Marketing text translation method and apparatus, electronic device, and medium

Abstract

Embodiments of the present application disclose a marketing text translation method and device, electronic equipment and a storage medium, which are adapted to the language, cultural customs, customer language habits and marketing scene characteristics of the target region for localized adaptation translation, solve the poor cultural adaptability problem of the existing method, make the translated text more consistent with the communication needs of the target region, and meanwhile, the first large model is used to realize the targeted adjustment of the reference text, the second large model is used to realize the semantic consistency detection of the translated text and the original text, through the division and cooperation of the two models, the core semantics and marketing appeal of the translated text are ensured to be consistent with the original text, the problems of semantic deviation and information loss are avoided, and thus the accuracy of the marketing text translation is provided.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, specifically to a marketing text translation method, apparatus, electronic device, and storage medium. Background Technology

[0002] With the development of natural language processing technology, machine translation technology has been applied to cross-language marketing text dissemination scenarios. By translating marketing text from the source language to the target language of the target region, marketing information can be transmitted across regions. Existing marketing text translation methods typically only perform a single-dimensional language conversion, without adapting to the cultural characteristics of the target region, leading to cultural conflicts in the translated text. Furthermore, existing methods lack effective translation result verification mechanisms, relying solely on simple text comparison to judge translation accuracy. This fails to accurately identify semantic deviations, missing core information, and other issues. The lack of iterative adjustment processes and threshold controls can result in endless iterations or incomplete adjustments, reducing translation efficiency. In addition, for multilingual marketing texts, existing methods do not identify the core expressive language, leading to confusion in the back-translation reference language and further affecting the accuracy of the translation results. All these problems result in insufficient accuracy and adaptability of marketing text translation, failing to meet the needs of cross-regional marketing communication. Summary of the Invention

[0003] This application provides a marketing text translation method, electronic device, apparatus, and storage medium to improve the accuracy and localization adaptability of marketing text translation, optimize translation efficiency, and realize the visualization of the translation process.

[0004] In a first aspect, embodiments of this application provide a marketing text translation method, including: The system obtains marketing text input by the user and the target region, and translates the marketing text based on the language and cultural characteristics of the target region to obtain translated text. The translated text is back-translated into its original language to obtain a reference text in the same language as the marketing text; Determine the degree of content difference between the marketing text and the reference text; If the content difference is greater than or equal to a preset value, the reference text is adjusted according to the content difference and the preset first model, and the process returns to the step of determining the content difference between the marketing text and the reference text until the content difference between the marketing text and the reference text is less than the preset value. The semantic consistency between the translated text and the marketing text is detected using a pre-defined second major model; When the translated text is semantically consistent with the marketing text, the translated text is output.

[0005] Optionally, in some embodiments of this application, the step of translating the marketing text based on the language and cultural characteristics of the target region to obtain translated text includes: Obtain the official language, regional cultural customs, target customer group's language usage habits, and marketing scenario adaptation characteristics of the target region; Based on the official language, regional cultural customs, target customer language usage habits, and marketing scenario adaptation characteristics, the marketing text is translated to obtain the translated text.

[0006] Optionally, in some embodiments of this application, determining the content difference between the marketing text and the reference text includes: Determine the preset analysis dimensions; Based on the aforementioned analytical dimensions, the marketing text and the reference text are subjected to dimensional difference quantification processing to obtain the difference quantification value of each dimension; The content difference between the marketing text and the reference text is obtained by weighting the difference values ​​of each dimension based on the preset weights of each dimension.

[0007] Optionally, in some embodiments of this application, adjusting the reference text based on the content difference and a preset first major model includes: Based on the first major model, the numerical value of the content difference degree and the distribution of the difference dimensions, the content in the reference text corresponding to the difference dimensions is adjusted. During the adjustment process, the core semantics of the marketing text and the expression habits of the original language are preserved, and the cultural characteristics of the target region are adapted.

[0008] Optionally, in some embodiments of this application, it further includes: When the content difference is greater than or equal to a preset value, the reference text is adjusted and the relevant steps are repeated. If the number of iterations reaches the threshold of the number of iterations but the content difference is still greater than or equal to the preset value, the first model outputs a reference text adjustment and optimization suggestion. Based on the adjustment and optimization suggestion, the reference text is manually adjusted, and the step of determining the content difference between the marketing text and the reference text is executed again until the content difference is less than the preset value.

[0009] Optionally, in some embodiments of this application, it further includes: Redundant information, garbled characters, and invalid symbols are removed from the marketing text to obtain standardized marketing text; The translation processing of the marketing text based on the language and cultural characteristics of the target region includes: translating the standardized marketing text based on the language and cultural characteristics of the target region.

