A machine translation assisted optimization method, device, equipment and medium

Abstract

The present invention discloses a machine translation auxiliary optimization method, device, equipment and medium. The method obtains a text to be translated, and matches and queries the text to be translated with an original text in a translation memory, marks the text to be translated with inconsistent matches as the marked text to be translated, and also marks the original text with inconsistent matches as the marked original text; based on the marked original text, obtains the corresponding translation from the translation memory as the marked translation, and the marked translation carries a word position identifier; performs machine translation on the marked text to be translated to obtain a machine-translated text, and the machine-translated text carries a word position identifier; based on the word position identifier, replaces the marked translation in the translation memory with the machine-translated text through a word alignment tool, obtains the translation of the text to be translated, combines the translation memory with the machine translation, and improves the translation accuracy and translation efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of machine translation, and in particular to a machine translation auxiliary optimization method, device, equipment and medium. Background Art

[0002] With the continuous advancement of globalization and informatization, the translation industry has also ushered in unprecedented challenges. Human translation alone can no longer meet people's needs. In order to meet the massive translation needs, machine translation has come into being. Although machine translation can improve translation speed, its translation accuracy needs to be improved, especially in fields such as medicine and communications where there are a large number of professional terms. Strong professional background support is required for translation. At present, when translating technical fields with relatively high professional requirements, most translation memories (TM) are used. The translation content matching the translation content is searched in the translation memory to obtain the translation, but the resources in the translation memory are limited, and the translation efficiency of translation using the translation memory is not high. Summary of the invention

[0003] The technical problem to be solved by the present invention is that the efficiency of translating in technical fields with relatively high professional requirements by simply using translation memory or machine translation in the prior art is not high. Therefore, the present invention provides a machine translation auxiliary optimization method, device, equipment and medium, which translates the text to be translated based on the translation memory combined with machine translation, and improves the translation efficiency on the basis of ensuring the translation accuracy.

[0004] The present invention is achieved through the following technical solutions:

[0005] A machine translation assisted optimization method, comprising:

[0006] Obtain the text to be translated, and match the text to be translated with the original text in the translation memory, mark the text to be translated that is inconsistent with the match as the marked text to be translated S1, and also mark the original text that is inconsistent with the match as the marked original text S2;

[0007] Based on the marked original text S2, a corresponding translation is obtained from the translation memory as the marked translation T2, wherein the marked translation T2 carries a word position identifier;

[0008] Determine mismatched words based on the marked text to be translated S1 and the marked original text S2, and perform machine translation on the mismatched words in the marked text to be translated S1 to obtain a machine translated text T1, wherein the machine translated text T1 carries a word position identifier;

[0009] Based on the word position identifier, the marked translation T2 in the translation memory is replaced with the machine translated text T1 through a word alignment tool to obtain the translation of the text to be translated.

[0010] Furthermore, the matching query between the text to be translated and the original text in the translation memory and marking the text to be translated that does not match the original text as marked text to be translated S1 includes:

[0011] Divide the text to be translated into sentences by using sentence segmentation markers, and count the number of sentences in the text to be translated;

[0012] When the number of sentences in the text to be translated is less than a preset number, the translation memory is polled based on each sentence in the text to be translated, and sentences with inconsistent matches are marked to form a marked text to be translated S1;

[0013] When the number of sentences in the text to be translated is greater than a preset number, the text to be translated and the original text in the translation memory are compared by a text comparison tool, and inconsistent texts are marked as marked text to be translated S1.

[0014] Furthermore, the step of obtaining a corresponding translation from a translation memory as the marked translation T2 based on the marked original text S2 includes:

[0015] Determine the mismatched words based on the to-be-translated marked text S1 and the marked original text S2, and determine the corresponding translations as the mismatched translations from the translation memory through a word alignment tool based on the marked original text S2;

[0016] The marked translation T2 is determined from the unmatched translations based on the unmatched words in the marked original text S2.

[0017] Furthermore, the non-matching words in the to-be-translated marked text S1 are machine translated to obtain the machine translated text T1, including:

[0018] The number of sentences in the to-be-translated marked text S1 is determined according to the sentence segmentation identifiers. When the number of sentences in the to-be-translated marked text S1 is 1, unmatched words are selected from the to-be-translated marked text S1 for machine translation to obtain a machine-translated text T1.

