generates display information for displaying the series. According to supplementary note 6. generator.
  • a computer-implemented extraction method comprising: a dividing step of dividing each of the first information and the first series into unit information in a dictionary that is a set of pairs of first information and second information; The second information corresponding to the first information that matches the unit information of the first series is extracted from the dictionary as constraint information used to generate a second series based on the first series.
  • An extraction method comprising a constraint information extraction step.
  • (Supplementary Note 10) A generation method executed by a computer, an extraction step of receiving the first series as input and extracting constraint information based on the first series and a dictionary that is a set of pairs of first information and second information; a generation step of generating a second sequence based on the constraint information and the first sequence; a display information generation step of generating display information for displaying the constraint information together with the second series in a modifiable format.
  • (Supplementary Note 11) A non-temporary storage medium storing a program for causing a computer to function as the extraction device according to any one of Additional Items 1 to 5.
  • the processor includes: inputting a constraint information list, outputting each element of a subset of one or more constraint information included in the constraint information list as a vocabulary constraint; generating one or more candidates for the second sequence using the first sequence and the vocabulary constraint;
  • a generation device that calculates a score indicating suitability as the second series for each of the one or more candidates.
  • the processor calculates the score based on at least one of a likelihood output by a model used to generate the candidate in the sequence generation unit and a likelihood obtained from the candidate by a reranking model.
  • the generating device according to supplementary note 1.
  • the processor In the case where the constraint information list includes constraint information having ambiguity, the processor generates the one or more candidates by performing a beam search with a vocabulary constraint that takes the ambiguity into account. 2.
  • the generating device according to 2.
  • At least one piece of constraint information is input to the processor in a format that allows two or more ambiguities, and the processor generates a vocabulary constraint while maintaining the ambiguity. Any one of Additional Notes 1 to 3.
  • a non-temporary storage medium storing a program for causing a computer to function as each part of the generation device according to any one of Supplementary Notes 1 to 4.

Abstract

A generation device for generating a second sequence from constraint information and a first sequence which is a sequence of information, said second sequence being a separate sequence of information, and wherein the generation device comprises: an input generation unit that uses a constraint information list as input and outputs each element of one or more constraint information subsets included in the constraint information list as a vocabulary constraint; a sequence generation unit that generates one or more candidates pertaining to the second sequence using the first sequence and the vocabulary constraints; and a re-ranking unit that calculates a score indicating the suitability as the second sequence for each of the one or more candidates.

