Translation method, electronic device, storage medium, and translation machine

Abstract

The application discloses a translation method, an electronic device, a storage medium and a translation machine, wherein the translation method comprises the following steps: obtaining input information of a plurality of target devices; performing parallel processing on the input information of the plurality of target devices to obtain translation information corresponding to target language categories of the plurality of input information; and outputting the translation information corresponding to the target language categories of the plurality of input information in parallel. The method can realize parallel processing and parallel translation output of multilingual information, improve the multilingual translation function of the translation machine, improve the processing efficiency of multidevice and multilingual translation, and thus improve the experience of multidevice synchronous interaction and cross-language communication.

Description

Technical Field

[0001] This invention relates to the field of communication translation technology, and in particular to a translation method, electronic device, storage medium, and translator. Background Technology

[0002] In the context of increasing globalization, the demand for cross-language communication is growing rapidly, especially in international business conferences, transnational cooperation, tourism, and multilingual social occasions, where users are placing higher demands on real-time and accurate multilingual translation services. However, existing translation devices use a serial processing mode, which can only process the input information of one user or device at a time, making it difficult to meet the needs of multiple users inputting simultaneously and translating multiple languages ​​in parallel. This processing method is prone to translation delays in multi-party interaction or conference scenarios, affecting the timeliness of information transmission and the experience of cross-language communication. Summary of the Invention

[0003] This invention aims to at least solve one of the technical problems existing in the prior art. Therefore, one objective of this invention is to propose a translation method that enables parallel processing and parallel translation output of multilingual information, improves the multilingual translation function of translation machines, increases the processing efficiency of multi-device, multilingual translation, and thus enhances the experience of synchronous interaction and cross-language communication across multiple devices.

[0004] The second objective of this invention is to provide an electronic device.

[0005] A third objective of this invention is to provide a computer-readable storage medium.

[0006] The fourth objective of this invention is to provide a translation machine.

[0007] To achieve the above objectives, a translation method according to a first aspect of the present invention includes: obtaining input information from multiple target devices; performing parallel processing on the input information from the multiple target devices to obtain translation information for target language types corresponding to the multiple input information; and outputting the translation information for the multiple input information for the target language types in parallel.

[0008] According to the translation method of this invention, input information from multiple target devices is acquired simultaneously, enabling the collection of input data in multiple languages. Then, parallel processing technology, deep learning models, and translation models are used to perform real-time language recognition and translation on the input information from each target device, ensuring that each input information can be simultaneously converted into a translation result in its corresponding target language. These translation results are synchronously transmitted back to their respective target devices through a parallel output mechanism, avoiding the time delay caused by sequential translation in traditional serial processing methods. This parallel processing method ensures that multilingual information can be processed efficiently and in real-time, ultimately achieving synchronous translation output between multiple devices. This enhances the multilingual translation function of the translator, improves the processing efficiency of multi-device, multilingual translation, and thus enhances the experience of synchronous interaction and cross-language communication across multiple devices.

[0009] In some embodiments, the parallel output of translation information corresponding to the target language of multiple input information includes: outputting the translation information separately through split-screen or multi-channel audio.

[0010] In some embodiments, the translation method further includes: obtaining the input language type corresponding to each input information; automatically translating the feedback information of the target language type into information of the target's input language type, and sending it to the target device corresponding to the target's input speech type.

[0011] In some embodiments, obtaining the input language type corresponding to each input information includes: extracting text information from the input information using a deep learning model and determining the input language type corresponding to the input information based on the text information.

[0012] In some embodiments, the translation method further includes sending the text, image, and voice data to a target device corresponding to the input voice type of the target.

[0013] In some embodiments, the input information of the plurality of target devices is multimodal information.

[0014] In some embodiments, the translation method further includes: sending the translation information to a target receiving device, the target receiving device including the target device and / or other communication terminals.

[0015] In some embodiments, the target device includes at least one of a mobile terminal and a wearable device.

[0016] To achieve the above objectives, an electronic device according to a second aspect of the present invention includes: at least one processor; a memory communicatively connected to the at least one processor; the memory storing a computer program executable by the at least one processor, wherein the at least one processor executes the computer program to implement the translation method described in the above embodiment.