[0010] Optionally, in some embodiments of this application, it further includes: Identify the language used in the standardized marketing text; The language used to express the text will be determined as the base language for the back-translation process of the translated text.

[0011] Optionally, in some embodiments of this application, it further includes: In the interactive mode of marketing text translation, the marketing text entered by the user and the selected target region information are displayed; After completing the translation and back-translation of the marketing text, the translated text, the reference text, and the content differences between the marketing text and the reference text are displayed simultaneously. During the iterative adjustment of the reference text, the reference text after each round of adjustment and the corresponding content difference change information are displayed in real time. After using the second largest model to complete the semantic consistency detection, the semantic consistency detection results between the translated text and the marketing text are displayed; If the detection result is semantically consistent, the translated text is determined as the final translated text and highlighted.

[0012] Optionally, in some embodiments of this application, it further includes: Display a control to enable the marketing text translation mode, which is associated with a target region selection sub-control and a translation parameter setting sub-control; In response to the activation operation of the control for the marketing text translation mode, the target region selection sub-control and the translation parameter setting sub-control are displayed. The target region selection sub-control is used to select the target region, and the translation parameter setting sub-control is used to set the preset value of content difference and the threshold of the number of iterations.

[0013] Optionally, in some embodiments of this application, the simultaneous display of the translated text, the reference text, and the content differences between the marketing text and the reference text includes: The display interface is divided into a marketing text display area, a translated text display area, a reference text display area, and a difference display area. There is no overlap between the display areas and the layout is adapted to each other. In the reference text display area, the parts that differ from the marketing text are marked and displayed, and in the difference degree display area, the quantitative value of the content difference degree and the difference dimension distribution information are displayed simultaneously.

[0014] Optionally, in some embodiments of this application, the method further includes: When displaying the semantic consistency detection results, a detection details control and a manual text adjustment control are shown; The detection details control is used to display the semantic consistency detection details of the second model from three dimensions: semantics, core marketing information, and sentiment tendency. The manual text adjustment control is used to manually modify the translated text or reference text.

[0015] Optionally, in some embodiments of this application, the step of determining the translated text as the final translated text and highlighting it if the detection result is semantically consistent includes: If the detection result is semantic inconsistency, in response to the operation of manually adjusting the control for the text, a text editing area is displayed, and the translated text or reference text is manually modified based on the text editing area; The modified translated text is then re-translated, and the steps of back-translation, difference determination, iterative adjustment, and semantic consistency detection are repeated until the detection result is semantically consistent. The translated text at this point is then identified as the final translated text and highlighted, while a translation completion indicator is displayed.

[0016] Secondly, embodiments of this application provide a marketing text translation device, comprising: The acquisition module is used to acquire the marketing text input by the user and the target region, and to translate the marketing text based on the language and cultural characteristics of the target region to obtain the translated text; The processing module is used to perform source language back-translation processing on the translated text to obtain a reference text in the same language as the marketing text; A determination module is used to determine the degree of content difference between the marketing text and the reference text; The adjustment module is used to adjust the reference text according to the content difference and the preset first large model if the content difference is greater than or equal to a preset value, and return to the step of determining the content difference between the marketing text and the reference text until the content difference between the marketing text and the reference text is less than the preset value. The detection module is used to detect the semantic consistency between the translated text and the marketing text using a preset second major model; The output module is used to output the translated text when the translated text is semantically consistent with the marketing text.

[0017] Accordingly, this application also provides an electronic device, including a memory, a processor, and a processor program stored in the memory and executable on the processor, wherein the processor executes the program as described in any of the methods above.

[0018] This application also provides a storage medium storing a processor program that, when executed by a processor, implements any of the methods described above.