Description

生成装置、生成方法、及びプログラムGeneration device, generation method, and program
　本発明は、機械翻訳の技術分野に関連するものである。 The present invention relates to the technical field of machine translation.
　あるドメインの文を他のドメイン（例：他言語）に変換する際に、指定された語句（制約語句）をすべて含ませることを目的とした制約を課したものを語彙制約付き機械翻訳という。語彙制約付き機械翻訳により、特定の語に対する訳を統一できることから、語彙制約付き機械翻訳は、一貫性が必要な特許／法務／技術文書などの翻訳においては特に重要となる技術である。 Vocabulary-constrained machine translation is when a sentence in one domain is translated into another domain (e.g., another language), with constraints imposed to ensure that all specified words (constraint words) are included. Since machine translation with vocabulary constraints can unify the translation of specific words, machine translation with vocabulary constraints is a particularly important technology in the translation of patents, legal documents, technical documents, etc. that require consistency.
　語彙制約付き機械翻訳手法では、与えられる制約語句を含むように翻訳文を生成する。一方、自動で制約語句を抽出する場合、人手で制約語句を抽出する場合のいずれにおいても、複数の制約語句の中には、適切でない制約語句（ノイズ）が含まれ得る。 In the machine translation method with vocabulary constraints, a translated sentence is generated to include the given constraint phrases. On the other hand, regardless of whether constraint terms are extracted automatically or manually, inappropriate constraint terms (noise) may be included in the plurality of constraint terms.
　ノイズを含んだ制約語句の全てを語彙制約として語彙制約付き機械翻訳手法を適用すると、間違った語句が翻訳文に含まれることになり、翻訳精度が低下することが想定される。なお、このような課題は、機械翻訳の分野に限らない、制約情報を用いて系列変換を行う分野全般で生じ得る課題である。 If a machine translation method with vocabulary constraints is applied with all constrained phrases containing noise as vocabulary constraints, incorrect words will be included in the translated sentence, and translation accuracy is expected to decrease. Note that such a problem is not limited to the field of machine translation, but can occur in any field in which sequence conversion is performed using constraint information.
　本発明は上記の点に鑑みてなされたものであり、制約情報を使用した系列変換を精度良く行うための技術を提供することを目的とする。 The present invention has been made in view of the above points, and it is an object of the present invention to provide a technique for accurately performing sequence conversion using constraint information.
　開示の技術によれば、制約情報と、情報の系列である第１系列とから別の情報の系列である第２系列を生成するための生成装置であって、
　制約情報リストを入力し、前記制約情報リストに含まれる１又は複数の制約情報の部分集合の各要素を語彙制約として出力する入力生成部と、
　前記第１系列と前記語彙制約とを用いて、前記第２系列についての１又は複数の候補を生成する系列生成部と、
　前記第２系列としての適切さを示すスコアを、前記１又は複数の候補のそれぞれについて算出するリランキング部と
　を備える生成装置が提供される。 According to the disclosed technology, there is provided a generation device for generating a second series that is another information series from constraint information and a first series that is an information series,
an input generation unit that inputs a constraint information list and outputs each element of a subset of one or more constraint information included in the constraint information list as a vocabulary constraint;
a sequence generation unit that generates one or more candidates for the second sequence using the first sequence and the vocabulary constraint;
and a reranking unit that calculates a score indicating suitability as the second series for each of the one or more candidates.
　開示の技術によれば、制約情報を使用した系列変換を精度良く行うための技術が提供される。 According to the disclosed technology, a technology for accurately performing sequence conversion using constraint information is provided.
語彙制約付き機械翻訳の例を示す図である。FIG. 2 is a diagram showing an example of machine translation with vocabulary constraints. 生成装置１００の構成例を示す図である。1 is a diagram showing a configuration example of a generation device 100. FIG. 生成装置１００の動作を説明するためのフローチャートである。3 is a flowchart for explaining the operation of the generation device 100. 抽出部１２０の構成例を示す図である。3 is a diagram showing an example configuration of an extraction unit 120. FIG. 抽出部１２０の構成例を示す図である。3 is a diagram showing an example configuration of an extraction unit 120. FIG. 生成装置１００の構成例を示す図である。1 is a diagram showing a configuration example of a generation device 100. FIG. 系列生成部１４０の構成例を示す図である。3 is a diagram showing an example of the configuration of a sequence generation unit 140. FIG. 機械翻訳モデルの構成例を示す図でる。It is a diagram showing an example of the configuration of a machine translation model. 系列生成部１４０の構成例を示す図である。3 is a diagram showing an example of the configuration of a sequence generation unit 140. FIG. 表示部５００による表示イメージを示す図である。5 is a diagram showing a display image on a display unit 500. FIG. 生成装置１００の構成例を示す図である。1 is a diagram showing a configuration example of a generation device 100. FIG. 実験で使用した各セッティングにおいてベースとなる詳細セッティングとハイパーパラメータを示す図である。FIG. 3 is a diagram showing detailed settings and hyperparameters that serve as a base for each setting used in the experiment. 評価結果を示す図である。It is a figure showing an evaluation result. 装置のハードウェア構成例を示す図である。It is a diagram showing an example of the hardware configuration of the device.
　以下、図面を参照して本発明の実施の形態（本実施の形態）を説明する。以下で説明する実施の形態は一例に過ぎず、本発明が適用される実施の形態は、以下の実施の形態に限られるわけではない。 Hereinafter, an embodiment of the present invention (this embodiment) will be described with reference to the drawings. The embodiments described below are merely examples, and embodiments to which the present invention is applied are not limited to the following embodiments.
　以下で説明する実施の形態では、本発明を機械翻訳に適用する例を示しているが、本発明は、制約情報を用いた系列変換であれば、どのような分野の系列変換にも適用可能である。例えば、要約タスク、発話文生成タスク、画像に説明文を付けるタスク等にも本発明を用いることが可能である。 The embodiment described below shows an example in which the present invention is applied to machine translation, but the present invention can be applied to sequence conversion in any field as long as it uses constraint information. It is. For example, the present invention can be used for summarization tasks, utterance generation tasks, tasks for adding explanatory text to images, and the like.
　また、以下で説明する実施の形態では、翻訳の単位を文としているが、翻訳の単位は任意のものとしてよい。 Furthermore, in the embodiment described below, the unit of translation is a sentence, but the unit of translation may be any unit.
　以下で説明する生成装置１００は、制約を付けた系列変換を行うような従来技術に対して特定の改善を提供するものであり、制約を付けた系列変換に係る技術分野の向上を示すものである。また、以下で説明する抽出装置は、制約情報の抽出において従来技術に対して特定の改善を提供するものであり、制約情報の抽出に係る技術分野の向上を示すものである。 The generation device 100 described below provides certain improvements over prior art techniques such as performing constrained sequence transformations, and represents an improvement in the technical field of constrained sequence transformations. be. Additionally, the extraction apparatus described below provides certain improvements over the prior art in extracting constraint information and represents an improvement in the field of technology related to extracting constraint information.
　（課題について）
　本実施の形態に係る構成と動作を詳細に説明する前に、まず、従来技術とそれに対する課題について説明する。なお、以下の課題の説明は公知技術ではない。また、以下で説明する課題は、実施形態の技術に関する課題である。 (About the assignment)
Before explaining the configuration and operation according to this embodiment in detail, first, the prior art and the problems associated with it will be explained. Note that the following explanation of the problem is not a known technique. Further, the problems described below are problems related to the technology of the embodiment.
　既に説明したように、あるドメインの文を他のドメイン（例：他言語）に変換する際に、指定された語句をすべて含ませることを目的とした制約を課したものを語彙制約付き機械翻訳と呼ぶ。参考として、図１に、語彙制約付き機械翻訳における入出力の例を示す。 As already explained, machine translation with lexical constraints imposes constraints on the purpose of including all specified words when converting sentences from one domain to another domain (e.g., another language). It is called. For reference, FIG. 1 shows an example of input and output in machine translation with vocabulary constraints.
　図１の例では、「光線一致に基づく定常波の幾何光学的理論を展開した。」という原言語文に対して、機械翻訳（MT Output）、制約語句（Constraints）、語彙制約付き機械翻訳（Constrained MT Output）が示されている。下線部分が制約語句を示す。 In the example in Figure 1, for the source language sentence ``We have developed a geometrical optical theory of standing waves based on ray coincidence.'', machine translation (MT Output), constraint words (Constraints), MT Output) is shown. The underlined portion indicates the constraint phrase.
　語彙制約付き機械翻訳の従来技術として、非特許文献１「Chen, G., Chen, Y., and Li, V. O. (2021). "Lexically Constrained Neural Machine Translation with Explicit Alignment Guidance." Proceedings of the AAAI Conference on Artificial Intelligence」には、人手で作成された制約語句に対する語彙制約付き機械翻訳手法が開示されている。非特許文献１に開示された手法は、ｓｏｆｔ法とも呼ばれる。非特許文献１に開示された手法では、制約語句が必ず翻訳文に含まれる保証はされない。 As a conventional technology for machine translation with vocabulary constraints, see Non-Patent Document 1 "Chen, G., Chen, Y., and Li, V. O. (2021). "Lexically Constrained Neural Machine Translation with Explicit Alignment Guidance." Proceedings of The ``AAAI Conference on Artificial Intelligence'' discloses a machine translation method with vocabulary constraints for manually created constraint phrases. The method disclosed in Non-Patent Document 1 is also called a soft method. The method disclosed in Non-Patent Document 1 does not guarantee that the constraint phrase will always be included in the translated sentence.
　非特許文献２「Matt Post and David Vilar. 2018. Fast Lexically Constrained Decoding with Dynamic Beam Allocation for Neural Machine Translation. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), pages 1314-1324, New Orleans, Louisiana. Association for Computational Linguistics」、及び参考文献１「Chousa, K. and Morishita, M. (2021). "Input Augmentation Improves Constrained Beam Search for Neural Machine Translation: NTT at WAT 2021." In Proceedings of the 8th Workshop on Asian Translation (WAT), pp. 53--61, Online. Association for Computational Linguistics.」にも、人手で作成された制約語句に対する語彙制約付き機械翻訳手法が開示されている。この手法では、制約語句が必ず翻訳文に含まれる保証がある。この手法は、ｈａｒｄ法とも呼ばれる。 Non-patent document 2 “Matt Post and David Vilar. 2018. Fast Lexically Constrained Decoding with Dynamic Beam Allocation for Neural Machine Translation. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Vol. ume 1 (Long Papers), pages 1314-1324, New Orleans, Louisiana. Association for Computational Linguistics” and Reference 1 “Chousa, K. and Morishita, M. (2021). “Input Augmentation Improves Constrained Beam Search for Neural Machine Translation : NTT at WAT 2021." In Proceedings of the 8th Workshop on Asian Translation (WAT), pp. 53--61, Online. Association for Computational Linguistics." A translation method is disclosed. This method guarantees that the constraint phrase will always be included in the translated sentence. This method is also called the hard method.
　制約語句を人手で作成するのではなく、自動で作成するユースケースもある。例えば、特許や科学技術論文などの固有名詞を多く含むドメインの文書の翻訳では過去の翻訳結果から作成される翻訳メモリや対訳辞書を用いることが多く、そのため対訳辞書から自動で抽出した制約語句によって語彙制約付き機械翻訳を行うというユースケースが考えられる。 There are also use cases where constraint terms are created automatically rather than manually. For example, when translating documents in domains that include many proper nouns, such as patents and scientific and technical papers, translation memories and bilingual dictionaries created from past translation results are often used. A possible use case is to perform machine translation with vocabulary constraints.
　一方で、制約語句を自動で抽出する場合には抽出された制約語句にノイズとなる語句が含まれることが考えられる。また、人手で制約語句を抽出する場合でも、ノイズは含まれ得る。 On the other hand, when automatically extracting constraint phrases, it is possible that the extracted constraint phrases include words that become noise. Further, even when constraint words and phrases are extracted manually, noise may be included.
　非特許文献１、２等に開示された従来の語彙制約付き機械翻訳手法では、与えられる制約語句が参照訳中に含まれることを仮定している。そのため、抽出された制約語句を語彙制約として語彙制約付き機械翻訳手法を適用すると、間違った語句が翻訳文に含まれる場合があり、翻訳精度が低下することが想定される。 Conventional machine translation methods with vocabulary constraints, such as those disclosed in Non-Patent Documents 1 and 2, assume that the given constraint phrase is included in the reference translation. Therefore, if a machine translation method with vocabulary constraints is applied using the extracted constraint phrases as vocabulary constraints, incorrect words may be included in the translated sentence, and translation accuracy is expected to decrease.
　上記の点から、以下では、ノイズを削減して適切に制約語句を抽出する技術、及び、ノイズが含まれ得る制約語句の集合を使用する場合でも、精度良く語彙制約付き機械翻訳を行うための技術を説明する。 Based on the above points, below we will discuss techniques for appropriately extracting constraint words by reducing noise, and methods for performing machine translation with vocabulary constraints with high accuracy even when using a set of constraint words that may contain noise. Explain the technology.
　（装置構成例、全体動作）
　図２に、本実施の形態における生成装置１００の構成例を示す。図２に示すように、生成装置１００は、入力部１１０、抽出部１２０、入力生成部１３０、系列生成部１４０、リランキング部１５０、及び出力部１６０を有する。 (Equipment configuration example, overall operation)
FIG. 2 shows a configuration example of the generation device 100 in this embodiment. As shown in FIG. 2, the generation device 100 includes an input section 110, an extraction section 120, an input generation section 130, a sequence generation section 140, a reranking section 150, and an output section 160.
　また、対訳辞書ＤＢ２００、及びモデルＤＢ３００が備えられている。対訳辞書ＤＢ２００には、対訳辞書が格納されており、モデルＤＢ３００には、学習済みの機械翻訳モデルが格納されている。対訳辞書ＤＢ２００、及びモデルＤＢ３００は、生成装置１００の外部に備えられていてもよいし（図２の例）、生成装置１００の内部に備えられていてもよい。 Additionally, a bilingual dictionary DB 200 and a model DB 300 are provided. The bilingual dictionary DB 200 stores bilingual dictionaries, and the model DB 300 stores trained machine translation models. The bilingual dictionary DB 200 and the model DB 300 may be provided outside the generation device 100 (as in the example in FIG. 2), or may be provided inside the generation device 100.
　図３のフローチャートを参照して、生成装置１００による全体の動作の流れについて説明する。Ｓ１０１において、入力部１１０により原言語文を入力する。Ｓ１０２において、抽出部１２０は、入力部１１０により入力された原言語文（入力文）と、対訳辞書ＤＢ２００から読み出した対訳辞書に基づいて、制約語句を自動的に抽出する。 The overall flow of operations by the generation device 100 will be described with reference to the flowchart in FIG. 3. In S101, a source language sentence is input using the input unit 110. In S102, the extraction unit 120 automatically extracts constraint phrases based on the source language sentence (input sentence) input by the input unit 110 and the bilingual dictionary read from the bilingual dictionary DB 200.
　Ｓ１０３において、入力生成部１３０は、制約語句の任意の組み合わせから複数の入力（語彙制約）を生成する。Ｓ１０４において、系列生成部１４０は、Ｓ１０３で生成した複数の入力と、モデルＤＢ３００から読み出した機械翻訳モデルを用いて、入力文に対する翻訳を行う。ここでは、Ｓ１０３で生成した複数の入力のそれぞれに対して翻訳結果が得られる。すなわち、系列生成部１４０は、ある系列と語彙制約とを用いて、予め学習済みの系列変換モデルに基づいて、別の系列についての１又は複数の候補を生成する。 In S103, the input generation unit 130 generates a plurality of inputs (vocabulary constraints) from arbitrary combinations of constraint words. In S104, the series generation unit 140 translates the input sentence using the plurality of inputs generated in S103 and the machine translation model read from the model DB 300. Here, translation results are obtained for each of the plurality of inputs generated in S103. That is, the sequence generation unit 140 uses a certain sequence and vocabulary constraints to generate one or more candidates for another sequence based on a previously learned sequence conversion model.
　Ｓ１０５において、リランキング部１５０が、各翻訳結果に対して入力文を用いてリランキング用のスコアを予測する。Ｓ１０６において、出力部１６０が、スコアの最も高い翻訳結果（目的言語文）を出力する。以下、主要な機能部についての構成と動作を詳細に説明する。 In S105, the reranking unit 150 predicts a reranking score for each translation result using the input sentence. In S106, the output unit 160 outputs the translation result (target language sentence) with the highest score. The configuration and operation of the main functional units will be explained in detail below.
　（抽出部１２０）
　まず、抽出部１２０について説明する。抽出部１２０は、原言語文と対訳辞書を入力として受け取り、原言語文と制約語句リストを出力する。なお、原言語文は出力しないこととしてもよい。 (Extraction unit 120)
First, the extraction unit 120 will be explained. The extraction unit 120 receives the source language sentence and the bilingual dictionary as input, and outputs the source language sentence and the constraint word list. Note that the source language sentence may not be output.
　図４は、抽出部１２０の構成図である。図４に示すように、抽出部１２０は、フィルタリング部１２１、分割部１２２、制約語句抽出部１２３を有する。また、抽出部１２０は、対訳辞書２００を参照する。なお、抽出部１２０は、フィルタリング部１２１を備えない構成としてもよい。 FIG. 4 is a block diagram of the extraction unit 120. As shown in FIG. 4, the extraction unit 120 includes a filtering unit 121, a division unit 122, and a constraint phrase extraction unit 123. The extraction unit 120 also refers to the bilingual dictionary 200. Note that the extraction section 120 may be configured without the filtering section 121.
　対訳辞書ＤＢ２００は、系列を変換する際に対応させる２つの語句のペアの集合を格納する。具体的には、翻訳を対象としている本実施の形態では、対訳辞書ＤＢ２００は、＜原言語語句、目的言語語句＞ペアの集合を格納している。原言語語句、及び目的言語語句はそれぞれ、複数の語からなる場合もある。本実施の形態では、１つの＜原言語語句、目的言語語句＞ペアを「対訳」と呼称する。原言語語句、及び目的言語語句をそれぞれ、原言語訳語、及び目的言語訳語と呼んでもよい。 The bilingual dictionary DB 200 stores a set of pairs of two words that are made to correspond when converting sequences. Specifically, in this embodiment, which targets translation, the bilingual dictionary DB 200 stores a set of <source language phrase, target language phrase> pairs. The source language phrase and the target language phrase may each consist of multiple words. In this embodiment, one <source language word/phrase, target language word/phrase> pair is referred to as a "bilingual translation". The source language phrase and the target language phrase may be called a source language translation word and a target language translation word, respectively.