[0019] Furthermore, when the number of sentences in the to-be-translated marked text S1 is not 1, after selecting unmatched words from the to-be-translated marked text S1 for machine translation to obtain the machine translated text T1, the machine translation auxiliary optimization method further includes:

[0020] According to the sentence order in the marked text to be translated, the words in the same sentence order are sorted to obtain the machine translated text T1.

[0021] Furthermore, the machine translation assisted optimization method also includes:

[0022] When the text to be translated matches the original text in the translation memory, the word alignment tool selects the corresponding translation from the translation memory based on the original text as the translation of the text to be translated.

[0023] A machine translation auxiliary optimization device, comprising:

[0024] The text matching module is used to obtain the text to be translated, and match the text to be translated with the original text in the translation memory, mark the text to be translated that is inconsistent with the match as the marked text to be translated S1, and also mark the original text that is inconsistent with the match as the marked original text S2;

[0025] A marked translation acquisition module, used to acquire a corresponding translation from a translation memory based on the marked original text S2 as a marked translation T2, wherein the marked translation T2 carries a word position identifier;

[0026] A machine translation module, for determining mismatched words based on the to-be-translated marked text S1 and the marked original text S2, and performing machine translation on the mismatched words in the to-be-translated marked text S1 to obtain a machine-translated text T1, wherein the machine-translated text T1 carries a word position identifier;

[0027] The translation acquisition module is used to replace the marked translation T2 in the translation memory with the machine translation text T1 based on the word position identifier through the word alignment tool to obtain the translation of the text to be translated.

[0028] Furthermore, the text matching module includes:

[0029] A sentence number counting unit is used to divide the text to be translated into sentences by using sentence segmentation marks, and to count the number of sentences in the text to be translated;

[0030] A first to-be-translated text marking unit, for, when the number of sentences in the to-be-translated text is less than a preset number, polling in a translation memory based on each sentence in the to-be-translated text, marking sentences with inconsistent matches, and forming a to-be-translated marked text S1;

[0031] The second text-to-be-translated marking unit is used for comparing the text-to-be-translated with the original text in the translation memory by using a text comparison tool when the number of sentences in the text-to-be-translated is greater than a preset number, and marking inconsistent texts as marked text-to-be-translated S1.

[0032] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned machine translation assisted optimization method when executing the computer program.

[0033] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the machine translation-assisted optimization method is implemented.

[0034] The present invention provides a machine translation auxiliary optimization method, device, equipment and medium. The method comprises the following steps: obtaining a text to be translated, matching and querying the text to be translated with an original text in a translation memory, marking the text to be translated with inconsistent matches as a marked text to be translated S1, and marking the original text with inconsistent matches as a marked original text S2; obtaining a corresponding translation from a translation memory as a marked translation T2 based on the marked original text S2, and the marked translation T2 carries a word position identifier; performing machine translation on the marked text to be translated S1 to obtain a machine translation text T1, and the machine translation text T1 carries a word position identifier; based on the word position identifier, replacing the marked translation T2 in the translation memory with the machine translation text T1 through a word alignment tool, obtaining a translation of the text to be translated, combining the translation memory with machine translation, and improving translation accuracy and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0036] Figure 1 The present invention is a flowchart of a machine translation assisted optimization method.

[0037] Figure 2 for Figure 1 A specific flow chart of step S10 in FIG.

[0038] Figure 3 The figure is a schematic diagram of the structure of a machine translation auxiliary optimization device of the present invention.

[0039] Figure 4 A schematic diagram of a computer device of the present invention. DETAILED DESCRIPTION

[0040] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0041] The present invention provides a machine translation assisted optimization method, which can be applied to different electronic devices, including but not limited to various personal computers, laptops, smart phones and tablet computers.

[0042] In one embodiment, if Figure 1 As shown, the present invention provides a machine translation assisted optimization method, comprising:

[0043] S10: Obtain the text to be translated, and match the text to be translated with the original text in the translation memory, mark the text to be translated that does not match as the marked text to be translated S1, and mark the original text that does not match as the marked original text S2;

[0044] S20: Based on the marked original text S2, a corresponding translation is obtained from the translation memory as the marked translation T2, wherein the marked translation T2 carries a word position identifier;

[0045] S30: determining mismatched words based on the to-be-translated marked text S1 and the marked original text S2, and performing machine translation on the mismatched words in the to-be-translated marked text S1 to obtain a machine-translated text T1, wherein the machine-translated text T1 carries a word position identifier;

[0046] S40: Based on the word position identifier, the marked translation T2 in the translation memory is replaced with the machine translated text T1 by a word alignment tool to obtain the translation of the text to be translated.

