Real-time translation interaction method, system and equipment based on double-sided screen equipment

Abstract

The invention discloses a real-time translation interaction method, system and device based on a double-sided screen device, and the method comprises the steps: translating source language content displayed on a main screen into a target language in real time through a translation server, and then displaying the target language on the main screen, an auxiliary screen or a projection screen; monitoring a control instruction based on the main screen and the auxiliary screen in real time, and synchronously displaying the control content displayed on one screen for the control object in a corresponding area of the same control object on the other screen; according to a trigger request of the voice translation touch button, a voice signal is obtained and translated in real time to obtain target language content, and the target language content is converted into voice through a TTS voice engine and played; a signal source direction is determined by using a time delay estimation algorithm, and environmental noise compensation is carried out by using an adaptive noise compensation algorithm. According to the technical scheme, flexible and convenient multi-mode translation interaction display is realized, the translation communication efficiency is remarkably improved, and the translation interaction experience in a complex scene is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of translation machines, and particularly to a real-time translation interaction method based on a dual-screen device, a real-time translation interaction system based on a dual-screen device, and a dual-screen device for real-time translation interaction. Background Art

[0002] With the acceleration of the globalization process, cross-border business exchanges have become increasingly frequent. In scenarios such as business presentations, product exhibitions, and technical exchanges, language barriers often become obstacles to communication. Traditional translation methods, such as manual translation or simple translation software, cannot meet the requirements of real-time, efficient, and intuitive cross-language communication. For example, manual translation is inefficient, costly, and difficult to achieve real-time interaction; most existing translation software can only display in a single language and cannot simultaneously display content in different languages, which is not conducive to the immediate understanding and communication between both parties.

[0003] Specifically, a single-screen device cannot synchronously display content in multiple languages and needs to frequently switch interfaces, reducing communication efficiency. The multi-screen system is separated from the translation function, with complex operations and high hardware costs. Existing translation services are not deeply integrated with the dual-screen hardware, and cannot achieve differential content display and real-time interaction on the dual screen. Summary of the Invention

[0004] To address the above problems, the present invention provides a real-time translation interaction method, system, and device based on a dual-screen device. Through the multi-mode translation interaction display of the main screen and the secondary screen in the dual-screen device for the source language content and the target language content obtained by translation, combined with the dual-screen synchronous display of manipulation instructions, there is no need to frequently switch interfaces, enabling a flexible and convenient interaction experience. At the same time, it can also achieve accurate voice translation in convenient and complex scenarios through pointing noise reduction, significantly improving the translation communication efficiency and enhancing the translation interaction experience in complex scenarios.

[0005] To achieve the above object, the present invention provides a real-time translation interaction method based on a dual-screen device, including:

[0006] For the source language content displayed on the main screen based on the user's manipulation instruction, after real-time translation into the target language by the translation server, the corresponding target language content is displayed on the main screen, the secondary screen, or the projection screen according to the preset display mode;

[0007] Real-time monitor the manipulation instructions based on the main screen and the secondary screen, and synchronously display the manipulation content displayed for the manipulation object on one of the screens based on the manipulation instruction in the corresponding area of the same manipulation object on the other screen;

[0008] Upon receiving a trigger request for the voice translation touch button set on the main screen, secondary screen, or dual-screen device, obtain a voice signal through a dual microphone array, determine the source language content corresponding to the voice signal through voice recognition, obtain the target language content through real-time translation by the translation server, and convert the target language content into voice through a TTS voice engine and play it;

[0009] Analyze the voice signal obtained by the dual microphone array using a time delay estimation algorithm to determine the signal source direction corresponding to the voice signal, and perform environmental noise compensation on the voice signal through an adaptive noise compensation algorithm.

[0010] In the above technical solution, preferably, the real-time translation interaction method based on the dual-screen device further includes:

[0011] Perform emotion feature recognition on the voice signal through an emotion recognition algorithm, and dynamically adjust the playback tone and rhythm of the TTS voice engine for the target language content according to the emotion features.

[0012] In the above technical solution, preferably, the real-time translation interaction method based on the dual-screen device further includes:

[0013] Based on the preset trigger method of the voice translation touch button independently set on the dual-screen device, wake up the AI voice assistant, and set the display mode and translation language of the dual-screen device through voice interaction commands with the AI voice assistant.