[0017] According to an embodiment of the present invention, an electronic device has at least one processor executing a computer program that implements the translation method described in the above embodiment. By acquiring input information from multiple target devices, it achieves simultaneous acquisition of input data in multiple languages. Then, using parallel processing technology, a deep learning model, and a translation model, it performs real-time language recognition and translation on the input information from each target device, ensuring that each input information can be simultaneously converted into a translation result for the corresponding target language. These translation results are synchronously transmitted back to their respective target devices through a parallel output mechanism, avoiding the time delay caused by sequential translation in traditional serial processing methods. This parallel processing method ensures that multilingual information can be processed efficiently and in real-time, ultimately achieving synchronous translation output between multiple devices. This enhances the multilingual translation function of the translator, improves the processing efficiency of multi-device, multilingual translation, and thus enhances the experience of synchronous interaction and cross-language communication across multiple devices.

[0018] To achieve the above objectives, a computer-readable storage medium according to a third aspect of the present invention stores a computer program thereon, which, when executed by a processor, implements the translation method described in the above embodiments.

[0019] According to the computer-readable storage medium of the present invention, by employing the translation method described in the above embodiments, parallel processing and parallel translation output of multilingual information can be realized, thereby improving the multilingual translation function of the translator, increasing the processing efficiency of multi-device and multilingual translation, and thus improving the experience of multi-device synchronous interaction and cross-language communication.

[0020] To achieve the above objectives, a translator according to a fourth aspect of the present invention includes: a receiving module for receiving input information; and a processing module connected to the receiving module for executing the translation method described in the above embodiment.

[0021] According to embodiments of the present invention, the processing module of the translator executes the translation method described in the above embodiments, acquiring input information from multiple target devices to simultaneously collect input data in multiple languages. Then, using parallel processing technology, deep learning models, and translation models, real-time language recognition and translation are performed on the input information from each target device, ensuring that each input information can be simultaneously converted into translation results for the corresponding target language. These translation results are synchronously transmitted back to their respective target devices through a parallel output mechanism, avoiding the time delay caused by sequential translation in traditional serial processing methods. This parallel processing method ensures that multilingual information can be processed efficiently and in real-time, ultimately achieving synchronous translation output between multiple devices, improving the multilingual translation function of the translator, increasing the processing efficiency of multi-device, multilingual translation, and thus enhancing the experience of synchronous interaction and cross-language communication across multiple devices.

[0022] In some embodiments, the processing module includes: an artificial intelligence unit, configured to perform speech recognition on the input information to extract text information from the input information and determine the input language type corresponding to the input information based on the text information; and a local translation unit, configured to obtain translation information of the target language type corresponding to the input information, and to automatically translate the feedback information of the target language type into information of the target input language type.

[0023] In some embodiments, the translator further includes an output module connected to the processing module for outputting translation information.

[0024] In some embodiments, the translator further includes a communication module connected to the processing module for sending translation information.

[0025] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0026] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart of a translation method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a translation processing flow according to an embodiment of the present invention; Figure 3 This is a block diagram of an electronic device according to an embodiment of the present invention; Figure 4 This is a block diagram of a translator according to an embodiment of the present invention.

[0027] Figure label: Translation machine 100; Receive module 1; Processing module 2; Output module 3; Communication module 4; Artificial intelligence unit 21; Local translation unit 22; 200 electronic devices; Processor 201; Memory 202. Detailed Implementation

[0028] The embodiments of the present invention are described in detail below. The embodiments described with reference to the accompanying drawings are exemplary. The embodiments of the present invention are described in detail below.

[0029] The following is for reference. Figures 1-2 A translation method according to embodiments of the present invention is described, which is used in a translation machine. The translation machine can be a hardware device or a software-based translation program or application. A hardware-based translation machine can connect to multiple target devices via wired or wireless means (such as Wi-Fi, Bluetooth, or 5G networks); while a software-based translation machine can run directly on multiple target devices, enabling parallel processing and translation output of multilingual information, thereby improving the experience of synchronous interaction across multiple devices and cross-language communication.

[0030] Figure 1 This is a flowchart of a translation method according to an embodiment of the present invention, such as... Figure 1 As shown, the translation method of this embodiment of the invention includes at least steps S1-S3.

[0031] S1, obtain input information from multiple target devices.