　なお、対訳辞書ＤＢ２００を、翻訳以外のタスクに使う場合には、その内容は、＜原言語語句、目的言語語句＞ペアの集合に限られない。 Note that when the bilingual dictionary DB 200 is used for tasks other than translation, its contents are not limited to a set of <source language phrase, target language phrase> pairs.
　フィルタリング部１２１は、対訳辞書からノイズとなる対訳をフィルタリングする。フィルタリング後の対訳辞書は対訳辞書ＤＢ２００に格納され、分割部１２２及び制約語句抽出部１２３は、フィルタリング後の対訳辞書を参照する。 The filtering unit 121 filters the bilingual translations that become noise from the bilingual dictionary. The bilingual dictionary after filtering is stored in the bilingual dictionary DB 200, and the dividing unit 122 and the constraint phrase extraction unit 123 refer to the bilingual dictionary after filtering.
　分割部１２２は、原言語文、及び、対訳辞書の原言語語句を形態素解析する。つまり、分割部１２２は、原言語文、及び、対訳辞書の原言語語句を単位情報に分割する。制約語句抽出部１２３は、原言語文に含まれる語句（分割で得られた単位情報の例）に対応する対訳を抽出し制約語句リストを作成する。以下、各部の処理をより詳細に説明する。 The dividing unit 122 morphologically analyzes the source language sentence and the source language phrases in the bilingual dictionary. That is, the dividing unit 122 divides the source language sentences and the source language phrases in the bilingual dictionary into unit information. The constraint phrase extraction unit 123 extracts bilingual translations corresponding to the phrases (examples of unit information obtained by division) included in the source language sentence and creates a constraint phrase list. The processing of each part will be explained in more detail below.
　＜抽出部１２０：フィルタリング部１２１＞
　フィルタリング部１２１は、以下（Ａ）～（Ｃ）に当てはまる対訳、又は対訳に含まれる語句を対訳辞書から削除する。ただし、フィルタリング部１２１は、（Ａ）～（Ｃ）の全部を実施することは必須ではなく、（Ａ）～（Ｃ）のうちの少なくとも１つを実施することとしてもよい。また、（Ａ）～（Ｃ）以外のフィルタリングを行ってもよい。特に、後述する変形例１，２においては、以下（Ｃ）の処理はスキップすることとしてもよい。 <Extraction unit 120: Filtering unit 121>
The filtering unit 121 deletes bilingual translations that fall under (A) to (C) below, or words included in the bilingual translations, from the bilingual dictionary. However, the filtering unit 121 does not necessarily need to implement all of (A) to (C), but may implement at least one of (A) to (C). Further, filtering other than (A) to (C) may be performed. In particular, in Modifications 1 and 2, which will be described later, the process (C) below may be skipped.
　（Ａ）名詞／名詞句以外の語句を含む対訳（動詞は活用があるためのこれを除く）
　（Ｂ）長さが１の語句からなる対訳
　単位などの一文字のものが（Ｂ）の例である。例えば、「目的言語：Ｃ、原言語：度」の対訳が（Ｂ）に該当する。 (A) Parallel translations that include words other than nouns/noun phrases (excluding verbs because they have conjugations)
(B) Parallel translation consisting of words with a length of 1 An example of (B) is a one-character translation such as a unit. For example, the parallel translation of "target language: C, source language: degree" corresponds to (B).
　（Ｃ）原言語と目的言語との間の対応関係に一意性が無いもの（例えば、原言語側の１つの語句に対して複数の訳語が存在するもの）
　（Ｃ）に該当する対訳については、対訳を削除する。あるいは、複数の訳語から、１つの訳語を残して他を削除することで、原言語語句と目的言語語句とが１対１になるようにする。複数の訳語から、１つの訳語を残して他を削除する方法についてはどのような方法を用いてもよく、例えば、１番最初に記載された訳語を残す、訳語の出現頻度が最も高いものを残す、などの方法を用いることができる。 (C) There is no unique correspondence between the source language and the target language (for example, there are multiple translations for one word in the source language)
For parallel translations that fall under (C), the parallel translations will be deleted. Alternatively, from a plurality of translated words, leave one translated word and delete the others, so that the source language phrase and the target language phrase are in a one-to-one relationship. Any method can be used to leave one translated word and delete the others from multiple translated words. For example, keep the first translated word listed, or select the one with the highest frequency of occurrence. You can use a method such as leaving it.
　例えば、「原言語：ｃｏｍｐｕｔｅｒ、目的言語：計算機、コンピュータ」の対訳は（Ｃ）に該当し、この場合、例えば、この対訳を削除するか、「目的言語：ｃｏｍｐｕｔｅｒ、原言語：計算機」のようにして、原言語語句と目的言語語句とが１対１になるようにする。 For example, the parallel translation of "source language: computer, target language: computer, computer" corresponds to (C). so that there is a one-to-one relationship between the source language phrase and the target language phrase.
　＜抽出部１２０：分割部１２２＞
　分割部１２２は、原言語文と、対訳辞書の原言語訳語を形態素単位に分割（トークナイズ）して、形態素境界に所定の記号（例：空白、"／"）を挿入する。この分割単位は、後に翻訳する際に行う分割処理の分割単位とは異なってもよい。 <Extraction unit 120: Division unit 122>
The dividing unit 122 divides (tokenizes) the source language sentence and the source language translation of the bilingual dictionary into morpheme units, and inserts predetermined symbols (eg, spaces, "/") at morpheme boundaries. This division unit may be different from the division unit of division processing performed later when translating.
　例えば、原言語文が"その限りではない"とすると、分割部１２２による処理後の原言語文は、"その／限り／で／は／ない"となる。 For example, if the source language sentence is "That is not the case," the source language sentence after processing by the dividing unit 122 will be "As long as it is, it is not."
　＜抽出部１２０：制約語句抽出部１２３＞
　制約語句抽出部１２３は、原言語文に含まれる語句に対応する対訳を抽出し、抽出した対訳を用いて制約語句リストを作成する。具体的な制約語句抽出方法の例を以下に説明する。なお、原言語文に含まれる語句に対応する制約語句を抽出できる方法であれば、辞書の形式や検索方法等は、以下で説明する方法に限定されず、別の方法であってもよい。 <Extraction unit 120: Constraint phrase extraction unit 123>
The constraint phrase extraction unit 123 extracts bilingual translations corresponding to the phrases included in the source language sentence, and creates a constraint phrase list using the extracted bilingual translations. A specific example of the constraint phrase extraction method will be described below. Note that the format of the dictionary, the search method, etc. are not limited to the method described below, and other methods may be used as long as the method can extract the constraint phrases corresponding to the words included in the source language sentence.
　本例では、対訳辞書として、原言語訳語の文字単位で、Ｔｒｉｅ木とよばれるデータ構造を使用して表現したものを用いることとする。 In this example, we will use a bilingual dictionary that is expressed in units of characters of the source language translation using a data structure called a Trie tree.
　制約語句抽出部１２３は、原言語文の文頭から、対訳辞書の原言語訳語の集合を対象に、前方一致検索を進める。原言語文に含まれる語句にマッチする原言語訳語を含む対訳（ペア）が見つかったら、その目的語訳語を制約語句として抽出する。前方一致検索の際には原言語訳語の語句の長さが最長となる対訳を選択する。 The constraint word extraction unit 123 performs a prefix match search from the beginning of the source language sentence to a set of source language translated words in the bilingual dictionary. When a pair of translations containing a source language translation that matches a word in the source language sentence is found, the target translation is extracted as a constraint word. When performing a prefix match search, the parallel translation with the longest length of the source language translation is selected.
　例えば、原言語文において、分割部１２２による形態素解析により分割された結果、３つの語句、つまり、「ＡＢＣ／ＧＨＩ／ＸＹＺ」が得られたとする。ここでのＡ、Ｂ、Ｃなどは文字である。制約語句抽出部１２３が「ＡＢＣ／ＧＨＩ／ＸＹＺ」を用いて対訳辞書の原言語訳語に対して検索を行うと、「ＡＢＣ／ＧＨＩ／ＸＹＺ」の文頭（あるいは前方）から一致するものがマッチする。 For example, assume that the source language sentence is divided by morphological analysis by the dividing unit 122, resulting in three words, ie, "ABC/GHI/XYZ". A, B, C, etc. here are letters. When the constraint phrase extraction unit 123 searches the source language translation of the bilingual dictionary using "ABC/GHI/XYZ", a match is found starting from the beginning (or front) of the sentence of "ABC/GHI/XYZ". .
　上記の検索の結果、例えば、「ＡＢ」、「ＡＢＣ」、「ＡＢＣＧ」「ＡＢＣ／ＧＨＩ」の４つの語句がマッチする場合でも、後述するように「ＡＢ」や「ＡＢＣＧ」は、形態素の単位が合わないため、マッチしないようにすることができる。この場合、残った「ＡＢＣ」「ＡＢＣ／ＧＨＩ」のうち、原言語訳語の語句の長さが最長である「ＡＢＣ／ＧＨＩ」とペアの目的語訳語を制約語句として抽出する。その後、「ＡＢＣ／ＧＨＩ」より後ろの部分である「ＸＹＺ」を用いて同様の処理を実行する。 As a result of the above search, for example, even if the four words "AB", "ABC", "ABCG", and "ABC/GHI" match, "AB" and "ABCG" are the units of morphemes, as will be explained later. Since they do not match, it is possible to prevent them from matching. In this case, from among the remaining "ABC" and "ABC/GHI", the target translated word paired with "ABC/GHI", which has the longest length of the source language translated word, is extracted as a constraint word. Thereafter, similar processing is performed using "XYZ" which is the part after "ABC/GHI".
　本実施の形態のように、分割部１２２により原言語文と対訳辞書の原言語訳語とを事前に形態素（単位情報の例）に分割し、形態素境界を考慮して検索を行うことで、分割単位の一致しない語句を誤抽出すること防ぐことができる。これは、原言語が日本語等のように、分かち書きされない言語の場合には特に効果が大きい。例えば原言語訳語である"はな"（花）が、原言語文の"その／限り／で／は／ない"にマッチすることを防ぐことができる。つまり、"はな"が、"は／な"にマッチできなくなる。 As in this embodiment, the dividing unit 122 divides the source language sentence and the source language translation of the bilingual dictionary into morphemes (an example of unit information) in advance, and performs a search taking morpheme boundaries into consideration. This can prevent incorrect extraction of words whose units do not match. This is particularly effective when the source language is a language that does not have separate lines, such as Japanese. For example, it is possible to prevent the source language translated word "hana" (flower) from matching the source language sentence "so/as long as/de/ha/nai". In other words, "hana" cannot match "ha/na".
　なお、ここで説明するような前方一致、最長一致、及び、単語分割の実施は、よりノイズが少なく、曖昧性を削減した制約語句抽出を実現するための手段の一例である。曖昧性を解消できる別の手段を用いてもよい。 Note that the implementation of prefix match, longest match, and word division as described here are examples of means for realizing constraint phrase extraction with less noise and reduced ambiguity. Other means of disambiguation may be used.
　例えば、分割部１２２で形態素解析を行う際に、品詞、原型、語幹、活用形、読み（発音）など曖昧性解消に必要な情報を分割した語句に付与し、当該付与した情報も使用してマッチングを行う。すなわち、マッチングの際に文字列だけでなく、その品詞等の付属情報を使うことで、例えば、原言語文にある"ｉｎ"という文字が、原言語訳語である前置詞ｉｎと名詞ｉｎｎ（宿）の両方にマッチする状況に置いて、その曖昧性を解消することができる。マッチング時に、曖昧性の解消を行うことが、翻訳精度の向上に重要な要素となる。 For example, when performing morphological analysis in the segmentation unit 122, information necessary for disambiguation, such as part of speech, prototype, stem, conjugation, and reading (pronunciation), is attached to the divided words and the attached information is also used. Perform matching. In other words, by using not only the character string but also attached information such as its part of speech during matching, for example, the character "in" in the source language sentence can be matched with the preposition in, which is the source language translation, and the noun inn (inn). The ambiguity can be resolved by placing the situation in which both of the above match. Resolving ambiguity during matching is an important element in improving translation accuracy.
　＜抽出部１２０の構成の他の例＞
　なお、抽出部１２０を、図４に示す構成に代えて、図５に示す構成としてもよい。図５に示す構成では、対訳辞書のフィルタリングを行うのではなく、制約語句抽出部１２１で抽出した制約語句に対して、フィルタリング部１２１がフィルタリングを行う。 <Other examples of the configuration of the extraction unit 120>
Note that the extraction unit 120 may have the configuration shown in FIG. 5 instead of the configuration shown in FIG. 4. In the configuration shown in FIG. 5, the filtering section 121 performs filtering on the constraint phrases extracted by the constraint phrase extraction section 121, instead of filtering the bilingual dictionary.
　フィルタリング処理については、前述したフィルタリング部１２１による処理と同様である。ただし、「対訳」は「制約語句」と読み替える。具体的には、フィルタリング部１２１は、以下（Ａ）～（Ｃ）に当てはまる制約語句を、制約語句抽出部１２３による抽出結果から削除する。ただし、フィルタリング部１２１は、（Ａ）～（Ｃ）の全部を実施することは必須ではなく、（Ａ）～（Ｃ）のうちの少なくとも１つを実施することとしてもよい。また、（Ａ）～（Ｃ）以外のルールを用いてもよい。特に、後述する変形例１，２を行う場合には以下（Ｃ）の処理はスキップすることとしてもよい。 The filtering process is similar to the process by the filtering unit 121 described above. However, "parallel translation" should be read as "constraint phrase". Specifically, the filtering unit 121 deletes constraint phrases that apply to (A) to (C) below from the extraction results by the constraint phrase extraction unit 123. However, the filtering unit 121 does not necessarily need to implement all of (A) to (C), but may implement at least one of (A) to (C). Further, rules other than (A) to (C) may be used. In particular, when performing Modifications 1 and 2, which will be described later, the following process (C) may be skipped.
　（Ａ）名詞／名詞句以外の語句を含む制約語句（動詞は活用があるためにこれを削除する）
　（Ｂ）長さが１の語句からなる制約語句
　（Ｃ）原言語と目的言語との間の対応関係に一意性が無いもの（例えば、原言語側の１つの語句に対して複数の制約語句が存在するもの）
　（Ｃ）を実施する場合、原言語側の１つの語句に対して複数の制約語句が存在する場合、例えば、当該複数の制約語句を削除する、あるいは、複数の制約語句から、１つの制約語句を残して他を削除することで、原言語語句と目的言語語句とが１対１になるようにする。 (A) Restricted phrases that include words other than nouns/noun phrases (verbs are deleted because they have conjugations)
(B) Constraint phrases consisting of a word with a length of 1. (C) Constraints in which there is no unique correspondence between the source language and the target language (for example, multiple constraint phrases for one word in the source language) exists)
When implementing (C), if there are multiple constraint phrases for one word in the source language, for example, delete the multiple constraint phrases, or select one constraint phrase from the multiple constraint phrases. By leaving the ``words'' and deleting the others, the source language phrases and target language phrases become one-to-one.
　（抽出部１２０、生成装置１００の他の構成例について）
　抽出部１２０は、生成装置１００とは独立の単独の装置であってもよい。この単独の装置を抽出装置と呼んでもよい。なお、生成装置１００に含まれている抽出部１２０についても、当該抽出部１２０を抽出装置と呼んでもよい。また、抽出部１２０を有する生成装置１００を抽出装置と呼んでもよい。また、抽出部１２０と抽出装置はいずれも、後述する実施例における表示情報生成部１７０と修正部１８０の両方又はいずれか１つを備えてもよい。 (Regarding other configuration examples of the extraction unit 120 and the generation device 100)
The extraction unit 120 may be a single device independent of the generation device 100. This single device may also be referred to as an extraction device. Note that the extraction unit 120 included in the generation device 100 may also be referred to as an extraction device. Further, the generation device 100 having the extraction unit 120 may be referred to as an extraction device. Furthermore, both the extraction unit 120 and the extraction device may include both or one of the display information generation unit 170 and the correction unit 180 in the embodiment described later.
　抽出部１２０が、生成装置１００とは独立の単独の装置として構成される場合において、生成装置１００は、抽出部１２０を含まないこととしてもよい。この場合の生成装置１００の構成を図６に示す。図６の構成においては、抽出装置により生成された制約語句リストが生成装置１００に入力される。ただし、図６の構成において、抽出装置により生成された制約語句リストではない制約語句リスト（例：ノイズが多く含まれる制約語句リスト）が生成装置１００に入力されることとしてもよい。 In the case where the extraction unit 120 is configured as a single device independent of the generation device 100, the generation device 100 may not include the extraction unit 120. The configuration of the generation device 100 in this case is shown in FIG. In the configuration of FIG. 6, the constraint phrase list generated by the extraction device is input to the generation device 100. However, in the configuration of FIG. 6, a constraint phrase list that is not the constraint phrase list generated by the extraction device (eg, a constraint phrase list that includes a lot of noise) may be input to the generation device 100.
　図６における入力生成部１３０、系列生成部１４０、及びリランキング部１５０の動作は、図２における入力生成部１３０、系列生成部１４０、及びリランキング部１５０の動作と同じである。 The operations of the input generation section 130, sequence generation section 140, and reranking section 150 in FIG. 6 are the same as the operations of the input generation section 130, sequence generation section 140, and reranking section 150 in FIG.
　（入力生成部１３０）
　次に、入力生成部１３０について説明する。入力生成部１３０は、制約語句リストを入力として受け取り、制約語句リストに含まれる語句についての部分集合の全ての要素をそれぞれ語彙制約とする。ただし、全ての要素のうちの一部の要素をそれぞれ語彙制約としてもよい。 (Input generation unit 130)
Next, the input generation unit 130 will be explained. The input generation unit 130 receives the constraint phrase list as input, and sets all elements of a subset of the words included in the constraint phrase list as vocabulary constraints. However, some elements among all the elements may be subjected to vocabulary constraints.
　最後に、入力生成部１３０は、抽出部１２０に入力された原言語文に対応する語彙制約として、上記の語彙制約を出力する。具体例を以下に示す。 Finally, the input generation unit 130 outputs the above vocabulary constraints as the vocabulary constraints corresponding to the source language sentence input to the extraction unit 120. A specific example is shown below.
　入力生成部１３０に、制約語句リストとして、｛Ａ，Ｂ，Ｃ｝が入力されるとする。ここで、Ａ、Ｂ、Ｃはそれぞれ制約語句である。 Assume that {A, B, C} is input to the input generation unit 130 as a constraint phrase list. Here, A, B, and C are each constraint phrases.
　入力生成部１３０は、｛Ａ，Ｂ，Ｃ｝の部分集合の要素として、｛｝，｛Ａ｝，｛Ｂ｝，｛Ｃ｝，｛Ａ，Ｂ｝，｛Ａ，Ｃ｝，｛Ｂ，Ｃ｝，｛Ａ，Ｂ，Ｃ｝を抽出し、これらのそれぞれを語彙制約として出力する。 The input generation unit 130 generates {}, {A}, {B}, {C}, {A, B}, {A, C}, {B, C}, {A, B, C} and output each of them as a vocabulary constraint.
　なお、｛｛｝，｛Ａ｝，｛Ｂ｝，｛Ｃ｝，｛Ａ，Ｂ｝，｛Ａ，Ｃ｝，｛Ｂ，Ｃ｝，｛Ａ，Ｂ，Ｃ｝｝は制約語彙集合であり、１つの｛...｝が、１つの語彙制約である。 Note that {{}, {A}, {B}, {C}, {A, B}, {A, C}, {B, C}, {A, B, C}} is a constraint vocabulary set. , one {...} is one vocabulary constraint.
　制約語句のリストＣの部分集合から作成される語彙制約は２｜Ｃ｜個となり、後述するように、その各語彙制約からそれぞれ複数の翻訳文の候補が得られる。 There are 2 |C| vocabulary constraints created from a subset of the list C of constraint phrases, and as will be described later, a plurality of translation sentence candidates are obtained from each of the vocabulary constraints.
　（系列生成部１４０）