[0014] In the above technical solution, preferably, the real-time translation interaction method based on the dual-screen device further includes:

[0015] Obtain the opening angle and horizontal direction between the main screen and the secondary screen through a sensor;

[0016] If the main screen and the secondary screen are in a folded state, turn off the main screen and the secondary screen;

[0017] If the main screen and the secondary screen are in an arched state, control the main screen and the secondary screen to be in a split-screen independent display mode or a mirror display mode;

[0018] If the main screen and the secondary screen are in a fully unfolded and horizontally tiled state, control the main screen and the secondary screen to be in an extended display mode or a mirror display mode;

[0019] If the main screen and the secondary screen are in a fully unfolded and vertical state, control the main screen and the secondary screen to be in an extended display mode or a mirror display mode according to the vertical direction.

[0020] In the above technical solution, preferably, the real-time translation interaction method based on the dual-screen device further includes:

[0021] Perform voiceprint recognition on the voice signal through a voiceprint recognition algorithm, and record the voice signal, the corresponding source language content, and the target language content respectively based on the voiceprint recognition label.

[0022] In the above technical solution, preferably, the process of displaying the corresponding target language content on the main screen, the secondary screen, or the screen projection according to the preset display mode specifically includes:

[0023] Display the source language content and / or the target language content in a preset area of the main screen, the secondary screen, or the screen projection according to the display modes of full-screen display, split-screen independent display, or screen projection.

[0024] In the above technical solution, preferably, real-time monitor the control instructions based on the main screen and the secondary screen, and synchronously display the control content displayed for the control object on one of the screens based on the control instructions in the corresponding area of the same control object on the other screen. The specific process includes:

[0025] Real-time monitor the interactive touch instructions and the brush annotation trajectories of the user based on the main screen and the secondary screen;

[0026] According to the interactive touch instruction for a certain control object on either the main screen or the secondary screen, synchronously execute the interactive touch instruction for the same control object on the other screen;

[0027] According to the brush annotation trajectory for the corresponding area of a certain control object on either the main screen or the secondary screen, synchronously display the brush annotation trajectory in the same area of the same control object on the other screen.

[0028] In the above technical solution, preferably, according to the trigger request of the voice translation touch button set on the main screen, the secondary screen, or the dual-screen device, the specific process includes:

[0029] Based on the virtual voice translation touch buttons set on the main screen and the secondary screen, real-time detect the trigger request instruction;

[0030] Based on the physical voice translation touch button independently set on the dual-screen device, real-time detect the trigger request instruction.

[0031] The present invention also proposes a real-time translation interaction system based on a dual-screen device, which applies the real-time translation interaction method based on a dual-screen device disclosed in any one of the above technical solutions, including:

[0032] A real-time translation module, which is used to perform real-time translation of the source language content displayed on the main screen based on the user's control instructions into the target language through a translation server, and then display the corresponding target language content on the main screen, the secondary screen or the projection screen according to the preset display mode;

[0033] A synchronous control module, which is used to monitor the control instructions based on the main screen and the secondary screen in real time, and synchronously display the control content displayed on one of the screens for the control object based on the control instructions in the corresponding area of the same control object on the other screen;

[0034] A voice translation module, which is used to obtain a voice signal through a dual microphone array according to the trigger request of the voice translation touch button set on the main screen, the secondary screen or the double-sided screen device, determine the source language content corresponding to the voice signal through voice recognition, obtain the target language content through the real-time translation of the translation server, and convert the target language content into voice and play it through a TTS voice engine;

[0035] A pointing noise reduction module, which is used to analyze the voice signal obtained by the dual microphone array by using a time delay estimation algorithm, determine the signal source direction corresponding to the voice signal, and perform environmental noise compensation on the voice signal through an adaptive noise compensation algorithm.