[0032] In some embodiments, the form of input information may include, but is not limited to: voice input, text input, image input, and video input. Voice input may involve acquiring voice information using a microphone or audio input device on the device. Text input may involve acquiring text information through text input boxes, SMS messages, emails, or uploading text files. Image input may involve acquiring text information by taking a picture with text on it using the device's camera, or by selecting and uploading a picture with text from the gallery. Video input may involve recording or uploading a video with subtitles to acquire text or voice information contained in the video.

[0033] In some embodiments, such as Figure 2 As shown, the target device may include, but is not limited to: headphones with microphones, smartphones, tablets, laptops, smartwatches and other wearable devices.

[0034] S2 processes the input information from multiple target devices in parallel to obtain translation information for the target language corresponding to the multiple input information.

[0035] Specifically, parallel processing techniques such as multithreading, batch processing, or distributed computing are employed to synchronously process input information from different target devices, ensuring that input information from different target devices can be processed at the same time, thereby improving translation efficiency. Then, a pre-trained translation model is used to translate each input information in real time, thereby obtaining translation information for multiple input information corresponding to different target languages.

[0036] S3 outputs translation information for multiple input information corresponding to different target languages ​​in parallel.

[0037] Specifically, by employing a multi-channel transmission mechanism, translation results can be simultaneously pushed to multiple target devices, enabling real-time reception by different end users. Depending on the type of target device, the translation results can be converted into the corresponding output format. For example, the translator can convert the translation result of voice input into voice data in the target language and play it through a speaker or target device, suitable for conference scenarios to achieve simultaneous output of multilingual voice; the translator can push the translated text result to a text window to achieve simultaneous display of information from multiple users in multiple languages, suitable for scenarios such as cross-border team collaboration and cross-language real-time chat; the translator can overlay the translation result as subtitles onto a video stream to achieve real-time presentation of multilingual subtitles, suitable for scenarios such as cross-language conferences, international forums, and live webcasts.

[0038] According to the translation method of this invention, input information from multiple target devices is acquired simultaneously, enabling the collection of input data in multiple languages. Then, parallel processing technology, deep learning models, and translation models are used to perform real-time language recognition and translation on the input information from each target device, ensuring that each input information can be simultaneously converted into a translation result in its corresponding target language. These translation results are synchronously transmitted back to their respective target devices through a parallel output mechanism, avoiding the time delay caused by sequential translation in traditional serial processing methods. This parallel processing method ensures that multilingual information can be processed efficiently and in real-time, ultimately achieving synchronous translation output between multiple devices. This enhances the multilingual translation function of the translator, improves the processing efficiency of multi-device, multilingual translation, and thus enhances the experience of synchronous interaction and cross-language communication across multiple devices.

[0039] In some embodiments, parallel output of translation information for multiple input information corresponding to target language types includes: outputting translation information separately through split-screen or multi-channel output.

[0040] Split-screen output refers to presenting translation information for different target devices or different languages ​​in different screen areas on the same display device. Split-screen can be vertical (such as displaying translation results for different target devices or languages ​​in two areas, one above and one below or one to the left and right), grid split-screen (such as dividing the screen into multiple areas, each displaying different translation results), or dynamic split-screen (such as automatically adjusting the screen division ratio according to the number of target devices and the translation languages ​​to optimize information display).

[0041] Specifically, when using split-screen output, the translator divides the translated text or subtitles into multiple regions, each corresponding to a target language. For example, in a meeting scenario where different participants use different languages, the translator can divide their respective translation results into several windows on the same display screen, with each window displaying the text information in the corresponding language. This allows the translation results of all languages ​​to be presented simultaneously and intuitively on one screen, enabling synchronized reading and interaction in multiple languages.

[0042] In some embodiments, multi-channel output can refer to outputting translation information for different target devices or target languages ​​through different audio channels. When using multi-channel output, the translator outputs the translated speech data separately through different audio channels. Each audio channel can correspond to a specific target language, allowing speech translations in different languages ​​to be played independently through different speakers or headphones during conferences or multilingual video calls, avoiding interference caused by mixed sound. For example, the translator can transmit the speech translation results separately to connected audio devices and use an audio processing module to separate the language signals before outputting them through multi-channel output, ensuring that each participant receives clear and independent speech translation.