[0019] This application provides a marketing text translation method, apparatus, electronic device, and storage medium. The method involves acquiring marketing text input by a user and a target region, translating the marketing text based on the language and cultural characteristics of the target region to obtain translated text, performing source language back-translation on the translated text to obtain reference text in the same language as the marketing text, determining the content difference between the marketing text and the reference text, and if the content difference is greater than or equal to a preset value, adjusting the reference text according to the content difference and a preset first major model, and returning to the step of determining the content difference between the marketing text and the reference text until the content difference between the marketing text and the reference text is less than the preset value. The second pre-set model is used to detect the semantic consistency between the translated text and the marketing text. When the translated text and the marketing text are semantically consistent, the translated text is output. In the solution provided in this application, localized adaptation translation is performed by combining the language, cultural customs, customer language habits and marketing scenario characteristics of the target region. This solves the problem of poor cultural adaptability of existing methods and makes the translated text more in line with the communication needs of the target region. At the same time, the first model is used to make targeted adjustments to the reference text, and the second model is used to detect the semantic consistency between the translated text and the original text. Through the division of labor and cooperation of the two models, the core semantics and marketing appeal of the translated text are ensured to be consistent with the original text, avoiding semantic deviation and information loss. Thus, the accuracy of marketing text translation is improved. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the marketing text translation method provided in this application embodiment; Figure 2 This is a schematic diagram of the marketing text translation device provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0022] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0023] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, components, features, and elements with the same names in different embodiments of this application may have the same meaning or different meanings, the specific meaning of which must be determined by its interpretation in that specific embodiment or further in conjunction with the context of that specific embodiment.

[0024] It should be noted that the marketing text mentioned in this application refers to various texts used for commercial marketing communication, including but not limited to product descriptions on e-commerce platforms, brand promotion texts on social media, promotional texts for offline activities, and text content on product posters. Its core characteristics are that it contains clear marketing appeals, key product / brand information, and corresponding emotional inclinations. The target region refers to the geographical scope to which the marketing text is intended for dissemination, which can be a country, province, city, or other geographical level. The first large model refers to a large natural language processing model with text difference analysis and targeted text adjustment capabilities, fine-tuned for marketing text translation scenarios to adapt to the language features and adjustment needs of marketing texts. The second large model refers to a large natural language processing model with semantic consistency detection capabilities, fine-tuned for marketing text semantic detection scenarios to achieve accurate detection from dimensions such as semantics, core marketing information, and emotional inclinations. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0025] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.

[0026] The following describes in detail the embodiments involved in this application. It should be noted that the order of description of the embodiments in this application is not intended to limit the priority of the embodiments.

[0027] This application provides a marketing text translation method, apparatus, storage medium, and smart terminal. Specifically, the marketing text translation method of this application can be executed by a smart terminal or a server, wherein the smart terminal can be a terminal. The terminal can be a smartphone, tablet computer, laptop computer, touch screen, game console, personal computer (PC), personal digital assistant (PDA), or other smart terminal. The terminal may also include a client, which can be a media playback client or an instant marketing text translation client, etc.

[0028] This application provides a marketing text translation method, which can be executed by an electronic device or a server. This application example illustrates the marketing text translation method executed by an electronic device. The electronic device includes a touchscreen display and a processor. The touchscreen display is used to present a graphical user interface (GUI) and receive operation commands generated by the user interacting with the GUI. When the user operates the GUI through the touchscreen display, the GUI can control the content locally on the electronic device in response to the received operation commands, or it can control the content on the server side in response to the received operation commands.

[0029] The marketing text translation solution provided in this application localizes and adapts the translation to the target region's language, cultural customs, customer language habits, and marketing scenario characteristics. This solves the problem of poor cultural adaptability in existing methods, making the translated text more in line with the communication needs of the target region. At the same time, it uses a first major model to make targeted adjustments to the reference text and a second major model to detect the semantic consistency between the translated text and the original text. Through the division of labor and cooperation between the two models, it ensures that the core semantics and marketing appeal of the translated text are consistent with the original text, avoiding semantic deviations and information loss. Thus, it provides accuracy for marketing text translation.

[0030] The following sections provide detailed descriptions of each example. It should be noted that the order in which the embodiments are described is not intended to limit the priority of the embodiments.

[0031] A marketing text translation method includes: acquiring marketing text input by a user and a target region, and translating the marketing text based on the language and cultural characteristics of the target region to obtain translated text; performing source language back-translation on the translated text to obtain reference text in the same language as the marketing text; determining the content difference between the marketing text and the reference text; if the content difference is greater than or equal to a preset value, adjusting the reference text according to the content difference and a preset first major model, and returning to the step of determining the content difference between the marketing text and the reference text, until the content difference between the marketing text and the reference text is less than the preset value; using a preset second major model to detect the semantic consistency between the translated text and the marketing text; and outputting the translated text when the translated text is semantically consistent with the marketing text.

[0032] Please see Figure 1 , Figure 1 This application provides a flowchart illustrating the marketing text translation method. The specific flow of this marketing text translation method is as follows: 101. Obtain the marketing text input by the user and the target region, and translate the marketing text based on the language and cultural characteristics of the target region to obtain the translated text.