[0047] The text to be translated refers to the text that needs to be translated, and the text to be translated can be in any language.

[0048] A translation memory refers to a collection of highly professional original texts and translated texts. The translations in the translation memory can be those translated by the original authors or those translated by some professional translators based on publicly available original texts.

[0049] It should be noted that when the text to be translated is matched with the original text in the translation memory, the inconsistent matches may be words, phrases or sentences. However, in order to facilitate the determination of the positions of the unmatched words, the marked text to be translated S1 and the marked original text S2 in this embodiment are in units of sentences. In order to ensure that the sentence format remains unchanged, the machine translated text T1 and the marked translation T2 are in units of words.

[0050] The machine translated text T1 carries a word position identifier which indicates the position of the sentence where the unmatched word in the to-be-translated marked text S1 is located in the to-be-translated text.

[0051] The marked translation T2 carries a word position identifier indicating the position of the sentence where the marked text T2 is located in the translation memory.

[0052] As an example, in step S10, the text to be translated is obtained, and the text to be translated is matched with the original text in the translation memory. If the match is inconsistent, it means that there is no translation corresponding to the text to be translated in the translation memory. The text to be translated with inconsistent matching is marked as marked text to be translated S1, and the original text with inconsistent matching is also marked as marked original text S2.

[0053] If the text to be translated matches the original text in the translation memory, it means that there is a translation corresponding to the text to be translated in the translation memory. Then, based on the original text that matches the text to be translated, the corresponding translation is selected from the translation memory as the translation of the text to be translated through the word alignment tool.

[0054] As an example, in step S20, when the text to be translated does not match the original text in the translation memory, the unmatched words are determined based on the marked text to be translated S1 and the marked original text S2, and based on the marked original text S2, the corresponding translation is determined from the translation memory as the unmatched translation through the word alignment tool; and the marked translation T2 is determined from the unmatched translation based on the unmatched words in the marked original text S2.

[0055] As an example, in step S30, after obtaining the marked text S1 to be translated, the number of sentences in the marked text S1 to be translated is first determined according to the sentence segmentation identifier. When the number of sentences in the marked text S1 to be translated is 1, unmatched words are selected from the marked text S1 to be translated for machine translation to obtain a machine-translated text T1; when the number of sentences in the marked text S1 to be translated is not 1, unmatched words in each sentence in the marked text S1 to be translated are selected for machine translation. After the translation is completed, the words in the same sentence order are sorted according to the sentence order in the marked text to be translated, so that the words in the same sentence order are stored together to form a machine-translated text T1, which is convenient for subsequent replacement and improves the accuracy of replacement.

[0056] Furthermore, when machine translation is performed on the unmatched words in the to-be-translated marked text S1, each unmatched word carries its position order in the sentence.

[0057] As an example, in step S40, after obtaining the machine translated text T1, based on the word position identifier, the marked translation T2 in the translation memory is replaced with the machine translated text T1 through a word alignment tool to obtain the translation of the text to be translated.

[0058] Furthermore, after obtaining the translation of the text to be translated, the result can be fed back to professional translators for correction to ensure the accuracy of the translation.

[0059] This embodiment obtains the translation of the text to be translated through steps S10 to S40, which can effectively improve the accuracy of the translation, especially for some fields with high professional requirements. If machine translation is used directly to translate the text to be translated, if some special sentence patterns such as inversion are encountered, the effect of machine translation may be very poor.

[0060] In one embodiment, if Figure 2As shown, in step S10, the text to be translated is matched with the original text in the translation memory, and the text to be translated that does not match is marked as the marked text to be translated S1, which specifically includes the following steps:

[0061] S11: dividing the text to be translated into sentences by using sentence segmentation marks, and counting the number of sentences in the text to be translated;

[0062] S12: When the number of sentences in the text to be translated is less than a preset number, the translation memory is polled based on each sentence in the text to be translated, and sentences with inconsistent matches are marked to form the marked text to be translated S1;

[0063] S13: When the number of sentences in the text to be translated is greater than a preset number, the text to be translated and the original text in the translation memory are compared by a text comparison tool, and inconsistent texts are marked as marked text to be translated S1.