[0036] The present invention also proposes a double-sided screen device for real-time translation interaction, and the double-sided screen device applies the real-time translation interaction method based on the double-sided screen device disclosed in any one of the above technical solutions.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0038] (1) Through the multi-mode translation interaction display of the source language content and the translated target language content on the main screen and the secondary screen of the double-sided screen device, combined with the dual-screen synchronous display of the control instructions, there is no need to frequently switch the interface, and a flexible and convenient interaction experience can be realized;

[0039] (2) Through pointing noise reduction, accurate voice translation in convenient and complex scenarios is realized, significantly improving the translation communication efficiency and enhancing the translation interaction experience in complex scenarios;

[0040] (3) Through emotion recognition, the dynamic adjustment of the playback tone and rhythm of the translation content is realized, enhancing the use experience of translation communication;

[0041] (4) Through the integration of the AI voice assistant, the convenient setting of the system mode is realized, enhancing the setting efficiency and use experience;

[0042] (5) Through the device state of the double-sided screen device, the display mode of the device is automatically controlled, reducing the control steps in the corresponding usage scenarios and enhancing the interaction efficiency and use experience;

[0043] (6) Through content recording based on voiceprint recognition, the traceability and recognition of translation records are improved, effectively enhancing the usage experience of translation records. Description of the Drawings

[0044] Figure 1 It is a schematic flowchart of a real-time translation interaction method based on a dual-screen device disclosed in an embodiment of the present invention;

[0045] Figure 2 It is a schematic flowchart of a translation processing process for a user's picture file disclosed in an embodiment of the present invention;

[0046] Figure 3 It is a schematic flowchart of a real-time translation interaction method based on a dual-screen device disclosed in another embodiment of the present invention. Detailed Embodiments

[0047] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0048] The following provides a more detailed description of the present invention with reference to the accompanying drawings:

[0049] As Figure 1 shown, a real-time translation interaction method based on a dual-screen device provided by the present invention includes:

[0050] For the source language content displayed on the main screen based on the user's manipulation instruction, it is translated into the target language in real time through a translation server, and then the corresponding target language content is displayed on the main screen, the secondary screen, or projected according to the preset display mode;

[0051] Real-time monitoring of the manipulation instructions based on the main screen and the secondary screen, and synchronously displaying the manipulation content displayed on one screen for a manipulation object based on the manipulation instruction in the corresponding area of the same manipulation object on the other screen;

[0052] According to the trigger request of the voice translation touch button set on the main screen, the secondary screen, or the dual-screen device, a voice signal is obtained through a dual microphone array, the source language content corresponding to the voice signal is determined through voice recognition, the target language content is obtained through real-time translation by the translation server, and the target language content is converted into voice and played through a TTS voice engine;

[0053] The time delay estimation algorithm is used to analyze the speech signal obtained by the dual-microphone array to determine the signal source direction corresponding to the speech signal, and the environmental noise is compensated for the speech signal through an adaptive noise compensation algorithm.

[0054] In this embodiment, multi-mode translation and interactive display of source language content and translated target language content are performed through the main screen and the sub-screen in the double-sided screen device, combined with the dual-screen synchronous display of control instructions, without the need for frequent interface switching, which can achieve a flexible and convenient interactive experience. At the same time, it can also achieve convenient and accurate voice translation in complex scenarios through pointing noise reduction, which significantly improves the efficiency of translation communication and enhances the translation interaction experience in complex scenarios.

[0055] Among them, if the source language content displayed on the main screen is text content, it is directly transmitted to the translation server. If the main screen displays picture content, the text content in the picture is extracted through OCR recognition and then transmitted to the translation server. After obtaining the translated target language content, the target language content is overwritten to the source language content position of the original picture.

[0056] like Figure 2 As shown, during the implementation process, a text file or image file can be uploaded to the dialogue window or dragged to the dialogue window, or text content can be directly entered in the dialogue window, and the dialogue window is used as an interface to process the above content and transmit it to the translation server.

[0057] Specifically, during the implementation process, the source language content is transmitted to the translation server for real-time translation. After being translated into the target language, the target language content is fed back to the double-sided screen device, and according to the display mode pre-selected by the user, the source language content and the target language content are displayed on the main screen at the same time, or the source language content is displayed on the main screen and the target language content is displayed on the secondary screen, or the source language content is displayed on the main screen and the target language content is displayed on the secondary screen and the projection screen, or the source language content is displayed on the main screen and the target language content is displayed on the secondary screen and the projection screen, or the source language content and the target language content are displayed on both the main screen and the secondary screen.

[0058] In addition, through the dual-screen synchronous display of control commands, when the two parties are face to face for translation and communication, the double-sided screens are arched, with the main screen and the sub-screen facing the two sides respectively. At this time, both parties can watch and control the same content in different languages at the same time, greatly improving the efficiency of translation and communication interaction.