[0033] Users input marketing text to be translated through the terminal's input interface and select the target region for the marketing text using the target region selection control. The terminal then transmits the user-input marketing text and target region information to the server via the network. The server receives and stores this information. The marketing text can be plain text or a text containing multiple languages. The target region information includes the region name, and the server can match the corresponding language and cultural characteristics based on the region name.

[0034] 102. Perform back-translation processing on the translated text to obtain a reference text in the same language as the marketing text.

[0035] Based on the target region information, obtain the official language, regional cultural customs, target customer group's language usage habits, and marketing scenario adaptation characteristics of the target region: the official language is the legally used language of the target region; regional cultural customs are the unique cultural traditions, language taboos, and expression habits of the target region; target customer group's language usage habits are the daily language expression characteristics of the target customer group that the marketing text is intended to reach (such as the online language habits of younger customers, the formal expression habits of business customers, etc.); and marketing scenario adaptation characteristics are the language requirements of the marketing text dissemination scenarios (such as the colloquial characteristics of social media scenarios, the written characteristics of e-commerce detail pages, etc.).

[0036] Before translating marketing text, it undergoes preprocessing to remove non-marketing redundant information, garbled characters, and invalid symbols. Non-marketing redundant information includes, but is not limited to, emojis, irrelevant web links, system identifiers, advertising characters, and meaningless spaces. Garbled characters include text errors caused by input or transmission issues, and invalid symbols include special characters without practical meaning. This preprocessing eliminates the interference of redundant information on the translation result, yielding a standardized marketing text with a neat format and clean information, which serves as the input text for subsequent translation processing.

[0037] If the standardized marketing text contains multilingual content, the server identifies its core descriptive language and uses natural language processing (NLP) technology to analyze the descriptive language corresponding to the core semantics of the text, determining this language as the base language for subsequent translation. If the standardized marketing text contains only a single language, that language is directly designated as the base language for translation. This step resolves the issue of inconsistent base languages ​​in the translation of multilingual marketing text, ensuring the accuracy of the translation results.

[0038] Furthermore, machine translation technology is used to convert the translated text into text in the same language as the original marketing text, resulting in a reference text. The reference text serves as a verification medium between the original marketing text and the translated text, and is used for subsequent discrepancy analysis.

[0039] 103. Determine the content difference between the marketing text and the reference text.

[0040] The content differences between the two were determined through multi-dimensional quantitative analysis and weighted calculation, specifically including: (1) Determine the preset analysis dimensions. The analysis dimensions are set in advance according to the characteristics of the marketing text, including but not limited to semantic expression, core marketing information, emotional tendency, sentence logic, etc. Each dimension is a key dimension that affects the dissemination effect of the marketing text. (2) Based on the above analysis dimensions, the original marketing text and the reference text are subjected to dimension difference quantification processing, that is, text comparison is performed on each dimension, the degree of difference between the two in that dimension is analyzed, and the degree of difference is converted into a quantitative value between 0 and 1 (the larger the value, the greater the difference), and the difference quantification value of each dimension is obtained. (3) Based on the preset weights of each dimension, the difference quantification values ​​of each dimension are weighted and calculated to obtain the overall content difference between the two. Among them, the weights of each dimension are preset according to the type of marketing text. For example, the emotional tendency dimension of brand promotion copy has a higher weight, and the core marketing information dimension of product introduction copy has a higher weight. The sum of the weights of each dimension is 1.

[0041] 104. If the content difference is greater than or equal to a preset value, the reference text is adjusted according to the content difference and the preset first major model, and the process returns to the step of determining the content difference between the marketing text and the reference text until the content difference between the marketing text and the reference text is less than the preset value.

[0042] The calculated content difference score is compared with a preset value, which can be set by the user through the translation parameter setting sub-control or a system default value (e.g., 0.2). This comparison determines whether the difference between the translated text and the original marketing text is within an acceptable range. If the content difference score is greater than or equal to the preset value, it indicates a significant difference between the translated text and the original marketing text, requiring iterative adjustments to the reference text. If the content difference score is less than the preset value, it indicates the difference is within an acceptable range, requiring no adjustment, and the process proceeds to the subsequent semantic consistency detection step.

[0043] Furthermore, if the content difference is greater than or equal to a preset value, then based on the first major model, the magnitude of the content difference, and the distribution of difference dimensions, targeted adjustments are made to the content in the reference text corresponding to the difference dimensions. For example, if the difference quantification value of the core marketing information dimension is the highest, then the core marketing information in the reference text is adjusted to ensure consistency with the original marketing text; if the difference in the sentiment dimension is large, then the sentiment expression of the reference text is adjusted to match the original marketing text. During the adjustment process, the core semantics of the original marketing text and the expression habits of the original language must be preserved, while ensuring that the translated text corresponding to the adjusted reference text is still suitable for the cultural characteristics of the target region.