[0064] The segmentation mark refers to a mark that can divide the text content into a sentence. The segmentation mark in this embodiment is a mark that can segment the text content in any language, and is not limited here. Taking Chinese as an example, the segmentation mark can be a punctuation mark ".", "!", "?", ":", or other forms of identifiers.

[0065] As an example, in step S11-step S13, the text to be translated is divided into sentences by sentence segmentation marking, and the number of sentences in the text to be translated is counted. When the number of sentences is small, less than a preset number, each sentence in the text to be translated can be polled in the translation memory to query whether there is an original sentence matching the sentence in the text to be translated in the translation memory; when the number of sentences is large, greater than the preset number, if the translation memory is polled, the matching efficiency is low. Therefore, when the sentence data in the text to be translated is greater than the preset number, it is not suitable for polling matching. In this embodiment, the text to be translated is compared with the original text in the translation memory by a text comparison tool, and the inconsistent text in the text to be translated is marked as the marked text to be translated S1, and the inconsistent original text in the translation memory is marked as the marked original text S2.

[0066] The text comparison tool refers to a tool that can compare the contents of two texts. This embodiment does not limit the specific text comparison tool used, and any online text comparison tool or offline text comparison tool that can compare the contents of two texts can be used.

[0067] For ease of understanding, the following example is used for explanation: The text to be translated is: Smoking is currently considered to be the most important high-risk factor for lung cancer. Tobacco contains more than 3,000 chemical substances, among which polycyclic aromatic hydrocarbons have strong carcinogenic activity.

[0068] The original text in the translation memory is: At present, smoking is considered to be the most important high-risk factor for lung cancer. There are more than 3,000 chemical substances in tobacco, among which nitrosamines have strong carcinogenic activity.

[0069] The text to be translated is matched with the original text in the translation memory, and the inconsistent text to be translated "There are more than 3,000 chemical substances in tobacco, among which polycyclic aromatic hydrocarbons have strong carcinogenic activity" is marked as the marked text to be translated S1, and the inconsistent original text "There are more than 3,000 chemical substances in tobacco, among which nitrosamines have strong carcinogenic activity" is also marked as the marked original text S2. Based on the marked text to be translated S1 and the marked original text S2, the mismatched words are determined to be "polycyclic aromatic hydrocarbons" and "nitrosamines". Based on the marked original text S2, the corresponding translation is obtained from the translation memory through the word alignment tool as the mismatched translation "There are more than 3,000 chemical substances in tobacco, among which nitrosamines have strong carcinogenic activity", and then according to the mismatched word "nitrosamine" in the marked original text S2, the mismatched translation "There are morethan 3,000 chemical substances in tobacco, among which nitrosamines have strong carcinogenic activity" is never matched. activity” and identified “nitrosamines” as the marker translation T2.

[0070] The unmatched word "polychain aromatic hydrocarbon compounds" in the to-be-translated marked text S1 is machine translated to obtain a machine translated text T1 "polychain aromatic hydrocarbon compounds". The machine translated text T1 carries a word position identifier, that is, the position of the unmatched word in the to-be-translated marked text S1 in the sentence to be translated. Based on the word position identifier, the position of the unmatched word in the original text in the corresponding translation memory can be determined. After the position of the unmatched word in the original text in the translation memory is determined, the position of the marked translation T2 can be determined by a word alignment tool. Finally, the marked translation T2 "nitrosamines" in the translation memory is replaced with the machine translated text T1 "polychain aromatic hydrocarbon compounds", and the translation of the to-be-translated text "At present, smoking is considered to be the most important high-risk factor for lung cancer. There are more than 3,000 chemical substances in tobacco, among which n polychain aromatic hydrocarbon compounds have strong carcinogenic activity."

[0071] The above-mentioned machine translation auxiliary optimization method is adopted. Since it is to translate the unmatched words in the translation mark text S1 when machine translation is performed, it is not to translate the entire sentence. At present, the translation of each word by machine translation is still relatively accurate. It is mainly when encountering special sentence patterns such as inverted sentences that a large translation deviation will occur. Therefore, the above-mentioned machine translation auxiliary optimization method is adopted. Under the premise of not changing the sentence format, according to the word position mark, the translation of the machine translated word is replaced to the position corresponding to the marked translation in the translation memory by the word alignment tool, which can effectively improve the translation efficiency and accuracy. In addition, the above-mentioned machine translation auxiliary optimization method does not need to determine the matching rate of each sentence in the text to be translated and the original text in the translation memory, because no matter how large the matching rate of the two is, all unmatched words can be accurately machine translated without changing the sentence format.