　次に、系列生成部１４０について説明する。系列生成部１４０は、モデルＤＢ３００から読み出した、学習済みの機械翻訳モデルを保持しているとする。また、系列生成部１４０は、下記の処理を語彙制約の数（語彙制約集合の要素数分）だけ繰り返す。例えば、語彙制約集合が｛｛｝，｛Ａ｝，｛Ｂ｝，｛Ｃ｝，｛Ａ，Ｂ｝，｛Ａ，Ｃ｝，｛Ｂ，Ｃ｝，｛Ａ，Ｂ，Ｃ｝｝であるとすると、８回繰り返す。 (Sequence generation unit 140)
Next, the sequence generation unit 140 will be explained. It is assumed that the series generation unit 140 holds a trained machine translation model read from the model DB 300. Furthermore, the sequence generation unit 140 repeats the following process by the number of lexical constraints (the number of elements in the lexical constraint set). For example, the vocabulary constraint set is {{}, {A}, {B}, {C}, {A, B}, {A, C}, {B, C}, {A, B, C}} Then, repeat 8 times.
　系列生成部１４０は、入力文（原言語文）と、語彙制約を入力として受け取る。系列生成部１４０は、語彙制約付き機械翻訳の既存手法を適用することにより、機械翻訳モデルを用いて翻訳文（目的言語文）を生成する。ここでは、複数の翻訳文を翻訳文候補（目的言語文候補）として生成する。また、翻訳文候補には、翻訳文としてのスコアが付けられている。 The sequence generation unit 140 receives an input sentence (source language sentence) and vocabulary constraints as input. The sequence generation unit 140 generates a translated sentence (target language sentence) using a machine translation model by applying an existing method of machine translation with vocabulary constraints. Here, a plurality of translated sentences are generated as translated sentence candidates (target language sentence candidates). In addition, a translation sentence candidate is given a score as a translation sentence.
　語彙制約付き機械翻訳の既存手法として、どの手法を用いてもよいが、例えば、ＬｅＣＡあるいはＬｅＣＡ＋ＬＣＤを使用することができる。ＬｅＣＡは非特許文献１に開示されており、ｓｏｆｔ法とも呼ばれる。ＬｅＣＡ＋ＬＣＤは、前述した参考文献１に開示されており、ｈａｒｄ法とも呼ばれる。 Any existing method for machine translation with vocabulary constraints may be used; for example, LeCA or LeCA+LCD may be used. LeCA is disclosed in Non-Patent Document 1 and is also called the soft method. LeCA+LCD is disclosed in Reference 1 mentioned above, and is also called the hard method.
　系列生成部１４０は、生成した複数の翻訳文候補を出力する。一例として、系列生成部１４０は、スコアの高い順に、予め定めた数の翻訳文候補を出力する。「予め定めた数」は１であってもよい。つまり、最も高いスコアの翻訳文のみを出力してもよい。ここでは、例えば、１つの語彙制約あたり３０個の翻訳文候補を出力する。 The series generation unit 140 outputs a plurality of generated translation sentence candidates. As an example, the sequence generation unit 140 outputs a predetermined number of translation sentence candidates in descending order of scores. The "predetermined number" may be one. In other words, only the translated sentence with the highest score may be output. Here, for example, 30 translation candidates are output for each vocabulary constraint.
　＜系列生成部１４０の構成例＞
　図７に、系列生成部１４０の構成例を示す。図７に示すように、系列生成部１４０は、系列変換部１４１と探索部１４２を有する。 <Configuration example of sequence generation unit 140>
FIG. 7 shows a configuration example of the sequence generation unit 140. As shown in FIG. 7, the sequence generation section 140 includes a sequence conversion section 141 and a search section 142.
　なお、翻訳文生成のためにｓｏｆｔ法を使う場合は、系列変換部１４１で語彙制約の情報を使用し、ｈａｒｄ法を使う場合は、ｈａｒｄ法の種類によって、系列変換部１４１で語彙制約を使う場合もあれば使わない場合もある。語彙制約についての系列変換部１４２への入力の矢印を点線で示している。ｈａｒｄ法のうち、前述したＬｅＣＡ＋ＬＣＤでは系列変換部１４１で語彙制約の情報を使用する。以下では、ＬｅＣＡ＋ＬＣＤを想定した構成／動作を説明する。 Note that when using the soft method to generate translated sentences, the sequence conversion unit 141 uses lexical constraint information, and when using the hard method, the sequence conversion unit 141 uses the lexical constraints depending on the type of hard method. Sometimes it's used, sometimes it's not. Arrows for inputting vocabulary constraints to the series conversion unit 142 are indicated by dotted lines. Among the hard methods, in the aforementioned LeCA+LCD, information on vocabulary constraints is used in the sequence conversion unit 141. The configuration/operation assuming LeCA+LCD will be described below.
　系列変換部１４１において、機械翻訳モデルとして、図８に示すように、符号化器（エンコーダ）と復号化器（デコーダ）とを有する、一般的なエンコーダ‐デコーダモデル（例えば、Transformer）をベースとするモデルを使用することができる。ただし、本発明は、エンコーダ‐デコーダモデル以外のモデルを使用しても実施可能である。 The sequence conversion unit 141 uses a general encoder-decoder model (for example, Transformer) as a machine translation model, which has an encoder and a decoder, as shown in FIG. model can be used. However, the invention can be implemented using models other than the encoder-decoder model.
　系列変換部１４１は、原言語文及び語彙制約を入力として受け取り、まず、語彙制約を用いて原言語文を拡張することにより、語彙制約の情報を付け加えた入力系列を作成し、それを機械翻訳モデルへの入力とする。 The sequence conversion unit 141 receives the source language sentence and vocabulary constraints as input, first expands the source language sentence using the vocabulary constraints to create an input sequence with information on the vocabulary constraints added, and then machine translates it. Use as input to the model.
　より具体的には、上記の拡張において、系列変換部１４１は、入力系列である原言語文Ｘと、各制約語句Ｃｉとを、下記のように＜ｓｅｐ＞という特別な区切りを示す文字列を介して結合（連結）することで語彙制約付きの入力系列を作成する。＜ｅｏｓ＞は文の終わりを表す文字列である。 More specifically, in the above expansion, the sequence conversion unit 141 converts the source language sentence X, which is the input sequence, and each constraint word C i into a character string indicating a special delimiter called <sep> as shown below. Create an input sequence with lexical constraints by combining (concatenating) via . <eos> is a character string representing the end of a sentence.
　［Ｘ，＜ｓｅｐ＞，Ｃ１，＜ｓｅｐ＞，Ｃ２，…，ＣＮ，＜ｅｏｓ＞］
　系列変換部１４１は、拡張した入力系列を機械翻訳モデルへの入力として、文を生成する。より具体的には、出力系列を構成し得る語の集合における各語の確率を出力する。 [X, <sep>, C 1 , <sep>, C 2 ,..., C N , <eos>]
The sequence conversion unit 141 generates a sentence by using the expanded input sequence as input to a machine translation model. More specifically, the probability of each word in a set of words that can constitute an output sequence is output.
　探索部１４２は、機械翻訳モデルにおける復号化器の出力確率を用いて、入力系列が与えられたときの生成確率が最大になる出力系列（の近似解）を探索する。探索部１４２は、ビームサーチを基としたグリッドビームサーチの手法を用いることで、出力系列が制約語彙をすべて満たすことを保証することを可能としている。 The search unit 142 uses the output probability of the decoder in the machine translation model to search for (an approximate solution of) the output sequence that maximizes the generation probability when the input sequence is given. The search unit 142 uses a grid beam search method based on beam search to ensure that the output sequence satisfies all of the constraint vocabulary.
　なお、探索部１４２がグリッドビームサーチを用いて探索を行うことは一例である。制約語句を含むように、語彙制約付き探索を行う処理方法であればどのような処理方法を使用してもよい。 Note that the searching unit 142 searching using grid beam search is an example. Any processing method may be used as long as it performs a lexically constrained search so as to include the constraint word/phrase.
　（リランキング部１５０）
　続いて、リランキング部１５０について説明する。リランキング部１５０は、系列生成部１４０で生成された１つ又は複数の翻訳文候補を入力として受け取る。例えば、系列生成部１４０が、１つの語彙制約あたりに３０個の翻訳文候補を生成するとして、８つの語彙制約があるとすると、リランキング部１５０は、その８つの語彙制約における１つの語彙制約ごとに３０文の翻訳文候補を入力として受け取る。 (Reranking section 150)
Next, the reranking unit 150 will be explained. The reranking unit 150 receives as input one or more translated sentence candidates generated by the series generation unit 140. For example, if the series generation unit 140 generates 30 translation candidates per lexical constraint and there are 8 lexical constraints, the reranking unit 150 generates 30 translation candidates per lexical constraint. Each time, 30 translation candidates are received as input.
　次に、リランキング部１５０は、入力文（原言語文）を用いて各翻訳文候補に対してスコアを計算し、スコアが最も良い翻訳文候補を、最終的な翻訳文として出力する。ここでスコアが最も高いものに絞り込まず、翻訳文とそのスコア全てを（又は一部を）出力してもよい。これにより、出力部１６０は、スコアを使ってランキング形式で翻訳文をユーザに提示することができる。 Next, the reranking unit 150 calculates a score for each translated sentence candidate using the input sentence (source language sentence), and outputs the translated sentence candidate with the highest score as the final translated sentence. Here, instead of narrowing down to the one with the highest score, all (or some) of the translated sentences and their scores may be output. Thereby, the output unit 160 can present the translated sentences to the user in a ranking format using the scores.
　リランキング部１５０によるスコアの計算の方法については、翻訳文のスコアを計算できる方法であればどのような方法を使用してもよいが、例えば、下記の例１、例２の方法を使用することができる。 As for the method of calculating the score by the reranking unit 150, any method may be used as long as it can calculate the score of the translated sentence, but for example, the methods in Example 1 and Example 2 below may be used. be able to.
　例１：
　リランキング部１５０は、系列生成部１４０において翻訳に用いた機械翻訳モデルの出力する翻訳文候補の尤度をスコアとして使用する。 Example 1:
The reranking unit 150 uses, as a score, the likelihood of translated sentence candidates output by the machine translation model used for translation in the sequence generation unit 140.
　例２：
　リランキング部１５０は、文末から文頭へと翻訳文を生成するRight-to-Left翻訳タスクをEncoder-DecoderモデルであるTransformerで学習した機械翻訳モデルをリランキングモデルとして使用し、そのリランキングモデルで翻訳文候補を強制的に出力させた際の尤度をスコアとして使用する。翻訳文候補を強制的に出力させることを、翻訳文候補を用いてforced decodingする、と言い換えてもよい。 Example 2:
The reranking unit 150 uses a machine translation model learned by Transformer, which is an encoder-decoder model, for a Right-to-Left translation task that generates a translated sentence from the end of a sentence to the beginning of a sentence. The likelihood when translation candidates are forced to be output is used as a score. Forcibly outputting translated sentence candidates may be rephrased as forced decoding using translated sentence candidates.
　すなわち、リランキングモデルのエンコーダに原言語文を入力し、リランキングモデルのデコーダへは、スコア（尤度）を評価したい翻訳文候補の単語を順次入力する。 That is, the source language sentence is input to the encoder of the reranking model, and the words of the translation sentence candidate whose score (likelihood) is to be evaluated are sequentially input to the decoder of the reranking model.
　なお、例１と例２において、機械翻訳モデルが出力する尤度は、もっともらしさを示す値であればどのような値であってもよい。機械翻訳モデルが出力する尤度は、確率であってもよいし、確率以外の値であってもよい。 Note that in Examples 1 and 2, the likelihood output by the machine translation model may be any value as long as it indicates plausibility. The likelihood output by the machine translation model may be a probability or a value other than probability.
　また、リランキング部１５０は、例１の尤度と例２の尤度の両方を使用してリランキングのスコアを計算してもよい。例えば、例１の尤度と例２の尤度の平均をリランキングのスコアとしてもよい。 Furthermore, the reranking unit 150 may calculate the reranking score using both the likelihood of Example 1 and the likelihood of Example 2. For example, the average of the likelihood of Example 1 and the likelihood of Example 2 may be used as the reranking score.
　（変形例１）
　次に、変形例１を説明する。変形例１では、抽出部１２０により生成される制約語句リストとして、１つの原言語の語句に対し、複数の目的言語語句が対応するものを用いることができる。このような制約語句リストを、複数訳語を許容する制約語句リストと呼んでもよい。例えば、抽出部１２０のフィルタリング部で（Ｃ）の手順を行わない場合、こうした制約語句リストが生成される場合がある。 (Modification 1)
Next, modification 1 will be explained. In the first modification, the constraint phrase list generated by the extraction unit 120 can be one in which a plurality of target language phrases correspond to one source language phrase. Such a constraint phrase list may be called a constraint phrase list that allows multiple translations. For example, if the filtering unit of the extraction unit 120 does not perform the step (C), such a constraint phrase list may be generated.
　例えば、ある原言語の語句に対する複数の目的言語語句として、ＡとＡ´があり、これらと、Ｂ、Ｃを含む「Ａ，Ａ´，Ｂ，Ｃ」が、抽出部１２０により制約語句リストの複数要素として生成されたとする。例えば、原言語文の語句がｃｏｍｐｕｔｅｒで、目的言語文の語句が計算機、コンピュータである場合、ＡとＡ´は、計算機とコンピュータに相当する。 For example, there are A and A' as a plurality of target language phrases for a certain source language phrase, and "A, A', B, C" including these and B and C are added to the constraint phrase list by the extraction unit 120. Suppose it is generated as multiple elements. For example, if the word in the source language sentence is "computer" and the word in the target language sentence is "computer", A and A' correspond to "computer" and "computer".
　ここでは、このような複数要素を有する制約語句リストを｛｛Ａ，Ａ´｝，｛Ｂ｝，｛Ｃ｝｝と表現する。 Here, such a constraint phrase list having multiple elements is expressed as {{A, A'}, {B}, {C}}.
　抽出部１２０から｛｛Ａ，Ａ´｝，｛Ｂ｝，｛Ｃ｝｝を入力された入力生成部１３０は、｛｝，｛Ａ｝，｛Ｂ｝，｛Ｃ｝，｛Ａ，Ｂ｝，｛Ａ，Ｃ｝，｛Ｂ，Ｃ｝，｛Ａ，Ｂ，Ｃ｝に加え、｛Ａ´｝，｛Ａ´，Ｂ｝，｛Ａ´，Ｃ｝，｛Ａ´，Ｂ，Ｃ｝も語彙制約として生成する。 The input generation unit 130 that receives {{A, A′}, {B}, {C}} from the extraction unit 120 generates {}, {A}, {B}, {C}, {A, B} , {A, C}, {B, C}, {A, B, C}, as well as {A'}, {A', B}, {A', C}, {A', B, C} is also generated as a vocabulary constraint.
　入力生成部１３０は、生成した複数の語彙制約のそれぞれを、系列生成部１４０に入力する。 The input generation unit 130 inputs each of the plurality of generated vocabulary constraints to the sequence generation unit 140.
　系列生成部１４０は、｛｝，｛Ａ｝，｛Ｂ｝，｛Ｃ｝，｛Ａ，Ｂ｝，｛Ａ，Ｃ｝，｛Ｂ，Ｃ｝，｛Ａ，Ｂ，Ｃ｝，｛Ａ´｝，｛Ａ´，Ｂ｝，｛Ａ´，Ｃ｝，｛Ａ´，Ｂ，Ｃ｝の１２個の語彙制約のそれぞれを語彙制約として使用することにより、１２回の語彙制約付き機械翻訳を行い、翻訳文候補を得る。例えば、仮に、１つの語彙制約に対して１つの翻訳文候補を生成する場合、１２個の翻訳文候補が得られることになる。 The sequence generation unit 140 generates {}, {A}, {B}, {C}, {A, B}, {A, C}, {B, C}, {A, B, C}, {A' By using each of the 12 vocabulary constraints of }, {A', B}, {A', C}, and {A', B, C} as vocabulary constraints, machine translation with vocabulary constraints is performed 12 times. and obtain translation candidates. For example, if one translated sentence candidate is generated for one vocabulary constraint, 12 translated sentence candidates will be obtained.
　語彙制約付き機械翻訳を行った後、これまでに説明した方法で、リランキング部１５０が、リランキング処理を行って、例えば、スコアの最も高い翻訳文候補を、最終的な翻訳文として出力する。 After performing the machine translation with vocabulary constraints, the reranking unit 150 performs the reranking process using the method described above, and outputs, for example, the translated sentence candidate with the highest score as the final translated sentence. .
　（変形例２）
　次に、変形例２を説明する。変形例２においても、抽出部１２０により生成される制約語句リストとして、１つの原言語の語句に対し、複数の目的言語語句が対応するものを用いることができる。 (Modification 2)
Next, modification 2 will be explained. Also in the second modification, the constraint phrase list generated by the extraction unit 120 can be one in which a plurality of target language phrases correspond to one source language phrase.
　変形例２では、系列生成部１４０の探索部１４２における翻訳文の探索処理において、１つの制約語句の表現型が複数存在することを許容して探索を行うようにしてもよい。つまり、各制約語句の候補からそれぞれ１つの要素を満たすように探索を行うようにしてもよい。具体的には下記のとおりである。 In the second modification, in the translation search process in the search unit 142 of the sequence generation unit 140, the search may be performed by allowing the existence of a plurality of expression types of one constraint phrase. In other words, the search may be performed such that one element is satisfied from each constraint word candidate. Specifically, the details are as follows.
　変形例２においても、ある原言語の語句に対する複数の目的言語語句として、ＡとＡ´があり、これらと、Ｂ、Ｃを含む「Ａ，Ａ´，Ｂ，Ｃ」が、抽出部１２０により制約語句リストの複数要素として生成されたとする。ここでは、制約語句リストとしては、｛Ａ，Ｂ，Ｃ｝が生成され、Ａは、Ａ´でもよいことを示す情報が抽出部１２０から入力生成部１３０に入力されるとする。あるいは、制約語句リストとしては、｛Ａ，Ａ´，Ｂ，Ｃ｝が生成され、ＡとＡ´はどちらでもよいことを示す情報が抽出部１２０から入力生成部１３０に入力されることとしてもよい。なお、上記では、１つの制約語句に対して、２つの曖昧性を許容する形式の例を示しているが、１つの制約語句に対して、３つ以上の曖昧性を許容する形式を用いてもよい。 In Modified Example 2 as well, there are A and A' as a plurality of target language phrases for a certain source language phrase, and "A, A', B, C" including these, B, and C are extracted by the extraction unit 120. Suppose that it is generated as multiple elements of a constraint word list. Here, it is assumed that {A, B, C} is generated as the constraint phrase list, and information indicating that A may be replaced by A' is input from the extraction unit 120 to the input generation unit 130. Alternatively, {A, A', B, C} may be generated as the constraint phrase list, and information indicating that either A or A' is acceptable may be input from the extraction unit 120 to the input generation unit 130. good. Note that although the above example shows an example of a format that allows two ambiguities for one constraint phrase, it is also possible to use a format that allows three or more ambiguities for one constraint phrase. Good too.
　例えば、Ａについて、３つの曖昧性を許容する場合、制約語句リストとしては、｛Ａ，Ｂ，Ｃ｝が生成され、Ａは、Ａ´でもよいし、Ａ´´でもよいことを示す情報が抽出部１２０から入力生成部１３０に入力されるとする。あるいは、制約語句リストとしては、｛Ａ，Ａ´，Ａ´´，Ｂ，Ｃ｝が生成され、ＡとＡ´とＡ´´はどれでもよいことを示す情報が抽出部１２０から入力生成部１３０に入力されることとしてもよい。 For example, if three ambiguities are allowed for A, {A, B, C} is generated as the constraint word list, and information indicating that A may be A' or A'' is generated. It is assumed that the input is input from the extraction unit 120 to the input generation unit 130. Alternatively, {A, A', A'', B, C} is generated as a constraint word list, and information indicating that any of A, A', and A'' is acceptable is sent from the extraction unit 120 to the input generation unit. 130 may be input.
　制約語句リスト｛Ａ，Ｂ，Ｃ｝に対し、ＡはＡ´でもよい場合、入力生成部１３０は、｛｝，｛｛Ａ，Ａ´｝｝，｛｛Ｂ｝｝，｛｛Ｃ｝｝，｛｛Ａ，Ａ´｝，｛Ｂ｝｝，｛｛Ａ，Ａ´｝，｛Ｃ｝｝，｛｛Ａ，Ａ´｝，｛Ｂ｝，｛Ｃ｝｝の７個の語彙候補制約を生成する。なお、変形例２では、ある原言語語句に対して、複数の目的言語語句（例：Ａ，Ａ´）が対応する場合があることから、訳語に曖昧性があり、制約として用いる語彙が確定していないので、語彙制約に代えて「語彙候補制約」と呼んでいる。すなわち、「語彙候補制約」は、曖昧性を保持した語彙制約である。なお、上記の語彙候補制約の表現形式は一例である。ＡでもよいしＡ´でもよい、ということが表現できれば、表現形式として上記の表現形式以外の表現形式を用いてもよい。 For the constraint word list {A, B, C}, if A may be A', the input generation unit 130 generates {}, {{A, A'}}, {{B}}, {{C}} , {{A, A'}, {B}}, {{A, A'}, {C}}, {{A, A'}, {B}, {C}} seven vocabulary candidate constraints generate. In addition, in Modification 2, there is ambiguity in the translated word because there are cases where multiple target language words (e.g., A, A') correspond to a certain source language word, and the vocabulary used as a constraint is determined. Therefore, we call it ``vocabulary candidate constraint'' instead of lexical constraint. That is, the "vocabulary candidate constraint" is a vocabulary constraint that maintains ambiguity. Note that the expression format of the vocabulary candidate constraints described above is an example. As long as it can be expressed that either A or A' is acceptable, expression formats other than those described above may be used as the expression format.