[0044] The content difference determination step is re-executed for the adjusted reference text, and the judgment is made again. At the same time, a threshold for the number of iterations is set (e.g., 5 times). If the number of iterations reaches the threshold but the content difference is still greater than or equal to the preset value, it means that the automatic adjustment of the model cannot solve the current difference problem. At this time, the first model outputs a reference text adjustment and optimization suggestion, which is displayed on the terminal. The user manually adjusts the reference text according to the suggestion. After manual adjustment, the content difference determination step is executed again until the content difference is less than the preset value.

[0045] 105. Use a pre-set second major model to detect the semantic consistency between the translated text and the marketing text.

[0046] The translated text and the original marketing text are input into the second model, which performs semantic consistency detection on the two. The detection dimensions include three core dimensions: semantics, core marketing information, and sentiment. The second model analyzes whether the translated text is consistent with the original marketing text in each dimension and outputs the overall semantic consistency detection result (consistent / inconsistent) and the detection details of each dimension.

[0047] 106. When the translated text is semantically consistent with the marketing text, output the translated text. If the detection result is semantically consistent, it means that the translated text not only fits the language and cultural characteristics of the target region, but also matches the core semantics and marketing appeal of the original marketing text. In this case, the translated text is output.

[0048] Optionally, in some embodiments of this application, it further includes: In the interactive mode of marketing text translation, the marketing text entered by the user and the selected target region information are displayed; After completing the translation and back-translation of the marketing text, the translated text, the reference text, and the content differences between the marketing text and the reference text are displayed simultaneously. During the iterative adjustment of the reference text, the reference text after each round of adjustment and the corresponding content difference change information are displayed in real time. After using the second largest model to complete the semantic consistency detection, the semantic consistency detection results between the translated text and the marketing text are displayed; If the detection result is semantically consistent, the translated text is determined as the final translated text and highlighted.

[0049] The terminal displays a marketing text translation mode activation control in the interactive interface. This control is a visual control such as a button or switch, and is associated with a target region selection sub-control and a translation parameter setting sub-control. When the user triggers an activation operation on the activation control (such as clicking or swiping), the terminal responds to the operation and displays the above two sub-controls in the interactive interface, allowing the user to select the target region and set translation parameters (preset value for content difference, threshold for number of iterations).

[0050] The terminal enters the interactive mode for marketing text translation. The interactive interface displays the marketing text to be translated entered by the user, as well as the target region information selected by the user through the target region selection sub-control, allowing the user to intuitively view the basic information entered.

[0051] After the server completes the translation and back-translation of the marketing text, the terminal receives the translated text, reference text, and content difference information transmitted by the server, and displays the above information synchronously in the interactive interface, specifically including: (1) The interactive interface is divided into a marketing text display area, a translation text display area, a reference text display area and a difference display area. There is no overlap between the display areas, and the layout is adapted to the screen size of the terminal to ensure the comfort of information viewing. (2) In the reference text display area, the parts of the reference text that differ from the original marketing text are marked (such as red, bold, underlined, etc.) so that users can clearly see the differences; (3) In the difference display area, the quantitative value of the content difference and the distribution information of the difference dimensions are displayed simultaneously (such as the quantitative value of the core marketing information dimension difference is 0.3, and the quantitative value of the emotional tendency dimension difference is 0.1), so that users can understand the specific situation of the difference.

[0052] During the server's iterative adjustments to the reference text, the terminal receives real-time updates of the adjusted reference text and corresponding content difference changes from the server for each round of adjustments. These updates are displayed in the reference text and difference display areas, allowing users to intuitively grasp the process and effects of the iterative adjustments. If the model automatically adjusts to the iteration threshold and outputs optimization suggestions, the terminal displays these suggestions on the interactive interface for users to manually adjust the reference text.

[0053] After the second model completes the semantic consistency detection between the translated text and the original marketing text, the terminal receives the detection results transmitted from the server and displays the results in the interactive interface; at the same time, the terminal displays detection details controls and manual text adjustment controls: (1) Detection details control: When the user triggers this control, the terminal displays the semantic consistency detection details of the second model from three dimensions: semantics, core marketing information, and sentiment tendency, including the detection results and differences of each dimension; (2) Manual text adjustment control: When the user triggers this control, the terminal displays a text editing area for the user to manually modify the translated text or reference text.