[0043] In some embodiments, for international conferences, participants from different countries can wear headphones with different audio channels to receive translation information in their respective languages, enabling seamless communication. For multilingual film / video playback, viewers can select different audio tracks to obtain translation information that matches their language. For VR (Virtual Reality) / AR (Augmented Reality) immersive experiences, in a multilingual environment, users can choose to listen to audio streams in different languages, achieving a more natural immersive interaction.

[0044] Therefore, this split-screen and multi-channel output method not only ensures the synchronization and independence of translated information, but also greatly improves the real-time nature and user experience of cross-language communication, enabling users of each target device to obtain clear and uninterrupted translation content according to their own language needs.

[0045] In some embodiments, the translator of the present invention is suitable for multilingual conference translation. Specifically, the translator can simultaneously translate the speech of multiple speakers. Unlike existing serial translation technologies, the present invention can process the input and output of multiple languages ​​simultaneously without translating them sequentially. The translation results are output via split-screen or multi-channel audio. Compared to existing technologies that can only process the speech of speakers one by one, this translator suitable for multilingual conference translation can significantly improve the synchronization efficiency of multilingual conferences.

[0046] In some embodiments, the translation device of the present invention is also suitable for meetings and seminars. Specifically, when multiple people participate in a discussion, they may express their opinions or interrupt at the same time, requiring the translation device to be able to distinguish and translate each person's speech in real time, which is especially important in international conferences and cross-language team collaborations.

[0047] In some embodiments, the translator of the present invention is also applicable to customer service centers. Specifically, when multiple customer service personnel are conversing with different customers simultaneously, an automated translator can help customer service personnel handle multilingual inquiries in real time, improving efficiency.

[0048] In some embodiments, the translator of the present invention is also applicable to guided tour services. Specifically, in multilingual guided tour venues such as museums and scenic spots, tour guides or staff may need to answer questions from tourists speaking different languages ​​simultaneously. The translator needs to be able to translate the speech input of multiple people in parallel in order to respond quickly to their needs.

[0049] In some embodiments, the translator of the present invention is also suitable for group discussions and distance learning. Specifically, in school group discussions or distance education, students may ask questions or communicate simultaneously. In this case, multilingual parallel translation can better support classroom interaction and ensure that everyone's language can be understood.

[0050] In some embodiments, the translation method further includes: obtaining the input language type corresponding to each input information, and automatically translating the feedback information of the target language type into information of the target's input language type, so as to send it to the target device corresponding to the target's input speech type.

[0051] In some embodiments, in multilingual communication scenarios, simply performing one-way translation is insufficient; it is also necessary to ensure that feedback information is correctly transmitted to the user of the original language to form a complete communication loop. Feedback information in the target language can refer to the process where, in two-way communication, after receiving input information from one end, the translation machine generates or processes the corresponding translation result based on that input, and then returns this result to the information sender in feedback form. Feedback information can include two scenarios: First, in this scenario, the translation machine first identifies and translates the input information from the target device, and then feeds back the translation result to the source device of the input information. For example, in an international conference, participant A speaks in English, and the translation machine automatically translates it into another target language. Simultaneously, when participant B replies in their native language (e.g., Chinese), the translation machine can automatically translate B's Chinese reply into English and feed it back to participant A. The feedback information here is the automatically translated response generated by the translation machine based on the input information.

[0052] Secondly, there's the feedback of information directly input locally. In some cases, the target device may directly input information in the local language (for example, a user directly inputs commands or comments on a translation device). The translation device then uses this information as feedback, transmitting it directly to the corresponding device in the target language. For instance, in some intelligent customer service scenarios, a user can directly input a question locally, and the customer service system translates the question into the customer service's working language, then the customer service representative responds, and the response is automatically translated back into the user's input language before being fed back to the user.

[0053] In some embodiments, obtaining the input language type corresponding to each input information includes: extracting text information from the input information using a deep learning model and determining the input language type corresponding to the input information based on the text information.

[0054] Specifically, the translator receives input information from the target device, which can exist in various forms, such as voice, images, or video. For voice input, the translator first converts the speech into text using Automatic Speech Recognition (ASR); while for image or video input, Optical Character Recognition (OCR) technology can be used to extract the embedded text. Regardless of the input format, the key is to convert the input information into standard text data.