[0059] Furthermore, in order to improve communication efficiency, real-time speech translation and reading are achieved through the conversion logic of speech to text, real-time text translation, and translation content to speech. At the same time, the time delay of the two voice signals collected by the dual-microphone array is combined to determine the source direction of the voice signal, and the environmental noise in other directions is blocked through an adaptive noise compensation algorithm to achieve environmental noise compensation for the voice signal and improve the recognition accuracy of the voice signal.

[0060] In the above embodiments, preferably, according to a preset display mode, the corresponding target language content is displayed on the main screen, the secondary screen or the projection screen. The specific process includes:

[0061] According to the display modes of full-screen display, split-screen independent display or projection screen display, the source language content and / or the target language content are displayed in a preset area of the main screen, the secondary screen or the projection screen.

[0062] During the implementation process, areas such as text boxes are pre-defined on the main screen and the secondary screen, which can be used to display the source language content and the target language content respectively. Among them, operations such as input, selection, modification, deletion and replacement of the source language content and the target language content can be realized in the text box area.

[0063] In the above embodiments, preferably, the manipulation instructions based on the main screen and the secondary screen are monitored in real time, and the manipulation content displayed for the manipulation object on one of the screens based on the manipulation instructions is synchronously displayed in the corresponding area of the same manipulation object on the other screen. The specific process includes:

[0064] Monitor the interactive touch instructions and the pen annotation tracks of the user based on the main screen and the secondary screen in real time;

[0065] According to the interactive touch instruction for a certain manipulation object on either the main screen or the secondary screen, execute the interactive touch instruction synchronously for the same manipulation object on the other screen;

[0066] According to the pen annotation track for the corresponding area of a certain manipulation object on either the main screen or the secondary screen, synchronously display the pen annotation track in the same area of the same manipulation object on the other screen.

[0067] During the implementation process, for example, it is monitored that the user makes a pen annotation on a certain word on the main screen, then according to the monitored annotation track and annotation position, the same-shaped track annotation is also made at the corresponding position of the same word on the secondary screen.

[0068] In the above embodiments, preferably, according to the trigger request of the voice translation touch button set on the main screen, the secondary screen or the double-sided screen device, the specific process includes:

[0069] Based on the virtual voice translation touch buttons set on the main screen and the secondary screen, detect the trigger request instruction in real time;

[0070] Based on the physical voice translation touch button independently set on the double-sided screen device, detect the trigger request instruction in real time.

[0071] During the implementation process, once a trigger request instruction is detected, the dual microphone array of the dual-screen device starts to acquire voice signals, and at the same time, the time delay estimation algorithm is used to identify and analyze the direction of the signal source, and the adaptive noise compensation algorithm is used to compensate the ambient noise of the voice signal in that direction.

[0072] As Figure 3 shown, in the above embodiment, preferably, the real-time translation interaction method based on the dual-screen device further includes:

[0073] Performing emotion feature recognition on the voice signal through an emotion recognition algorithm, and dynamically adjusting the playback tone and rhythm of the TTS voice engine for the target language content according to the emotion features.

[0074] During the implementation process, the emotion recognition algorithm itself belongs to the prior art. According to the emotion features obtained by recognizing the voice signal through the emotion recognition algorithm, the emotion features can be expressed using emotion feature tags or emotion feature magnitudes, and the playback tone and playback rhythm of the TTS voice engine can be adjusted according to a preset dynamic adjustment scale or dynamic adjustment algorithm. For example, if the emotion feature of the voice signal is recognized as cheerful, the playback tone of the target language content is appropriately increased and the playback rhythm is appropriately accelerated.

[0075] In the above embodiment, preferably, the real-time translation interaction method based on the dual-screen device further includes:

[0076] Based on the preset triggering method of the voice translation touch button independently set on the dual-screen device, waking up the AI voice assistant, and setting the display mode and translation language of the dual-screen device through voice interaction instructions with the AI voice assistant.

[0077] During the implementation process, it can be set that when the voice translation touch button is short-pressed (within 1 second), the voice translation function is triggered, and when the voice translation touch button is long-pressed (more than 2 seconds), the AI voice assistant function is triggered. After waking up the AI voice assistant, through voice interaction with the AI voice assistant, the AI voice assistant has the permission to control and set the display mode and translation language of the dual-screen device.