[0054] If the semantic consistency detection result is consistent, the terminal determines the translated text as the final translated text and highlights it in the translated text display area (such as blue highlighting, background highlighting, etc.). At the same time, a translation completion mark is displayed in the interactive interface to prompt the user that the translation process is complete. If the semantic consistency detection result is inconsistent, the terminal responds to the user's manual adjustment of the text control, allowing the user to manually modify the translated text or reference text in the text editing area. After the user completes the modification, the terminal transmits the modified text to the server. The server then re-executes the steps of back-translation, difference determination, iterative adjustment, and semantic consistency detection on the modified translated text until the detection result is semantically consistent, at which point the final translated text is highlighted. To facilitate better implementation of the marketing text translation method of this application embodiment, this application embodiment also provides a marketing text translation device, wherein the meanings of the terms are the same as those in the marketing text translation system described above, and specific implementation details can be found in the description of the system embodiment.

[0055] Please see Figure 2 , Figure 2The diagram below illustrates the structure of a marketing text translation device provided in this application embodiment. Specifically, the marketing text translation device may include an acquisition module 201, a processing module 202, a determination module 203, an adjustment module 204, a detection module 205, and an output module 206, as follows: The acquisition module 201 is used to acquire the marketing text input by the user and the target region, and to translate the marketing text based on the language and cultural characteristics of the target region to obtain translated text; Processing module 202 is used to perform source language back-translation processing on the translated text to obtain reference text in the same language as the marketing text; The determination module 203 is used to determine the content difference between the marketing text and the reference text; The adjustment module 204 is used to adjust the reference text according to the content difference degree and the preset first large model if the content difference degree is greater than or equal to the preset value, and return to the step of determining the content difference degree between the marketing text and the reference text until the content difference degree between the marketing text and the reference text is less than the preset value. The detection module 205 is used to detect the semantic consistency between the translated text and the marketing text using a preset second major model; The output module 206 is used to output the translated text when the translated text is semantically consistent with the marketing text.

[0056] This application provides a marketing text translation device. An acquisition module 201 acquires marketing text input by a user and a target region, and translates the marketing text based on the language and cultural characteristics of the target region to obtain translated text. A processing module 202 performs source language back-translation on the translated text to obtain reference text in the same language as the marketing text. A determination module 203 determines the content difference between the marketing text and the reference text. An adjustment module 204, if the content difference is greater than or equal to a preset value, adjusts the reference text according to the content difference and a preset first large model, and returns to the step of determining the content difference between the marketing text and the reference text, until the content difference between the marketing text and the reference text is less than the preset value. The detection module 205 uses a preset second major model to detect the semantic consistency between the translated text and the marketing text; the output module 206 outputs the translated text when the semantics of the translated text and the marketing text are consistent. In the solution provided in this application, localized adaptation translation is carried out by combining the language, cultural customs, customer language habits and marketing scenario characteristics of the target region, which solves the problem of poor cultural adaptability of existing methods and makes the translated text more in line with the communication needs of the target region. At the same time, the first major model is used to realize the targeted adjustment of the reference text, and the second major model realizes the semantic consistency detection between the translated text and the original text. Through the division of labor and cooperation of the two models, the core semantics and marketing appeal of the translated text are ensured to be consistent with the original text, avoiding semantic deviation and information loss. Thus, the accuracy of marketing text translation is provided. Furthermore, embodiments of this application also provide an electronic device, such as... Figure 3 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically: The electronic device may include components such as a processor 301 with one or more processing cores, a memory 302 with one or more processor-readable storage media, a power supply 303, and an input unit 304. Those skilled in the art will understand that... Figure 3 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: Processor 301 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in memory 302, and by calling data stored in memory 302, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Optionally, processor 301 may include one or more processing cores; preferably, processor 301 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless marketing text translation. It is understood that the aforementioned modem processor may not be integrated into processor 301.

[0057] The memory 302 can be used to store software programs and modules. The process 301 executes various functional applications and marketing text translation methods by running the software programs and modules stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 302 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 302 may also include a memory controller to provide the process 301 with access to the memory 302.