　図９に示すように、系列生成部１４０は、原言語文とともに、語彙候補制約を入力として受け取る。系列生成部１４０は、｛｝，｛｛Ａ，Ａ´｝｝，｛｛Ｂ｝｝，｛｛Ｃ｝｝，｛｛Ａ，Ａ´｝，｛Ｂ｝｝，｛｛Ａ，Ａ´｝，｛Ｃ｝｝，｛｛Ａ，Ａ´｝，｛Ｂ｝，｛Ｃ｝｝の７個の語彙候補制約のそれぞれを語彙候補制約として使用して、７回の語彙制約付き機械翻訳を行い、翻訳文候補を得る。例えば、仮に、１つの語彙候補制約に対して１つの翻訳文候補を生成する場合、７個の翻訳文候補が得られることになる。 As shown in FIG. 9, the sequence generation unit 140 receives the source language sentence and the vocabulary candidate constraints as input. The sequence generation unit 140 generates {}, {{A, A′}}, {{B}}, {{C}}, {{A, A′}, {B}}, {{A, A′} , {C}}, {{A, A'}, {B}, {C}}, each of the seven vocabulary candidate constraints was used as a vocabulary candidate constraint, and machine translation with vocabulary constraints was performed seven times. , obtain translation candidates. For example, if one translated sentence candidate is generated for one vocabulary candidate constraint, seven translated sentence candidates will be obtained.
　語彙制約付き機械翻訳を行った後、これまでに説明した方法で、リランキング部１５０が、リランキング処理を行って、例えば、スコアの最も高い翻訳文候補を、最終的な翻訳文として出力する。 After performing the machine translation with vocabulary constraints, the reranking unit 150 performs the reranking process using the method described above, and outputs, for example, the translated sentence candidate with the highest score as the final translated sentence. .
　｛Ａ，Ａ´｝を含む語彙候補制約を用いる場合、系列生成部１４０の探索部１４２では、Ａという単語がＡ´でもよいとして、探索を実行する。つまり、曖昧性を考慮した探索を実行する。探索には、例えばhttps://aclanthology.org/D17-1098/に開示されている参考文献２「Peter Anderson, Basura Fernando, Mark Johnson, and Stephen Gould. 2017. Guided Open Vocabulary Image Captioning with Constrained Beam Search. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 936-945, Copenhagen, Denmark. Association for Computational Linguistics」の手法を用いることができる。この手法は、「曖昧性を考慮した探索」の手法の例である。 When using a vocabulary candidate constraint including {A, A'}, the search unit 142 of the sequence generation unit 140 performs a search assuming that the word A may be A'. In other words, a search is performed that takes ambiguity into account. For searching, for example, reference 2 “Peter Anderson, Basura Fernando, Mark Johnson, and Stephen Gould. 2017. Guided Open Vocabulary Image Captioning with Constrained Beam Search In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 936-945, Copenhagen, Denmark. Association for Computational Linguistics can be used. This method is an example of a "search considering ambiguity" method.
　参考文献２に開示されている手法では、ＡとＡ´のどちらでもよいという訳語の曖昧性を考慮した語彙制約付きビームサーチを行う。つまり、ビームサーチの際にＡとＡ´の曖昧性が解消される。 The method disclosed in Reference 2 performs a beam search with lexical constraints that takes into account the ambiguity of the translated word, which may be either A or A'. In other words, ambiguity between A and A' is resolved during beam search.
　なお、参考文献２に開示されている手法は、言語生成手法ではあるが、翻訳技術ではない。当該手法を翻訳のデコーディング時の探索に応用した従来技術は存在しない。 Note that the method disclosed in Reference 2 is a language generation method, but is not a translation technology. There is no prior art that applies this method to search during translation decoding.
　なお、これまでの実施の形態の説明、及び、変形例１及び２の説明において、ある原言語の語句に対する複数の目的言語語句（例：Ａ，Ａ´）は、計算機とコンピュータのような同義語だけでなく、Ｔｒｕｎｋに対する車のトランク、象の鼻、幹、幹線、など、同義語以外の語句であってもよい。探索部１４２における探索時には単語の意味を考慮していないので、ＡとＡ´が全く関係ない語句であってもよい。また、本発明に係る技術を翻訳以外のタスクで使用する際には、元の系列の語に対して、変換後系列における何が複数の語に対応するかについては、任意の基準で設定すればよい。 In addition, in the description of the embodiment so far and the description of Modifications 1 and 2, multiple target language phrases (e.g. A, A') for a certain source language phrase are synonyms such as "calculator" and "computer". In addition to the word ``Trunk'', it may be a word or phrase other than a synonym, such as ``car trunk'', ``elephant's trunk'', trunk, trunk line, etc. Since the meaning of words is not taken into account when searching in the search unit 142, A and A' may be completely unrelated words. Furthermore, when using the technology according to the present invention for tasks other than translation, arbitrary criteria can be used to determine which words in the converted series correspond to words in the original series. Bye.
　また、これまでの実施の形態の説明、及び、変形例１及び２の説明において、単語の変形（複数形、時制の変化、など）を考慮して形態素解析時に、原型でない語句を原型に変換するようにしてもよい。 In addition, in the description of the embodiment so far and the description of Modifications 1 and 2, words that are not the original form are converted to the original form during morphological analysis in consideration of word transformations (plural forms, changes in tense, etc.). You may also do so.
　一例として、対訳辞書が英－和である場合において、「corn－トウモロコシ、魚の目」という辞書エントリがあるとする。この場合、「We roasted corns over the charcoal.」という原言語文が生成装置１００に入力されたと想定する。この場合、抽出部１２０において、形態素単位でマッチングをかけたときに、cornsについては、cornが形態素として含まれるので、対訳辞書にマッチする。しかし、例えば、対訳辞書のエントリがfeetであり、入力文における形態素が、footである場合、マッチしない。そこで、footを原型に戻してマッチングをかけることにより、このような問題を解消できる。 As an example, suppose that the bilingual dictionary is English-Japanese, and there is a dictionary entry for "corn". In this case, it is assumed that the source language sentence "We roasted corns over the charcoal." is input to the generating device 100. In this case, when the extraction unit 120 performs matching on a morpheme basis, corns is matched in the bilingual dictionary because corn is included as a morpheme. However, for example, if the bilingual dictionary entry is feet and the morpheme in the input sentence is foot, there will be no match. Therefore, by returning the foot to its original form and performing matching, this problem can be resolved.
　（実施例）
　次に、より具体的な例として、これまでに説明した技術を用いた実施例を説明する。本実施例では、後述する表示部５００（表示、及び入力操作が可能な装置）において、制約語句の編集（修正、追加）を行って、その都度、その制約語句に対する目的言語文（翻訳文）を確認することが可能である。 (Example)
Next, as a more specific example, an embodiment using the techniques described above will be described. In this embodiment, each time a constraint phrase is edited (corrected or added) on a display unit 500 (device capable of display and input operations), which will be described later, a target language sentence (translation sentence) for the constraint phrase is created. It is possible to confirm.
　＜表示イメージ＞
　まず、表示部５００での表示イメージを、図１０を参照して説明する。図１０に示す例において、ユーザは、原言語文として、「分路巻線のみに補助巻線を持つ超電導単相単巻変圧器を試作した。」を入力し、「送信」を押す。 <Display image>
First, a display image on the display unit 500 will be described with reference to FIG. 10. In the example shown in FIG. 10, the user inputs "A prototype superconducting single-phase autotransformer having an auxiliary winding only in the shunt winding." is input as the source language sentence, and presses "Send".
　表示部５００には、入力した原言語文に対する複数の制約語句（制約語句リスト）が表示される。ここで制約語句として表示される語句は、フィルタリング部１２１によるフィルタリング後の語句である。その右側に、フィルタリングされた制約語句が、「追加しますか？」という形で表示される。 The display unit 500 displays a plurality of constraint phrases (constraint phrase list) for the input source language sentence. The words and phrases displayed here as constraint words are the words and phrases that have been filtered by the filtering unit 121. On the right side, the filtered constraint phrases are displayed in the form of "Do you want to add?"
　ユーザは、チェックボックスをマークすることで、修正（又は削除）あるいは追加したい制約語句を選択し、対応するボタンを押すことで、選択した制約語句の修正（又は削除）／追加を行なうことができる。また、ユーザ自身が作成した制約語句を追加することも可能である。 Users can select the constraint phrases they wish to modify (or delete) or add by marking checkboxes, and modify (or delete)/add the selected constraint phrases by pressing the corresponding button. . It is also possible to add constraint phrases created by the user himself.
　表示イメージにおける「更新」を押すことで、その時点での制約語句を使用した目的言語文を表示させることができる。 By pressing "Update" on the display image, you can display the target language sentence using the current constraint phrase.
　＜装置構成、動作＞
　上記のような表示を実現するための生成装置１００の構成例を図１１に示す。図１１に示すように、本実施例の生成装置１００は、抽出部１２０、表示情報生成部１７０、修正部１８０、生成部１９０、対訳辞書ＤＢ２００、制約語句リストＤＢ４００を有する。なお、修正部１８０は、表示情報生成部１７０の中に含まれていてもよい。 <Device configuration and operation>
FIG. 11 shows a configuration example of a generation device 100 for realizing the above display. As shown in FIG. 11, the generation device 100 of this embodiment includes an extraction section 120, a display information generation section 170, a modification section 180, a generation section 190, a bilingual dictionary DB 200, and a constraint phrase list DB 400. Note that the modification unit 180 may be included in the display information generation unit 170.
　なお、対訳辞書ＤＢ２００、制約語句リストＤＢ４００は、生成装置１００の外部に備えられてもよい。また、生成部１９０も、生成装置１００の外部（例えば別サーバ）に備えられてもよい。また、生成装置１００は、制約語句のリストを表示部５００に表示する目的で使用されてもよい。その場合、生成装置１００には、図１１に示す機能部のうち、抽出部１２０と表示情報生成部１７０のみが備えられることとしてもよい。生成装置１００を抽出装置と呼んでもよい。各部の機能は下記のとおりである。 Note that the bilingual dictionary DB 200 and the constraint phrase list DB 400 may be provided outside the generation device 100. Furthermore, the generation unit 190 may also be provided outside the generation device 100 (eg, another server). Further, the generation device 100 may be used for the purpose of displaying a list of constraint words on the display unit 500. In that case, the generation device 100 may include only the extraction unit 120 and the display information generation unit 170 among the functional units shown in FIG. The generation device 100 may also be called an extraction device. The functions of each part are as follows.
　抽出部１２０は、図４又は図５に示した抽出部１２０である。原言語文を入力として、制約語句リストを出力する。出力された制約語句リストは、制約語句リストＤＢ４００に格納されるとともに、表示情報生成部１７０に入力される。また、抽出部１２０は、フィルタリングした制約語句を、フィルタ語句リストとして出力してもよい。出力されたフィルタ語句リストは、表示情報生成部１７０に入力される。 The extraction unit 120 is the extraction unit 120 shown in FIG. 4 or 5. It takes the source language sentence as input and outputs a list of constraint words. The output constraint phrase list is stored in the constraint phrase list DB 400 and is input to the display information generation section 170. Further, the extraction unit 120 may output the filtered constraint words as a filter word list. The output filter word list is input to the display information generation section 170.
　表示情報生成部１７０は、制約語句リストを表示部５００に表示するための情報（制約語句リスト提示用情報と呼ぶ）を生成する。制約語句リスト提示用情報には、制約語句リストが含まれる。また、制約語句リスト提示用情報に、フィルタ語句リストの情報を、削除してしまった情報又はフィルタ候補語句又は追加候補として含めてもよい。制約語句リスト提示用情報は、表示情報生成部１７０から表示部５００に送信され、表示部５００に入力される。また、表示情報生成部１７０は、制約語句を修正可能な形式で、当該制約語句を用いて生成された目的言語文（翻訳文）とともに表示するための表示情報を生成してもよい。また、生成装置１００が、表示部５００から追加又は修正がなされた制約語句を受信した際に、表示情報生成部１７０は、受信した制約語句に基づいて生成された目的言語文（翻訳文）を取得し、当該目的言語文（翻訳文）を表示するための表示情報を生成してもよい。 The display information generation unit 170 generates information for displaying the constraint phrase list on the display unit 500 (referred to as constraint phrase list presentation information). The constraint word/phrase list presentation information includes a constraint word/phrase list. Further, the information for presenting the constraint word/phrase list may include information on the filter word/phrase list as deleted information, filter candidate words, or additional candidates. The constraint word list presentation information is transmitted from the display information generation section 170 to the display section 500 and input to the display section 500. Further, the display information generation unit 170 may generate display information to be displayed together with the target language sentence (translated sentence) generated using the constraint phrase in a format in which the constraint phrase can be modified. Further, when the generation device 100 receives the added or modified constraint phrase from the display unit 500, the display information generation unit 170 generates a target language sentence (translated sentence) based on the received constraint phrase. Display information for displaying the target language sentence (translated sentence) may be generated.
　また、表示情報生成部１７０は、ユーザが制約語句リストを確認する際の「修正支援情報」を生成し、それを表示部５００に送信してもよい。修正支援情報には、ユーザが入力した原言語文、抽出された制約語句リスト、抽出された制約語句リストに基づいて生成された目的言語文の少なくとも１つを含む。 Additionally, the display information generation unit 170 may generate “correction support information” for the user to check the constraint phrase list, and transmit it to the display unit 500. The modification support information includes at least one of a source language sentence input by the user, an extracted constraint phrase list, and a target language sentence generated based on the extracted constraint phrase list.
　修正部１８０は、ユーザが、提示された制約語句リストを修正した情報として、追加制約語句又は修正制約語句の少なくとも一方を、表示部５００から受信する。 The modification unit 180 receives from the display unit 500 at least one of the additional constraint phrases and the modified constraint phrases as information that the user has modified the presented constraint phrase list.
　修正部１８０は、受信した情報を元に、制約語句リストＤＢ４００に保存された情報を修正する。また、制約語句リストが修正された際に、修正された制約語句リストを元に、語彙制約付き機械翻訳によって再度目的言語文の生成を行い、表示情報生成部１７０で、当該目的言語文を有する修正支援情報を生成し、これを表示部５００に送信することで、表示部５００で表示してもよい。 The modification unit 180 modifies the information stored in the constraint phrase list DB 400 based on the received information. In addition, when the constraint phrase list is modified, a target language sentence is generated again by machine translation with vocabulary constraints based on the modified constraint phrase list, and the display information generation unit 170 generates a target language sentence. The modification support information may be generated and displayed on the display section 500 by transmitting it to the display section 500.
　生成部１９０は、入力生成部１３０、系列生成部１４０、及びリランキング部１５０を有する。これまでに説明したとおり、生成部１９０は、これらの機能部により、制約語句リストＤＢ４００から読み出した制約語句リストと、表示部５００から受信した原言語文とに基づいて、語彙制約を考慮した目的言語文（翻訳文）を生成し、生成した目的言語文を表示情報生成部１７０に入力する。 The generation unit 190 includes an input generation unit 130, a sequence generation unit 140, and a reranking unit 150. As explained above, the generation unit 190 uses these functional units to generate an objective that takes into account vocabulary constraints based on the constraint phrase list read from the constraint phrase list DB 400 and the source language sentence received from the display unit 500. A language sentence (translated sentence) is generated, and the generated target language sentence is input to the display information generation section 170.
　表示部５００は、例えば、ディスプレイを有するコンピュータ（端末）である。表示部５００は、生成装置１００とネットワークを介して接続される。図１０を参照して説明したように、表示部５００は、ユーザから原言語文を受け付けるとともに、制約語句リスト等を表示する。また、表示部５００は、制約語句や原言語文について、追加・修正指示を受け付ける。また、表示部５００は、原言語文、最終的な目的言語文、及び、最終的な制約語句リストをセットで出力することも可能である。 The display unit 500 is, for example, a computer (terminal) having a display. The display unit 500 is connected to the generation device 100 via a network. As described with reference to FIG. 10, the display unit 500 receives a source language sentence from the user and displays a list of constraint words and the like. The display unit 500 also accepts instructions for adding and modifying constraint words and sentences in the source language. The display unit 500 can also output a source language sentence, a final target language sentence, and a final constraint phrase list as a set.
　上記実施例における生成装置１００により、語彙制約付き機械翻訳の結果を確認しながら制約語句リストの修正を行うことをインタラクティブに繰り返す事で、よりユーザのイメージに近い目的言語文（翻訳文）を生成することができる。 The generation device 100 in the above embodiment generates a target language sentence (translated sentence) that is closer to the user's image by interactively repeating the process of modifying the restricted word list while checking the results of machine translation with vocabulary constraints. can do.
　（実験結果）
　以下の実験結果の説明では、「本実施の形態における生成装置１００」を提案手法あるいは提案システムと呼ぶ。 (Experimental result)
In the following explanation of the experimental results, "the generation device 100 in this embodiment" will be referred to as the proposed method or the proposed system.
　提案手法により自動抽出した語彙制約に対する翻訳候補のリランキングによる語彙制約付き機械翻訳手法の有効性を確認するため、日英翻訳を対象として対訳辞書から自動抽出した語彙制約による語彙制約付き機械翻訳の精度評価を行った。 In order to confirm the effectiveness of the machine translation method with lexical constraints based on re-ranking of translation candidates based on the lexical constraints automatically extracted using the proposed method, we conducted a machine translation method with lexical constraints using lexical constraints automatically extracted from a bilingual dictionary for Japanese-English translation. Accuracy evaluation was performed.
　＜対訳辞書について＞
　語彙制約を抽出するために用いる対訳辞書として、汎用的な辞書であるEDR日英対訳辞書（EDR-JE）および日英翻訳システムALT-J/E の対訳辞書を用いた。 <About the bilingual dictionary>
As bilingual dictionaries used to extract vocabulary constraints, we used the EDR Japanese-English bilingual dictionary (EDR-JE), which is a general-purpose dictionary, and the bilingual dictionary from the Japanese-English translation system ALT-J/E.
　＜モデル＞
　評価に用いる翻訳モデルとしては以下のものを使用した。 <Model>