[0072] In one embodiment, a machine translation assisted optimization device is provided, and the machine translation assisted optimization device corresponds one-to-one to a machine translation assisted optimization method in the above embodiment. Figure 3As shown, the machine translation auxiliary optimization device includes a text matching module 10, a marked translation acquisition module 20, a machine translation module 30 and a translation acquisition module 40. The detailed description of each functional module is as follows:

[0073] The text matching module 10 is used to obtain the text to be translated, and match the text to be translated with the original text in the translation memory, mark the text to be translated that is inconsistent with the match as the marked text to be translated S1, and also mark the original text that is inconsistent with the match as the marked original text S2;

[0074] A marked translation acquisition module 20, for acquiring a corresponding translation from a translation memory based on the marked original text S2 as a marked translation T2, wherein the marked translation T2 carries a word position identifier;

[0075] A machine translation module 30 is used to determine mismatched words based on the marked text to be translated S1 and the marked original text S2, and to perform machine translation on the mismatched words in the marked text to be translated S1 to obtain a machine translated text T1, wherein the machine translated text T1 carries a word position identifier;

[0076] The translation acquisition module 40 is used to replace the marked translation T2 in the translation memory with the machine translation text T1 based on the word position identifier and through the word alignment tool to acquire the translation of the text to be translated.

[0077] Furthermore, the text matching module 10 includes a sentence quantity counting unit, a first text-to-be-translated marking unit and a second text-to-be-translated marking unit.

[0078] A sentence number counting unit is used to divide the text to be translated into sentences by using sentence segmentation marks, and to count the number of sentences in the text to be translated;

[0079] A first to-be-translated text marking unit, for, when the number of sentences in the to-be-translated text is less than a preset number, polling in a translation memory based on each sentence in the to-be-translated text, marking sentences with inconsistent matches, and forming a to-be-translated marked text S1;

[0080] The second text-to-be-translated marking unit is used for comparing the text-to-be-translated with the original text in the translation memory by using a text comparison tool when the number of sentences in the text-to-be-translated is greater than a preset number, and marking inconsistent texts as marked text-to-be-translated S1.

[0081] For the specific definition of a machine translation assisted optimization device, please refer to the definition of a machine translation assisted optimization method above, which will not be repeated here. Each module in the above-mentioned machine translation assisted optimization device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0082] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a computer-readable storage medium and an internal memory. The computer-readable storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the computer-readable storage medium. The database of the computer device is used to store data involved in a machine translation-assisted optimization method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a machine translation-assisted optimization method is implemented.

[0083] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a machine translation assisted optimization method in the above embodiment are implemented, such as Figure 1 Steps S10-S40 shown, or Figure 2 Alternatively, when the processor executes the computer program, the functions of each module / unit of the machine translation auxiliary optimization device in the above embodiment are realized, for example Figure 3 The functions of the modules 10 to 40 are shown in Figure 1. To avoid repetition, they will not be described here.

[0084] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of a machine translation assisted optimization method in the above embodiment are implemented, for example: Figure 1 Steps S10-S40 shown, or Figure 2 Alternatively, when the processor executes the computer program, the functions of each module / unit in the embodiment of the machine translation auxiliary optimization device are realized, for example Figure 3 The functions of the modules 10 to 40 are shown in Figure 1. To avoid repetition, they will not be described here.