[0058] The electronic device also includes a power supply 303 that supplies power to various components. Preferably, the power supply 303 can be logically connected to the processor 301 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 303 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0059] The electronic device may also include an input unit 304, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0060] Although not shown, the electronic device may also include a display unit, etc., which will not be described in detail here. Specifically, in the embodiments of this application, the processing 301 in the electronic device loads the executable files corresponding to the processes of one or more applications into the memory 302 according to the following instructions, and the processing 301 runs the applications stored in the memory 302 to realize various functions, as follows: The system acquires marketing text and target region input by the user, and translates the marketing text based on the language and cultural characteristics of the target region to obtain translated text. It then performs back-translation of the translated text into the original language to obtain reference text in the same language as the marketing text. The system determines the content difference between the marketing text and the reference text. If the content difference is greater than or equal to a preset value, the system adjusts the reference text according to the content difference and a preset first major model, and returns to the step of determining the content difference between the marketing text and the reference text until the content difference is less than the preset value. A preset second major model is used to detect the semantic consistency between the translated text and the marketing text. When the translated text is semantically consistent with the marketing text, the system outputs the translated text.

[0061] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0062] This application's embodiments combine the target region's language, cultural customs, customer language habits, and marketing scenario characteristics to perform localized adaptation translation, solving the problem of poor cultural adaptability in existing methods. This makes the translated text more in line with the communication needs of the target region. At the same time, a first major model is used to make targeted adjustments to the reference text, and a second major model is used to detect the semantic consistency between the translated text and the original text. Through the division of labor and cooperation between the two models, it is ensured that the core semantics and marketing appeals of the translated text are consistent with the original text, avoiding semantic deviations and information loss. Thus, it provides accuracy for marketing text translation.

[0063] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a processor-readable storage medium and loaded and executed by a processor.

[0064] Therefore, embodiments of this application provide a storage medium storing a plurality of instructions that can be loaded by a processor to execute steps in any of the marketing text translation methods provided in embodiments of this application. For example, the instructions can execute the following steps: The system acquires marketing text and target region input by the user, and translates the marketing text based on the language and cultural characteristics of the target region to obtain translated text. It then performs back-translation of the translated text into the original language to obtain reference text in the same language as the marketing text. The system determines the content difference between the marketing text and the reference text. If the content difference is greater than or equal to a preset value, the system adjusts the reference text according to the content difference and a preset first major model, and returns to the step of determining the content difference between the marketing text and the reference text until the content difference is less than the preset value. A preset second major model is used to detect the semantic consistency between the translated text and the marketing text. When the translated text is semantically consistent with the marketing text, the system outputs the translated text.

[0065] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0066] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0067] Since the instructions stored in the storage medium can execute the steps in any of the marketing text translation methods provided in the embodiments of this application, the beneficial effects that any of the marketing text translation methods provided in the embodiments of this application can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0068] The above provides a detailed description of a marketing text translation method, apparatus, electronic device, and storage medium provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for translating marketing texts, characterized in that, include: The system obtains the marketing text input by the user and the target region, and translates the marketing text based on the language and cultural characteristics of the target region to obtain the translated text. The translated text is back-translated into its original language to obtain a reference text in the same language as the marketing text; Determine the degree of content difference between the marketing text and the reference text; If the content difference is greater than or equal to a preset value, the reference text is adjusted according to the content difference and the preset first model, and the process returns to the step of determining the content difference between the marketing text and the reference text until the content difference between the marketing text and the reference text is less than the preset value. The semantic consistency between the translated text and the marketing text is detected using a pre-defined second major model; When the translated text is semantically consistent with the marketing text, the translated text is output.

2. The marketing text translation method according to claim 1, characterized in that, The process of translating the marketing text based on the language and cultural characteristics of the target region to obtain the translated text includes: Obtain the official language, regional cultural customs, target customer group's language usage habits, and marketing scenario adaptation characteristics of the target region; Based on the official language, regional cultural customs, target customer language usage habits, and marketing scenario adaptation characteristics, the marketing text is translated to obtain the translated text.

3. The marketing text translation method according to claim 1, characterized in that, Determining the content difference between the marketing text and the reference text includes: Determine the preset analysis dimensions; Based on the aforementioned analytical dimensions, the marketing text and the reference text are subjected to dimensional difference quantification processing to obtain the difference quantification value of each dimension; The content difference between the marketing text and the reference text is obtained by weighting the difference values ​​of each dimension based on the preset weights of each dimension.

4. The marketing text translation method according to claim 1, characterized in that, The adjustment of the reference text based on the content difference and the preset first major model includes: Based on the first major model, the numerical value of the content difference degree and the distribution of the difference dimensions, the content in the reference text corresponding to the difference dimensions is adjusted. During the adjustment process, the core semantics of the marketing text and the expression habits of the original language are preserved, and the cultural characteristics of the target region are adapted.