The following translation models were used for evaluation.
　・Transformer
　・LeCA + {EDR-JE, ALT-J/E}
　・LeCA+LCD + {EDR-JE, ALT-J/E}
　翻訳モデルの学習および評価に使用する対訳コーパスにはASPECを用いた。各モデルの詳細な設定およびハイパーパラメータについては、図１２に示すものを用いた。 ・Transformer
・LeCA + {EDR-JE, ALT-J/E}
・LeCA+LCD + {EDR-JE, ALT-J/E}
ASPEC was used as the bilingual corpus used for training and evaluating the translation model. For detailed settings and hyperparameters of each model, those shown in FIG. 12 were used.
　辞書から抽出した制約に対しては２｜Ｃ｜個の語彙制約からそれぞれ生成文の上位３０文を集めたものを用いた。翻訳候補のリランキングの際に用いるスコアにはRerankerにより、原言語文と翻訳候補からリランキングモデルの計算するスコアを用いた。 For the constraints extracted from the dictionary, a collection of the top 30 sentences generated from each of the 2 |C| vocabulary constraints was used. The score used for reranking translation candidates was the score calculated by the reranking model from the source language sentence and translation candidates using Reranker.
　リランキングモデルには文末から文頭へと翻訳文を生成するRight-to-Left 翻訳タスクをTransformer（big）で学習したモデルを使用した。その際のリランキングのスコアとしては入力された翻訳候補をforced decodingした際の尤度を用いた。各手法の評価には翻訳精度の自動評価尺度であるBLEUを用いた。 For the reranking model, we used a model trained with Transformer (big) for the Right-to-Left translation task, which generates translated sentences from the end of the sentence to the beginning of the sentence. The re-ranking score used was the likelihood when forced decoding was performed on the input translation candidates. BLEU, an automatic evaluation scale for translation accuracy, was used to evaluate each method.
　＜実験結果について＞
　対訳辞書により自動抽出した語彙制約を用いた際の各手法の翻訳精度を図１３に示す。リランキングモデルによるスコアを用いるRerankerにおいて、ベースライン（Transformer）に対してLeCAやLeCA+LCDが翻訳精度を向上させることができていることがわかる。また、図１３から、辞書の種類に依らずに、翻訳精度が高いことがわかる。 <About experimental results>
FIG. 13 shows the translation accuracy of each method when using vocabulary constraints automatically extracted by a bilingual dictionary. It can be seen that LeCA and LeCA+LCD are able to improve translation accuracy compared to the baseline (Transformer) in Reranker, which uses scores based on the reranking model. Moreover, from FIG. 13, it can be seen that the translation accuracy is high regardless of the type of dictionary.
　（ハードウェア構成例）
　本実施の形態で説明したいずれの装置（生成装置１００、抽出装置）も、例えば、コンピュータにプログラムを実行させることにより実現できる。このコンピュータは、物理的なコンピュータであってもよいし、クラウド上の仮想マシンであってもよい。 (Hardware configuration example)
Any of the devices (generating device 100, extracting device) described in this embodiment can be realized, for example, by causing a computer to execute a program. This computer may be a physical computer or a virtual machine on the cloud.
　すなわち、当該装置は、コンピュータに内蔵されるＣＰＵやメモリ等のハードウェア資源を用いて、当該装置で実施される処理に対応するプログラムを実行することによって実現することが可能である。上記プログラムは、コンピュータが読み取り可能な記録媒体（可搬メモリ等）に記録して、保存したり、配布したりすることが可能である。また、上記プログラムをインターネットや電子メール等、ネットワークを通して提供することも可能である。 That is, the device can be realized by using hardware resources such as a CPU and memory built into a computer to execute a program corresponding to the processing performed by the device. The above program can be recorded on a computer-readable recording medium (such as a portable memory) and can be stored or distributed. It is also possible to provide the above program through a network such as the Internet or e-mail.
　図１４は、上記コンピュータのハードウェア構成例を示す図である。図１４のコンピュータは、それぞれバスＢＳで相互に接続されているドライブ装置１０００、補助記憶装置１００２、メモリ装置１００３、ＣＰＵ１００４、インタフェース装置１００５、表示装置１００６、入力装置１００７、出力装置１００８等を有する。なお、当該コンピュータは、更にＧＰＵを備えてもよい。 FIG. 14 is a diagram showing an example of the hardware configuration of the computer. The computer in FIG. 14 includes a drive device 1000, an auxiliary storage device 1002, a memory device 1003, a CPU 1004, an interface device 1005, a display device 1006, an input device 1007, an output device 1008, etc., which are interconnected by a bus BS. Note that the computer may further include a GPU.
　当該コンピュータでの処理を実現するプログラムは、例えば、ＣＤ－ＲＯＭ又はメモリカード等の記録媒体１００１によって提供される。プログラムを記憶した記録媒体１００１がドライブ装置１０００にセットされると、プログラムが記録媒体１００１からドライブ装置１０００を介して補助記憶装置１００２にインストールされる。但し、プログラムのインストールは必ずしも記録媒体１００１より行う必要はなく、ネットワークを介して他のコンピュータよりダウンロードするようにしてもよい。補助記憶装置１００２は、インストールされたプログラムを格納すると共に、必要なファイルやデータ等を格納する。 A program that realizes processing on the computer is provided, for example, on a recording medium 1001 such as a CD-ROM or a memory card. When the recording medium 1001 storing the program is set in the drive device 1000, the program is installed from the recording medium 1001 to the auxiliary storage device 1002 via the drive device 1000. However, the program does not necessarily need to be installed from the recording medium 1001, and may be downloaded from another computer via a network. The auxiliary storage device 1002 stores installed programs as well as necessary files, data, and the like.
　メモリ装置１００３は、プログラムの起動指示があった場合に、補助記憶装置１００２からプログラムを読み出して格納する。ＣＰＵ１００４は、メモリ装置１００３に格納されたプログラムに従って、ライトタッチ維持装置１００に係る機能を実現する。インタフェース装置１００５は、ネットワーク等に接続するためのインタフェースとして用いられる。表示装置１００６はプログラムによるＧＵＩ（Ｇｒａｐｈｉｃａｌ　Ｕｓｅｒ　Ｉｎｔｅｒｆａｃｅ）等を表示する。入力装置１００７はキーボード及びマウス、ボタン、又はタッチパネル等で構成され、様々な操作指示を入力させるために用いられる。出力装置１００８は演算結果を出力する。 The memory device 1003 reads and stores the program from the auxiliary storage device 1002 when there is an instruction to start the program. CPU 1004 implements functions related to light touch maintenance device 100 according to programs stored in memory device 1003. The interface device 1005 is used as an interface for connecting to a network or the like. A display device 1006 displays a GUI (Graphical User Interface) and the like based on a program. The input device 1007 is composed of a keyboard, a mouse, buttons, a touch panel, or the like, and is used to input various operation instructions. An output device 1008 outputs the calculation result.
　（実施の形態のまとめ、効果等）
　以上説明したとおり、本実施の形態で説明した技術により、語彙制約付き機械翻訳に用いる制約語句を、低いノイズで適切に自動抽出することが可能となる。また、本実施の形態で説明した技術により、語彙制約付き機械翻訳において、精度良く翻訳を行うことができる。 (Summary of embodiments, effects, etc.)
As described above, the technology described in this embodiment makes it possible to appropriately automatically extract constraint phrases used in machine translation with vocabulary constraints with low noise. Furthermore, with the technology described in this embodiment, it is possible to perform translation with high accuracy in machine translation with vocabulary constraints.
　以上の実施形態に関し、更に以下の付記１と付記２を開示する。 Regarding the above embodiments, the following additional notes 1 and 2 are further disclosed.
　＜付記１＞
（付記項１）
　メモリと、
　前記メモリに接続された少なくとも１つのプロセッサと、
　を含み、
　前記プロセッサは、
　第１情報と第２情報とのペアの集合である辞書における前記第１情報と、第１系列のそれぞれを単位情報に分割し、
　前記第１系列の単位情報にマッチする前記第１情報に対応する前記第２情報を、前記辞書から、前記第１系列に基づいて第２系列を生成するために使用される制約情報として抽出する
　抽出装置。
（付記項２）
　前記プロセッサは、前記辞書から、予め定めたルールに合致するペアを削除し、当該削除処理が施された辞書を使用する
　付記項１に記載の抽出装置。
（付記項３）
　前記予め定めたルールに合致するペアは、名詞以外の語句又は名詞句以外の語句を含むペア、長さが１の語句からなるペア、及び、第１情報と第２情報との対応関係に一意性が無いペアのうちの少なくともいずれか１つである
　付記項２に記載の抽出装置。
（付記項４）
　前記プロセッサは、曖昧性を解消するように、前記第１系列の単位情報と前記第１情報とのマッチングを実行する
　付記項１ないし３のうちいずれか１項に記載の抽出装置。
（付記項５）
　前記プロセッサは、前記制約情報を表示部に送信するための表示情報を生成し、前記表示部に表示された制約情報に対して追加又は修正がなされた制約情報を受信する
　付記項１ないし４のうちいずれか１項に記載の抽出装置。
（付記項６）
　メモリと、
　前記メモリに接続された少なくとも１つのプロセッサと、
　を含み、
　前記プロセッサは、
　第１系列を入力として、当該第１系列と、第１情報と第２情報とのペアの集合である辞書とに基づいて、制約情報を抽出し、
　前記制約情報と前記第１系列とに基づいて第２系列を生成し、
　前記制約情報を修正可能な形式で、前記第２系列とともに表示するための表示情報を生成する
　生成装置。
（付記項７）
　追加又は修正がなされた制約情報を受信した際に、前記プロセッサは、受信した制約情報に基づいて生成された系列を取得し、当該系列を表示するための表示情報を生成する
　付記項６に記載の生成装置。
（付記項８）
前記プロセッサは、予め定めたルールに基づきフィルタリングされた制約情報を追加候補として表示するための表示情報を生成する
　付記項６又は７に記載の生成装置。
（付記項９）
　コンピュータが実行する抽出方法であって、
　第１情報と第２情報とのペアの集合である辞書における前記第１情報と、第１系列のそれぞれを単位情報に分割する分割ステップと、
　前記第１系列の単位情報にマッチする前記第１情報に対応する前記第２情報を、前記辞書から、前記第１系列に基づいて第２系列を生成するために使用される制約情報として抽出する制約情報抽出ステップと
　を備える抽出方法。
（付記項１０）
　コンピュータが実行する生成方法であって、
　第１系列を入力として、当該第１系列と、第１情報と第２情報とのペアの集合である辞書とに基づいて、制約情報を抽出する抽出ステップと、
　前記制約情報と前記第１系列とに基づいて第２系列を生成する生成ステップと、
　前記制約情報を修正可能な形式で、前記第２系列とともに表示するための表示情報を生成する表示情報生成ステップと
　を備える生成方法。
（付記項１１）
　コンピュータを、付記項１ないし５のうちいずれか１項に記載の抽出装置として機能させるためのプログラムを記憶した非一時的記憶媒体。 <Additional note 1>
(Additional note 1)
memory and
at least one processor connected to the memory;
including;
The processor includes:
dividing each of the first information and the first series in a dictionary which is a set of pairs of first information and second information into unit information;
The second information corresponding to the first information that matches the unit information of the first series is extracted from the dictionary as constraint information used to generate a second series based on the first series. Extraction device.
(Additional note 2)
The extraction device according to supplementary note 1, wherein the processor deletes pairs that match a predetermined rule from the dictionary, and uses the dictionary that has been subjected to the deletion process.
(Additional note 3)
Pairs that match the predetermined rules include words other than nouns, pairs containing words other than noun phrases, pairs consisting of words with a length of 1, and pairs that are unique in the correspondence between the first information and the second information. The extraction device according to Supplementary Note 2, wherein at least one of the pairs has no gender.
(Additional note 4)
The extraction device according to any one of Supplementary Notes 1 to 3, wherein the processor performs matching between the unit information of the first series and the first information so as to resolve ambiguity.
(Additional note 5)
The processor generates display information for transmitting the constraint information to the display unit, and receives constraint information added or modified to the constraint information displayed on the display unit. Additional notes 1 to 4. The extraction device according to any one of these.
(Additional note 6)
memory and
at least one processor connected to the memory;
including;
The processor includes:
extracting constraint information based on the first series and a dictionary that is a set of pairs of first information and second information with the first series as input;
generating a second sequence based on the constraint information and the first sequence;
A generation device that generates display information for displaying the constraint information together with the second series in a modifiable format.
(Supplementary Note 7)
When receiving added or modified constraint information, the processor obtains a series generated based on the received constraint information, and generates display information for displaying the series. According to supplementary note 6. generator.
(Supplementary Note 8)
The generating device according to supplementary note 6 or 7, wherein the processor generates display information for displaying filtered constraint information as additional candidates based on a predetermined rule.
(Supplementary Note 9)
A computer-implemented extraction method, comprising:
a dividing step of dividing each of the first information and the first series into unit information in a dictionary that is a set of pairs of first information and second information;
The second information corresponding to the first information that matches the unit information of the first series is extracted from the dictionary as constraint information used to generate a second series based on the first series. An extraction method comprising a constraint information extraction step.
(Supplementary Note 10)
A generation method executed by a computer,
an extraction step of receiving the first series as input and extracting constraint information based on the first series and a dictionary that is a set of pairs of first information and second information;
a generation step of generating a second sequence based on the constraint information and the first sequence;
a display information generation step of generating display information for displaying the constraint information together with the second series in a modifiable format.
(Supplementary Note 11)
A non-temporary storage medium storing a program for causing a computer to function as the extraction device according to any one of Additional Items 1 to 5.
　＜付記２＞
（付記項１）
　制約情報と、情報の系列である第１系列とから別の情報の系列である第２系列を生成するための生成装置であって、
　メモリと、
　前記メモリに接続された少なくとも１つのプロセッサと、
　を含み、
　前記プロセッサは、
　制約情報リストを入力し、前記制約情報リストに含まれる１又は複数の制約情報の部分集合の各要素を語彙制約として出力し、
　前記第１系列と前記語彙制約とを用いて、前記第２系列についての１又は複数の候補を生成し、
　前記第２系列としての適切さを示すスコアを、前記１又は複数の候補のそれぞれについて算出する
　生成装置。
（付記項２）
　前記プロセッサは、前記系列生成部において前記候補の生成に用いたモデルにより出力される尤度と、リランキングモデルにより前記候補から得られる尤度のうちの少なくとも一方に基づいて、前記スコアを算出する
　付記項１に記載の生成装置。
（付記項３）
　前記制約情報リストに曖昧性を有する制約情報が含まれる場合において、前記プロセッサは、曖昧性を考慮した語彙制約付きのビームサーチを行うことにより、前記１又は複数の候補を生成する
　付記項１又は２に記載の生成装置。
（付記項４）
　少なくとも１つの制約情報が、２以上の曖昧性を許容する形式で前記プロセッサに入力され、前記プロセッサは、当該曖昧性を保持したまま語彙制約を生成する
　付記項１ないし３のうちいずれか１項に記載の生成装置。
（付記項５）
　制約情報と、情報の系列である第１系列とから別の情報の系列である第２系列を生成するためのコンピュータが実行する生成方法であって、
　制約情報リストを入力し、前記制約情報リストに含まれる１又は複数の制約情報の部分集合の各要素を語彙制約として出力する入力生成ステップと、
　前記第１系列と前記語彙制約とを用いて、前記第２系列についての１又は複数の候補を生成する系列生成ステップと、
　前記第２系列としての適切さを示すスコアを、前記１又は複数の候補のそれぞれについて算出するリランキングステップと
　を備える生成方法。
（付記項６）
　コンピュータを、付記項１ないし４のうちいずれか１項に記載の生成装置における各部として機能させるためのプログラムを記憶した非一時的記憶媒体。 <Additional note 2>
(Additional note 1)
A generation device for generating constraint information and a second series that is another information series from a first series that is an information series,
memory and
at least one processor connected to the memory;
including;
The processor includes:
inputting a constraint information list, outputting each element of a subset of one or more constraint information included in the constraint information list as a vocabulary constraint;
generating one or more candidates for the second sequence using the first sequence and the vocabulary constraint;
A generation device that calculates a score indicating suitability as the second series for each of the one or more candidates.
(Additional note 2)
The processor calculates the score based on at least one of a likelihood output by a model used to generate the candidate in the sequence generation unit and a likelihood obtained from the candidate by a reranking model. The generating device according to supplementary note 1.
(Additional note 3)
In the case where the constraint information list includes constraint information having ambiguity, the processor generates the one or more candidates by performing a beam search with a vocabulary constraint that takes the ambiguity into account. 2. The generating device according to 2.
(Additional note 4)
At least one piece of constraint information is input to the processor in a format that allows two or more ambiguities, and the processor generates a vocabulary constraint while maintaining the ambiguity. Any one of Additional Notes 1 to 3. The generator described in .
(Additional note 5)
A generation method performed by a computer for generating constraint information and a second series that is another information series from a first series that is an information series, the method comprising:
an input generation step of inputting a constraint information list and outputting each element of a subset of one or more constraint information included in the constraint information list as a vocabulary constraint;
a sequence generation step of generating one or more candidates for the second sequence using the first sequence and the vocabulary constraint;
A generation method comprising: a reranking step of calculating a score indicating suitability as the second series for each of the one or more candidates.
(Additional note 6)
A non-temporary storage medium storing a program for causing a computer to function as each part of the generation device according to any one of Supplementary Notes 1 to 4.
　以上、本実施の形態について説明したが、本発明はかかる特定の実施形態に限定されるものではなく、特許請求の範囲に記載された本発明の要旨の範囲内において、種々の変形・変更が可能である。 Although the present embodiment has been described above, the present invention is not limited to such specific embodiment, and various modifications and changes can be made within the scope of the gist of the present invention as described in the claims. It is possible.
１００　生成装置
１１０　入力部
１２０　抽出部
１２１　フィルタリング部
１２２　分割部
１２３　制約語句抽出部
１３０　入力生成部
１４０　系列生成部
１４１　系列変換部
１４２　探索部
１５０　リランキング部
１６０　出力部
１７０　表示情報生成部
１８０　修正部
１９０　生成部
２００　対訳辞書ＤＢ
３００　モデルＤＢ
４００　制約語句リストＤＢ
５００　表示部
１０００　ドライブ装置
１００１　記録媒体
１００２　補助記憶装置
１００３　メモリ装置
１００４　ＣＰＵ
１００５　インタフェース装置
１００６　表示装置
１００７　入力装置
１００８　出力装置 100 Generation device 110 Input section 120 Extraction section 121 Filtering section 122 Division section 123 Constraint phrase extraction section 130 Input generation section 140 Sequence generation section 141 Sequence conversion section 142 Search section 150 Reranking section 160 Output section 170 Display information generation section 180 Modification section 190 Generation unit 200 Bilingual dictionary DB
300 model DB
400 Constraint phrase list DB
500 Display unit 1000 Drive device 1001 Recording medium 1002 Auxiliary storage device 1003 Memory device 1004 CPU
1005 Interface device 1006 Display device 1007 Input device 1008 Output device