[0055] Furthermore, the translation machine utilizes a deep learning model to process the extracted text information. This deep learning model can employ pre-trained language models (such as Transformer, BERT, etc.) for feature extraction and semantic understanding. By analyzing the grammatical structure, vocabulary distribution, and contextual information of the text, it determines which language the text belongs to. This step can automatically identify multiple languages ​​without requiring manual user input, ensuring high accuracy and real-time performance in determining the input language.

[0056] Furthermore, the translation machine can use the identified input language type as a label and associate it with subsequent translation tasks, enabling the translation model to automatically select an appropriate translation strategy or target language conversion path based on this language information. Users do not need to manually set the input language type.

[0057] Furthermore, after language recognition is completed, the translation machine can pass the input information to the translation model and automatically select the translation language according to the user-preset receiving language to complete the translation task. This not only automates and intelligently processes the input information but also provides an accurate linguistic foundation for subsequent two-way translation and feedback mechanisms.

[0058] In some embodiments, the translation method further includes sending the text, image, and voice data to a target device corresponding to the type of input voice input.

[0059] Specifically, after the translation process is complete, the translator can output the translation results in a format suitable for the target device's reception and display, based on the characteristics of the target device and the user's preset preferences. For example, if the target device is mainly used for text display, the translation results can be sent in text form; if the target device has image display capabilities, the translator can present the translated text information as an image, such as overlaying the translated content onto relevant pictures or video frames; and for devices that support voice output, the translation results can be processed through text-to-speech (TTS) and then fed back in voice form.

[0060] In some embodiments, the translator can also dynamically select the output format according to the specific scenario. For example, in a conference scenario, to ensure real-time communication between the speaker and the audience, both text and voice formats may be used simultaneously so that different devices and users can obtain the best interactive experience. Through this multimodal output method, the present invention not only ensures the efficient transmission of translation results, but also greatly improves the flexibility and adaptability of cross-device and cross-language communication.

[0061] In some embodiments, the input information from multiple target devices is multimodal information. That is, the input information can exist simultaneously or alternately in various forms, including but not limited to voice, text, images, and video. Multimodal input means that the target devices are no longer limited to a single input form, but can capture information through multiple sensing methods. For example, in a meeting or multilingual communication scenario, participants can speak by voice while simultaneously uploading text notes, taking pictures with text, or recording video clips using smartphones or tablets. This diverse information input method can more comprehensively express the content of the communication, compensating for the shortcomings of a single input method. After unified processing, this multimodal input information can provide a richer data foundation for subsequent automatic speech recognition, translation, and feedback output, thereby improving the accuracy of translation results and the flexibility of applications.

[0062] In some embodiments, the translator of the present invention not only supports voice and text input, but also has powerful image recognition translation capabilities. Users can take pictures of menus, road signs, or documents, and the system will recognize the text in the image and translate it. Therefore, the translator of the present invention greatly expands its application scenarios, such as translating menus or signs while traveling, or translating documents in the workplace.

[0063] In some embodiments, the translation method further includes: sending translation information to a target receiving device, the target receiving device including a target device and / or other communication terminals. These other communication terminals may be devices with communication capabilities, capable of receiving data and performing secondary processing or distribution. Examples include cloud servers, cloud platforms, IoT devices, and remote control terminals.

[0064] Specifically, after translation is completed, the translation device sends the translated information in an appropriate format (text, voice, image, video subtitles, etc.) to the target receiving device via a preset communication protocol (such as Wi-Fi, Bluetooth, 5G network, Internet, etc.). The target receiving device can be a terminal device directly connected to the translation device (such as a smartphone, tablet, smartwatch, conference terminal, etc.), or other terminals within the same communication network (such as a personal computer, server, cloud server, or dedicated communication terminal) to achieve multi-point distribution and synchronous reception of information. For example, in a multinational conference scenario, the translation device sends the translation results to various terminals in the conference system through split-screen display or multi-channel output, so that participants can receive the translated information in a timely manner regardless of the device they use; in an intelligent customer service system, customer service personnel and customers communicate using different devices, and the translated information is displayed synchronously across terminals through cloud push or direct transmission, thereby ensuring barrier-free communication between the two parties. In this way, the translation method of the present invention fully utilizes network communication and multi-terminal collaboration technology, realizes flexible and widespread dissemination of translated information, and further improves the applicability and interactive experience of the system in multi-device and multi-user scenarios.