[0078] In the above embodiment, preferably, the real-time translation interaction method based on the dual-screen device further includes:

[0079] Obtaining the opening angle and horizontal direction between the main screen and the secondary screen through a sensor;

[0080] If the main screen and the secondary screen are in a folded state, turn off the main screen and the secondary screen;

[0081] If the main screen and the secondary screen are in an arched state, control the main screen and the secondary screen to be in a split-screen independent display mode or a mirror display mode;

[0082] If the main screen and the secondary screen are in a fully expanded and horizontally tiled state, then control the main screen and the secondary screen to be in an extended display mode or a mirror display mode;

[0083] If the main screen and the secondary screen are in a fully expanded and vertical state, then control the main screen and the secondary screen to be in an extended display mode or a mirror display mode according to the vertical direction.

[0084] During the implementation process, automatically adjust the display mode of the dual-screen device according to the relative opening angle between the main screen and the secondary screen and the placement state of the dual-screen device, without manual adjustment. The automatic adjustment strategy can be set and changed according to the usage scenario and usage habits.

[0085] In the above implementation manner, preferably, the real-time translation interaction method based on the dual-screen device further includes:

[0086] Perform voiceprint recognition on the voice signal through a voiceprint recognition algorithm, and respectively record the voice signal, the corresponding source language content, and the target language content based on the voiceprint recognition label.

[0087] During the implementation process, according to the voiceprint recognition result, tags can be added to the voice signals of multiple users during a single use, such as "User 1", "User 2", or "User A", "User B". During the implementation process, it can also be customized. During the process of organizing the translation content records, it is also possible to synchronously modify different tags, such as modifying all "User 1" in a single record to the user name corresponding to the voice signal.

[0088] The present invention also proposes a real-time translation interaction system based on a dual-screen device, which applies the real-time translation interaction method based on the dual-screen device disclosed in any one of the above implementation manners, including:

[0089] A real-time translation module, which is used to, for the source language content displayed on the main screen based on a user control instruction, perform real-time translation into the target language through a translation server, and then display the corresponding target language content on the main screen, the secondary screen, or the projection screen according to a preset display mode;

[0090] A synchronous control module, which is used to monitor the control instructions based on the main screen and the secondary screen in real time, and synchronously display the control content displayed for the control object on one screen based on the control instruction in the corresponding area of the same control object on the other screen;

[0091] A voice translation module, which is used to, according to the trigger request of the voice translation touch button set on the main screen, the secondary screen, or the dual-screen device, obtain the voice signal through a dual microphone array, determine the source language content corresponding to the voice signal through voice recognition, obtain the target language content through the real-time translation of the translation server, and convert the target language content into voice through a TTS voice engine and play it;

[0092] A noise reduction module, which is used to analyze the speech signal obtained by the dual microphone array by using the time delay estimation algorithm, determine the signal source direction corresponding to the speech signal, and perform environmental noise compensation on the speech signal through the adaptive noise compensation algorithm.

[0093] According to the real-time translation interaction system based on the dual-screen device disclosed in the above embodiment, the functions to be realized by each module respectively correspond to the steps of the real-time translation interaction method based on the dual-screen device disclosed in the above embodiment. During the implementation process, refer to the above embodiment for operation, and details are not described herein again.

[0094] The present invention also proposes a dual-screen device for real-time translation interaction, and the dual-screen device applies the real-time translation interaction method based on the dual-screen device disclosed in any one of the above embodiments.

[0095] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A real-time translation interaction method based on a dual-screen device, characterized in that, Including: For the source language content displayed on the main screen based on the user's control instructions, after real-time translation into the target language by the translation server, the corresponding target language content is displayed on the main screen, the secondary screen or the projection screen according to the preset display mode; Real-time monitor the control instructions based on the main screen and the secondary screen, and synchronously display the control content displayed on one of the screens for the control object based on the control instructions in the corresponding area of the same control object on the other screen; According to the trigger request of the voice translation touch button set on the main screen, the secondary screen or the double-sided screen device, obtain the voice signal through the dual microphone array, determine the source language content corresponding to the voice signal through voice recognition, obtain the target language content through the real-time translation of the translation server, and convert the target language content into voice and play it through the TTS voice engine; Use the time delay estimation algorithm to analyze the voice signal obtained by the dual microphone array, determine the signal source direction corresponding to the voice signal, and perform environmental noise compensation on the voice signal through the adaptive noise compensation algorithm.