Claims (6)

  1. 　制約情報と、情報の系列である第１系列とから別の情報の系列である第２系列を生成するための生成装置であって、
    　制約情報リストを入力し、前記制約情報リストに含まれる１又は複数の制約情報の部分集合の各要素を語彙制約として出力する入力生成部と、
    　前記第１系列と前記語彙制約とを用いて、前記第２系列についての１又は複数の候補を生成する系列生成部と、
    　前記第２系列としての適切さを示すスコアを、前記１又は複数の候補のそれぞれについて算出するリランキング部と
    　を備える生成装置。     A generation device for generating constraint information and a second series that is another information series from a first series that is an information series,
    an input generation unit that inputs a constraint information list and outputs each element of a subset of one or more constraint information included in the constraint information list as a vocabulary constraint;
    a sequence generation unit that generates one or more candidates for the second sequence using the first sequence and the vocabulary constraint;
    and a reranking unit that calculates a score indicating suitability as the second series for each of the one or more candidates.
  2. 　前記リランキング部は、前記系列生成部において前記候補の生成に用いたモデルにより出力される尤度と、リランキングモデルにより前記候補から得られる尤度のうちの少なくとも一方に基づいて、前記スコアを算出する
    　請求項１に記載の生成装置。     The reranking unit calculates the score based on at least one of the likelihood output by the model used to generate the candidates in the sequence generation unit and the likelihood obtained from the candidates by the reranking model. The generating device according to claim 1.
  3. 　前記制約情報リストに曖昧性を有する制約情報が含まれる場合において、前記系列生成部は、曖昧性を考慮した語彙制約付きのビームサーチを行うことにより、前記１又は複数の候補を生成する
    　請求項１に記載の生成装置。     When the constraint information list includes constraint information having ambiguity, the sequence generation unit generates the one or more candidates by performing a beam search with vocabulary constraints that take ambiguity into consideration. 1. The generating device according to 1.
  4. 　少なくとも１つの制約情報が、２以上の曖昧性を許容する形式で前記入力生成部に入力され、前記入力生成部は、当該曖昧性を保持したまま語彙制約を生成する
    　請求項１に記載の生成装置。     Generation according to claim 1, wherein at least one piece of constraint information is input to the input generation unit in a format that allows two or more ambiguities, and the input generation unit generates the lexical constraint while maintaining the ambiguity. Device.
  5. 　制約情報と、情報の系列である第１系列とから別の情報の系列である第２系列を生成するための生成装置が実行する生成方法であって、
    　制約情報リストを入力し、前記制約情報リストに含まれる１又は複数の制約情報の部分集合の各要素を語彙制約として出力する入力生成ステップと、
    　前記第１系列と前記語彙制約とを用いて、前記第２系列についての１又は複数の候補を生成する系列生成ステップと、
    　前記第２系列としての適切さを示すスコアを、前記１又は複数の候補のそれぞれについて算出するリランキングステップと
    　を備える生成方法。     A generation method executed by a generation device for generating constraint information and a second series that is another information series from a first series that is an information series, the method comprising:
    an input generation step of inputting a constraint information list and outputting each element of a subset of one or more constraint information included in the constraint information list as a vocabulary constraint;
    a sequence generation step of generating one or more candidates for the second sequence using the first sequence and the vocabulary constraint;
    A generation method comprising: a reranking step of calculating a score indicating suitability as the second series for each of the one or more candidates.
  6. 　コンピュータを、請求項１ないし４のうちいずれか１項に記載の生成装置における各部として機能させるためのプログラム。 A program for causing a computer to function as each part of the generation device according to any one of claims 1 to 4.
PCT/JP2022/026407 2022-06-30 2022-06-30 Generation device, generation method, and program WO2024004184A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2022/026407 WO2024004184A1 (en) 2022-06-30 2022-06-30 Generation device, generation method, and program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2022/026407 WO2024004184A1 (en) 2022-06-30 2022-06-30 Generation device, generation method, and program