[0065] In some embodiments, the target device includes at least one of a mobile terminal and a wearable device. The mobile terminal may refer to portable devices such as smartphones, tablets, and laptops, which have robust network connectivity, data processing, and display capabilities, enabling them to receive, process, and display translated information in real time. Wearable devices include small, wearable, and intelligent portable electronic devices such as smartwatches, smart glasses, and fitness trackers.

[0066] In some embodiments, the translation method of this invention is tailored to the characteristics of mobile terminals and wearable devices, supporting multimodal input and output. For example, on smartphones or tablets, users can capture speech through a built-in microphone, take images or videos through a camera, and then upload the text, image, and video information to the translation device. The translation results can be presented on a large screen in the form of text, speech, or images, making it easy for users to read and listen. In wearable devices such as smartwatches or smart glasses, translation information can be output through a small screen display or voice broadcast, enabling users to quickly obtain real-time translation content during exercise, travel, or meetings.

[0067] In some embodiments, the translator of the present invention has the ability to collaborate with multiple smart devices, including smartphones, tablets, smartwatches, etc., and can seamlessly connect to wearable devices such as smart headphones and smart glasses to achieve information sharing between devices.

[0068] In some embodiments, the translation method of this invention supports cross-device synchronization and is suitable for environments with multi-person collaboration and multi-scenario transitions, such as company meetings and teaching environments. In large conferences, users can simultaneously input and receive output through different devices.

[0069] In some embodiments, users can choose to receive information from specific target devices to avoid redundancy. For example, in a customer service scenario, customers only receive responses from the customer service terminal, while customer service representatives need to receive information from all customer terminals to ensure effective multi-party communication.

[0070] In some embodiments, to enhance user privacy, the translator supports local processing of voice and translation data, ensuring secure data transmission. Compared to most existing translation devices that rely on cloud processing, this invention improves data security through local processing, making it particularly suitable for enterprise or individual users with high privacy requirements. It can provide high-quality translation services even without a network connection, avoiding the limitations of existing devices that cannot be used without internet access, making it suitable for travel or scenarios where internet access is unavailable.

[0071] In some embodiments, the translator of the present invention can adjust the voice style or translation method according to the user's needs or the scenario, avoiding a monotonous and boring interactive experience. For example, some users may prefer the translation results to be presented in a more formal tone, while others may prefer a more natural and conversational expression.

[0072] In some embodiments, the translator of the present invention supports text input of any length, making it suitable for translating long articles, legal documents, and other content. It provides a level of flexibility that existing translation devices cannot meet, and is particularly suitable for translating complex or professional documents.

[0073] In some embodiments, the translator of the present invention can detect and switch languages ​​in real time during the translation process, automatically adapting to the context and avoiding interruption of the translation process due to language switching.

[0074] The following is for reference. Figure 3 An electronic device according to an embodiment of the present invention is described.

[0075] Figure 3 This is a block diagram of an electronic device according to an embodiment of the present invention, such as... Figure 3 As shown, the electronic device of this embodiment includes at least one processor and a memory.

[0076] In some embodiments, at least one processor can be one processor, or multiple processors such as two processors, three processors, five processors, etc. The processor can be a single-core or multi-core processor, responsible for executing computer programs and implementing functions such as input information acquisition, translation, data push, and result display.

[0077] In some embodiments, the processor may be a central processing unit (CPU), a graphics processing unit (GPU), or a digital signal processor (DSP), etc., in an electronic device. This depends specifically on the design and purpose of the electronic device.

[0078] In some embodiments, the memory is a device for storing data and programs, and may be random access memory (RAM), read-only memory (ROM), flash memory, or other storage media. The memory is used to store translation-related data, computer programs, and intermediate processing results.

[0079] In some embodiments, a computer program is stored in the memory and can be executed by at least one processor, which executes the computer program to implement the translation method described in the above embodiments.