2. The real-time translation interaction method based on a dual-screen device according to claim 1, wherein It also includes: Perform emotion feature recognition on the voice signal through the emotion recognition algorithm, and dynamically adjust the playback tone and rhythm of the TTS voice engine for the target language content according to the emotion features.

3. The real-time translation interaction method based on a dual-screen device according to claim 1, characterized in that It also includes: Based on the preset trigger method of the voice translation touch button independently set on the double-sided screen device, wake up the AI voice assistant, and set the display mode and translation language of the double-sided screen device through the voice interaction instruction with the AI voice assistant.

4. The real-time translation interaction method based on a dual-screen device according to claim 1, characterized in that It also includes: Obtain the opening angle and horizontal direction between the main screen and the secondary screen through the sensor; If the main screen and the secondary screen are in the folded state, turn off the main screen and the secondary screen; If the main screen and the secondary screen are in the arch-standing state, control the main screen and the secondary screen to be in the split-screen independent display mode or the mirror display mode; If the main screen and the secondary screen are in the fully unfolded and horizontally tiled state, control the main screen and the secondary screen to be in the extended display mode or the mirror display mode; If the main screen and the secondary screen are in the fully unfolded and vertical state, control the main screen and the secondary screen to be in the extended display mode or the mirror display mode according to the vertical direction.

5. The real-time translation interaction method based on a dual-screen device according to claim 1, wherein It also includes: Perform voiceprint recognition on the voice signal through the voiceprint recognition algorithm, and respectively record the voice signal and the corresponding source language content and target language content based on the voiceprint recognition label.

6. The real-time translation interaction method based on a dual-screen device according to claim 1, characterized in that The specific process of displaying the corresponding target language content on the main screen, the secondary screen or the projection screen according to the preset display mode includes: According to the display modes of same-screen display, split-screen independent display or projection display, display the source language content and / or target language content in the preset area of the main screen, the secondary screen or the projection screen.

7. The real-time translation interaction method based on a dual-screen device according to claim 1, characterized in that Real-time monitor the control instructions based on the main screen and the secondary screen, and synchronously display the control content displayed on one of the screens for the control object based on the control instructions in the corresponding area of the same control object on the other screen. The specific process includes: Real-time monitor the user's interactive touch instructions and brush annotation trajectories based on the main screen and the secondary screen; Execute the interactive touch command for the same control object on the other screen synchronously according to the interactive touch command for a certain control object on any one of the main screen and the secondary screen. Synchronously display the brush annotation trajectory of the corresponding area of a certain control object on any one of the main screen and the secondary screen on the same area of the same control object on the other screen.

8. The real-time translation interaction method based on a dual-screen device according to claim 1, wherein, According to the trigger request of the voice translation touch button set on the main screen, the secondary screen or the dual-screen device, the specific process includes: Based on the virtual voice translation touch buttons set on the main screen and the secondary screen, detect the trigger request command in real time. Based on the physical voice translation touch button independently set on the dual-screen device, detect the trigger request command in real time.

9. A real-time translation interaction system based on a dual-screen device, characterized in that, Applying the real-time translation interaction method based on the dual-screen device according to any one of claims 1 to 8, includes: A real-time translation module, which is used to translate the source language content displayed on the main screen based on the user's control instruction into the target language in real time through a translation server, and then display the corresponding target language content on the main screen, the secondary screen or the projection screen according to the preset display mode. A synchronous control module, which is used to monitor the control instructions based on the main screen and the secondary screen in real time, and synchronously display the control content displayed for the control object on one of the screens on the corresponding area of the same control object on the other screen. A voice translation module, which is used to obtain a voice signal through a dual microphone array according to the trigger request of the voice translation touch button set on the main screen, the secondary screen or the dual-screen device, determine the source language content corresponding to the voice signal through voice recognition, obtain the target language content through the real-time translation of the translation server, and convert the target language content into voice through a TTS voice engine and play it. A pointing noise reduction module, which is used to analyze the voice signal obtained by the dual microphone array by using the time delay estimation algorithm, determine the signal source direction corresponding to the voice signal, and perform environmental noise compensation on the voice signal through the adaptive noise compensation algorithm.

10. A double-sided screen device for real-time translation interaction, characterized in that, The dual-screen device applies the real-time translation interaction method based on the dual-screen device according to any one of claims 1 to 8.

Images