Publications (1)

Publication Number Publication Date
WO2024004184A1 true WO2024004184A1 (en) 2024-01-04

Family

ID=89382574

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2022/026407 WO2024004184A1 (en) 2022-06-30 2022-06-30 Generation device, generation method, and program

Country Status (1)

Country Link
WO (1) WO2024004184A1 (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0696114A (en) * 1992-09-11 1994-04-08 Toshiba Corp Machine translation system and document editor
JP2016189154A (en) * 2015-03-30 2016-11-04 日本電信電話株式会社 Translation method, device, and program

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0696114A (en) * 1992-09-11 1994-04-08 Toshiba Corp Machine translation system and document editor
JP2016189154A (en) * 2015-03-30 2016-11-04 日本電信電話株式会社 Translation method, device, and program

Similar Documents

Publication Publication Date Title
US5895446A (en) Pattern-based translation method and system
JP3189186B2 (en) Translation device based on patterns
US20080040095A1 (en) System for Multiligual Machine Translation from English to Hindi and Other Indian Languages Using Pseudo-Interlingua and Hybridized Approach
JP2005216126A (en) Text generation method and text generation device of other language
JPS62163173A (en) Mechanical translating device
JP2000353161A (en) Method and device for controlling style in generation of natural language
CN110678868B (en) Translation support system, translation support apparatus, translation support method, and computer-readable medium
US20030139920A1 (en) Multilingual database creation system and method
US20030083860A1 (en) Content conversion method and apparatus
US20050273316A1 (en) Apparatus and method for translating Japanese into Chinese and computer program product
Dhanani et al. FAST-MT Participation for the JOKER CLEF-2022 Automatic Pun and Humour Translation Tasks
Scannell Statistical models for text normalization and machine translation
Al-Mannai et al. Unsupervised word segmentation improves dialectal Arabic to English machine translation
Mara English-Wolaytta Machine Translation using Statistical Approach
JP2018072979A (en) Parallel translation sentence extraction device, parallel translation sentence extraction method and program
Yeong et al. Using dictionary and lemmatizer to improve low resource English-Malay statistical machine translation system
WO2024004184A1 (en) Generation device, generation method, and program
WO2024004183A1 (en) Extraction device, generation device, extraction method, generation method, and program
JP2009157888A (en) Transliteration model generation device, transliteration apparatus, and computer program therefor
Núñez et al. Phonetic normalization for machine translation of user generated content
Anto et al. Text to speech synthesis system for English to Malayalam translation
JP4829685B2 (en) Translation phrase pair generation apparatus, statistical machine translation apparatus, translation phrase pair generation method, statistical machine translation method, translation phrase pair generation program, statistical machine translation program, and storage medium
Langlais et al. General-purpose statistical translation engine and domain specific texts: Would it work?
Ouvrard et al. Collatinus & Eulexis: Latin & Greek Dictionaries in the Digital Ages.
Sankaravelayuthan et al. A Comprehensive Study of Shallow Parsing and Machine Translation in Malaylam

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22949457

Country of ref document: EP

Kind code of ref document: A1