[0080] According to an embodiment of the present invention, an electronic device has at least one processor executing a computer program that implements the translation method described in the above embodiment. By acquiring input information from multiple target devices, it achieves simultaneous acquisition of input data in multiple languages. Then, using parallel processing technology, a deep learning model, and a translation model, it performs real-time language recognition and translation on the input information from each target device, ensuring that each input information can be simultaneously converted into a translation result for the corresponding target language. These translation results are synchronously transmitted back to their respective target devices through a parallel output mechanism, avoiding the time delay caused by sequential translation in traditional serial processing methods. This parallel processing method ensures that multilingual information can be processed efficiently and in real-time, ultimately achieving synchronous translation output between multiple devices. This enhances the multilingual translation function of the translator, improves the processing efficiency of multi-device, multilingual translation, and thus enhances the experience of synchronous interaction and cross-language communication across multiple devices.

[0081] This invention also proposes a computer-readable storage medium storing a computer program thereon, which, when executed, implements the translation method described in the above embodiments.

[0082] The computer-readable storage medium in the embodiments of the present invention may include, but is not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be described in detail here.

[0083] According to the computer-readable storage medium of the present invention, by employing the translation method described in the above embodiments, parallel processing and parallel translation output of multilingual information can be realized, thereby improving the multilingual translation function of the translator, increasing the processing efficiency of multi-device and multilingual translation, and thus improving the experience of multi-device synchronous interaction and cross-language communication.

[0084] The following is for reference. Figure 4 A translator according to an embodiment of the present invention is described.

[0085] Figure 4 This is a block diagram of a translator according to an embodiment of the present invention, such as... Figure 4 As shown, the translator includes a receiving module and a processing module.

[0086] In some embodiments, the receiving module can be used to receive input information, which can be multimodal data, including voice, text, images, and video. For example, in a conference scenario, the receiving module can simultaneously receive voice input sent by participants via smartphones or smart headsets, as well as image and text information transmitted via tablets or cameras. The receiving module can support multiple communication protocols (such as Wi-Fi, Bluetooth, 5G, etc.) to ensure that data can be transmitted to the translation device in real time from different sources.

[0087] In some embodiments, the processing module is connected to the receiving module to execute the translation method described in the above embodiments. Specifically, the processing module first preprocesses the received input information, for example, by converting speech to text using Automatic Speech Recognition (ASR) or extracting text from an image using Optical Character Recognition (OCR). Then, the processing module automatically detects the input language using a deep learning model and, based on user settings or the automatic recognition results, calls the corresponding translation model for real-time translation. Finally, the processing module pushes the translation results back to the target device or other communication terminals in the form of text, voice, images, or video subtitles through output methods such as split-screen or multi-channel audio, achieving synchronous interaction between multiple devices and multiple languages.

[0088] According to embodiments of the present invention, the processing module of the translator executes the translation method described in the above embodiments, acquiring input information from multiple target devices to simultaneously collect input data in multiple languages. Then, using parallel processing technology, deep learning models, and translation models, real-time language recognition and translation are performed on the input information from each target device, ensuring that each input information can be simultaneously converted into translation results for the corresponding target language. These translation results are synchronously transmitted back to their respective target devices through a parallel output mechanism, avoiding the time delay caused by sequential translation in traditional serial processing methods. This parallel processing method ensures that multilingual information can be processed efficiently and in real-time, ultimately achieving synchronous translation output between multiple devices, improving the multilingual translation function of the translator, increasing the processing efficiency of multi-device, multilingual translation, and thus enhancing the experience of synchronous interaction and cross-language communication across multiple devices.

[0089] In some embodiments, such as Figure 4 As shown, the processing module includes an artificial intelligence unit and a local translation unit.

[0090] The artificial intelligence unit can be used to perform speech recognition on the input information, extract text information from the input information, and determine the type of input language based on the text information.

[0091] Specifically, the artificial intelligence unit is responsible for the initial processing of the input information collected by the target device. This unit uses Automatic Speech Recognition (ASR) technology to convert speech input into text; for image or video input, it uses Optical Character Recognition (OCR) technology to extract the text information. The extracted text data is then further processed by a pre-trained deep learning model (e.g., a Transformer- or BERT-based model). By analyzing the text's grammar, vocabulary, and contextual information, the system automatically determines the language of the input information. This step ensures that the system can accurately identify the original language of the input information without requiring the user to manually set the input language, providing an accurate linguistic foundation for subsequent translation.