[0085] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0086] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0087] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A machine translation assisted optimization method, characterized in that: include: Obtain the text to be translated, and match the text to be translated with the original text in the translation memory, mark the text to be translated that is inconsistent with the match as the marked text to be translated S1, and also mark the original text that is inconsistent with the match as the marked original text S2; Based on the marked original text S2, a corresponding translation is obtained from the translation memory as the marked translation T2, wherein the marked translation T2 carries a word position identifier; Determine mismatched words based on the marked text to be translated S1 and the marked original text S2, and perform machine translation on the mismatched words in the marked text to be translated S1 to obtain a machine-translated text T1, wherein the machine-translated text T1 carries a word position identifier; based on the word position identifier, replace the marked translation T2 in the translation memory with the machine-translated text T1 through a word alignment tool to obtain a translation of the text to be translated; The step of matching the text to be translated with the original text in the translation memory and marking the text to be translated that does not match the original text as marked text to be translated S1 includes: The text to be translated is divided into sentences by using sentence segmentation marks, and the number of sentences in the text to be translated is counted; when the number of sentences in the text to be translated is less than a preset number, the translation memory is polled based on each sentence in the text to be translated, and the sentences with inconsistent matches are marked to form a marked text to be translated S1; when the number of sentences in the text to be translated is greater than the preset number, the text to be translated is compared with the original text in the translation memory by using a text comparison tool, and the inconsistent text is marked as the marked text to be translated S1; The method performs machine translation on the unmatched words in the marked text to be translated S1 to obtain the machine translated text T1, including: determining the number of sentences in the marked text to be translated S1 according to the sentence segmentation identifier, and when the number of sentences in the marked text to be translated S1 is 1, selecting unmatched words from the marked text to be translated S1 for machine translation to obtain the machine translated text T1.

2. The machine translation assisted optimization method according to claim 1, characterized in that: The step of obtaining a corresponding translation from a translation memory as a marked translation T2 based on the marked original text S2 includes: Determine the mismatched words based on the to-be-translated marked text S1 and the marked original text S2, and determine the corresponding translations as the mismatched translations from the translation memory through a word alignment tool based on the marked original text S2; The marked translation T2 is determined from the unmatched translations based on the unmatched words in the marked original text S2.

3. The machine translation assisted optimization method according to claim 1, characterized in that: When the number of sentences in the to-be-translated marked text S1 is not 1, after selecting unmatched words from the to-be-translated marked text S1 for machine translation to obtain the machine translated text T1, the machine translation auxiliary optimization method further includes: According to the sentence order in the marked text to be translated, the words in the same sentence order are sorted to obtain the machine translated text T1.

4. The machine translation assisted optimization method according to claim 1, characterized in that: The machine translation assisted optimization method also includes: When the text to be translated matches the original text in the translation memory, the word alignment tool selects the corresponding translation from the translation memory based on the original text as the translation of the text to be translated.

5. A machine translation auxiliary optimization device, characterized in that: include: The text matching module is used to obtain the text to be translated, and match the text to be translated with the original text in the translation memory, mark the text to be translated that is inconsistent with the match as the marked text to be translated S1, and also mark the original text that is inconsistent with the match as the marked original text S2; A marked translation acquisition module, used to acquire a corresponding translation from a translation memory based on the marked original text S2 as a marked translation T2, wherein the marked translation T2 carries a word position identifier; A machine translation module, for determining mismatched words based on the to-be-translated marked text S1 and the marked original text S2, and performing machine translation on the mismatched words in the to-be-translated marked text S1 to obtain a machine-translated text T1, wherein the machine-translated text T1 carries a word position identifier; A translation acquisition module is used to replace the marked translation T2 in the translation memory with the machine translation text T1 based on the word position identifier and through the word alignment tool to obtain the translation of the text to be translated; The text matching module comprises: A sentence number counting unit is used to divide the text to be translated into sentences by using sentence segmentation marks, and to count the number of sentences in the text to be translated; A first to-be-translated text marking unit, for, when the number of sentences in the to-be-translated text is less than a preset number, polling in a translation memory based on each sentence in the to-be-translated text, marking sentences with inconsistent matches, and forming a to-be-translated marked text S1; The second to-be-translated text marking unit is used for comparing the to-be-translated text with the original text in the translation memory through a text comparison tool when the number of sentences in the to-be-translated text is greater than a preset number, and marking inconsistent texts as to-be-translated marked texts S1; The machine translation module performs machine translation on the unmatched words in the to-be-translated marked text S1 to obtain a machine translated text T1, including: The number of sentences in the to-be-translated marked text S1 is determined according to the sentence segmentation identifiers. When the number of sentences in the to-be-translated marked text S1 is 1, unmatched words are selected from the to-be-translated marked text S1 for machine translation to obtain a machine-translated text T1.

6. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the machine translation assisted optimization method according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the machine translation assisted optimization method according to any one of claims 1 to 4 is implemented.

Images