5. The marketing text translation method according to claim 1, characterized in that, Also includes: When the content difference is greater than or equal to a preset value, the reference text is adjusted and the relevant steps are repeated. If the number of iterations reaches the threshold of the number of iterations but the content difference is still greater than or equal to the preset value, the first model outputs a reference text adjustment and optimization suggestion. Based on the adjustment and optimization suggestion, the reference text is manually adjusted, and the step of determining the content difference between the marketing text and the reference text is executed again until the content difference is less than the preset value.

6. The marketing text translation method according to claim 1, characterized in that, After obtaining the marketing text input by the user and the target region, and before translating the marketing text based on the language and cultural characteristics of the target region, the process further includes: Redundant information, garbled characters, and invalid symbols are removed from the marketing text to obtain standardized marketing text; The translation processing of the marketing text based on the language and cultural characteristics of the target region includes: translating the standardized marketing text based on the language and cultural characteristics of the target region.

7. The marketing text translation method according to claim 6, characterized in that, Also includes: Identify the language used in the standardized marketing text; The language used to express the text will be determined as the base language for the back-translation process of the translated text.

8. The marketing text translation method according to any one of claims 1 to 7, characterized in that, Also includes: In the interactive mode of marketing text translation, the marketing text entered by the user and the selected target region information are displayed; After completing the translation and back-translation of the marketing text, the translated text, the reference text, and the content differences between the marketing text and the reference text are displayed simultaneously. During the iterative adjustment of the reference text, the reference text after each round of adjustment and the corresponding content difference change information are displayed in real time. After using the second largest model to complete the semantic consistency detection, the semantic consistency detection results between the translated text and the marketing text are displayed; If the detection result is semantically consistent, the translated text is determined as the final translated text and highlighted.

9. The marketing text translation method according to claim 8, characterized in that, Also includes: Display a control to enable the marketing text translation mode, which is associated with a target region selection sub-control and a translation parameter setting sub-control; In response to the activation operation of the control for the marketing text translation mode, the target region selection sub-control and the translation parameter setting sub-control are displayed. The target region selection sub-control is used to select the target region, and the translation parameter setting sub-control is used to set the preset value of content difference and the threshold of the number of iterations.

10. The marketing text translation method according to claim 8, characterized in that, The synchronized display of the translated text, the reference text, and the content differences between the marketing text and the reference text includes: The display interface is divided into a marketing text display area, a translated text display area, a reference text display area, and a difference display area. There is no overlap between the display areas and the layout is adapted to each other. In the reference text display area, the parts that differ from the marketing text are marked and displayed, and in the difference degree display area, the quantitative value of the content difference degree and the difference dimension distribution information are displayed simultaneously.

11. The marketing text translation method according to claim 8, characterized in that, The method further includes: When displaying the semantic consistency detection results, a detection details control and a manual text adjustment control are shown; The detection details control is used to display the semantic consistency detection details of the second model from three dimensions: semantics, core marketing information, and sentiment tendency. The manual text adjustment control is used to manually modify the translated text or reference text.

12. The marketing text translation method according to claim 11, characterized in that, If the detection result is semantically consistent, the translated text is determined as the final translated text and highlighted, including: If the detection result is semantic inconsistency, in response to the operation of manually adjusting the control for the text, a text editing area is displayed, and the translated text or reference text is manually modified based on the text editing area; The modified translated text is then re-translated, and the steps of back-translation, difference determination, iterative adjustment, and semantic consistency detection are repeated until the detection result is semantically consistent. The translated text at this point is then identified as the final translated text and highlighted, while a translation completion indicator is displayed.

13. A marketing text translation device, characterized in that, The device includes: The acquisition module is used to acquire the marketing text input by the user and the target region, and to translate the marketing text based on the language and cultural characteristics of the target region to obtain the translated text; The processing module is used to perform source language back-translation processing on the translated text to obtain a reference text in the same language as the marketing text; A determination module is used to determine the degree of content difference between the marketing text and the reference text; The adjustment module is used to adjust the reference text according to the content difference and the preset first large model if the content difference is greater than or equal to a preset value, and return to the step of determining the content difference between the marketing text and the reference text until the content difference between the marketing text and the reference text is less than the preset value. The detection module is used to detect the semantic consistency between the translated text and the marketing text using a preset second major model; The output module is used to output the translated text when the translated text is semantically consistent with the marketing text.

14. An electronic device, characterized in that, include: A memory, a processor, and a processor program stored in the memory and executable on the processor, wherein the processor executes the program as steps of the marketing text translation method as described in any one of claims 1 to 13.

15. A storage medium, characterized in that, The computer processing program is stored and can be loaded by a processor to execute the marketing text translation method as described in any one of claims 1 to 13.

Images