[0092] In some embodiments, the local translation unit can be used to obtain translation information for the target language corresponding to the input information, and to automatically translate the feedback information for the target language into information for the target input language.

[0093] Specifically, the local translation unit utilizes the input language type determined by the artificial intelligence unit to translate the input information into a predetermined target language and obtain the corresponding translation information. Simultaneously, this unit also has a feedback function, automatically translating the feedback information generated in the target language back into the original input language, thereby achieving bidirectional language conversion and interaction. For example, in a multilingual conference scenario, the system can translate a participant's speech (e.g., English) into the target language (e.g., Chinese) and push it to other participants; when other participants reply in Chinese, the local translation unit automatically translates the Chinese feedback back into English and sends it back to the original speaker, ensuring that both parties can communicate in their respective familiar languages.

[0094] In some embodiments, such as Figure 4 As shown, the translator also includes an output module. The output module is connected to the processing module and is used to output translated information.

[0095] In some embodiments, the output module can be a display screen or a voice playback module. Specifically, the display screen can be used to display translated information in the form of text or images, while the voice playback module can convert the translated text into voice output through text-to-speech technology to meet the needs of different users and scenarios.

[0096] In some embodiments, the translator further includes a communication module. The communication module is connected to the processing module and is used to send translation information.

[0097] In some embodiments, the communication module acts as a data transmission bridge between the translator and the device. It transmits the translated information, processed by the processing module, as digital data via wired (e.g., Ethernet, USB (Universal Serial Bus)) or wireless (e.g., Wi-Fi, Bluetooth, 5G) communication methods. This not only enables the output of translated information within the local device (e.g., via a display screen or voice playback module) but also allows for network transmission, pushing the translated information to remote devices, cloud services, or other terminals, achieving cross-device and cross-platform information sharing and synchronization.

[0098] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example.

[0099] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A translation method, characterized in that, For use in a translation machine, the translation method includes: Obtain input information from multiple target devices; Input information from multiple target devices is processed in parallel to obtain translation information for target language types corresponding to the multiple input information. The translation information for the target language type is output in parallel from multiple input information.

2. The translation method according to claim 1, characterized in that, The parallel output of translation information corresponding to the target language for multiple input information items includes: The translation information can be output separately via split-screen or multi-channel audio.

3. The translation method according to claim 1, characterized in that, The translation method also includes: Obtain the input language type corresponding to each input information; The feedback information of the target language type is automatically translated into the information of the target's input language type, and sent to the target device corresponding to the target's input speech type.

4. The translation method according to claim 3, characterized in that, Obtaining the input language type corresponding to each input information includes: The text information in the input information is extracted using a deep learning model, and the type of input language corresponding to the input information is determined based on the text information.

5. The translation method according to claim 3, characterized in that, The translation method also includes: The input voice type is sent to the target device corresponding to the target in at least one of the forms of text, image, and voice.

6. The translation method according to any one of claims 1-5, characterized in that, The input information of the multiple target devices is multimodal information.

7. The translation method according to any one of claims 1-5, characterized in that, The translation method also includes: The translation information is sent to a target receiving device, which includes the target device and / or other communication terminals.

8. The translation method according to any one of claims 1-5, characterized in that, The target device includes at least one of a mobile terminal and a wearable device.

9. An electronic device, characterized in that, include: At least one processor; A memory that is communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which, when executing the computer program, implements the translation method according to any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the translation method according to any one of claims 1-8.

11. A translator, characterized in that, include: The receiving module is used to receive input information; A processing module, connected to the receiving module, is used to execute the translation method according to any one of claims 1-8.

12. The translator according to claim 11, characterized in that, The processing module includes: An artificial intelligence unit is used to perform speech recognition on input information to extract text information from the input information and determine the type of input language corresponding to the input information based on the text information; The local translation unit is used to obtain translation information for the target language corresponding to the input information, and to automatically translate the feedback information for the target language into information for the target input language.

13. The translator according to claim 11, characterized in that, The translator also includes: The output module, connected to the processing module, is used to output translation information.

14. The translator according to claim 11, characterized in that, The translator also includes: A communication module, connected to the processing module, is used to send translation information.

Images