Abstract generation method, device, electronic device and storage medium

Abstract

A method, device, electronic device and storage medium for generating a summary. The method for generating a summary includes: generating a first summary about a first text, and generating a second summary about a second text (201); the second text is a translation text corresponding to the first text; translating the first summary to obtain a third summary, and translating the second summary to obtain a fourth summary (202); determining a first similarity between the second summary and the third summary, and determining a second similarity between the first summary and the fourth summary (203); based on the first similarity and the second similarity, selecting the second summary or the third summary as a summary translation of the first text (204); wherein the first text, the first summary and the fourth summary correspond to a first language; and the second text, the second summary and the third summary correspond to a second language.

Description

Technical Field

[0001] The present application relates to machine translation technology, and in particular to a summary generation method, device, electronic device and storage medium. Background Art

[0002] In cross-language speeches or conferences, speakers speak for a long time and have a lot of content. Therefore, in addition to simultaneous interpretation of the speech content, a summary translation related to the speech content is often generated to allow participants to quickly understand the speech content. In related technologies, the generated summary translation cannot accurately reflect the speech content. Summary of the invention

[0003] To solve the related technical problems, the embodiments of the present application provide a summary generation method, device, electronic device and storage medium.

[0004] The present application provides a method for generating a summary, including:

[0005] Generate a first summary about the first text, and generate a second summary about the second text; the second text is a translation text corresponding to the first text;

[0006] translating the first abstract to obtain a third abstract, and translating the second abstract to obtain a fourth abstract;

[0007] determining a first similarity between the second abstract and the third abstract, and determining a second similarity between the first abstract and the fourth abstract;

[0008] Based on the first similarity and the second similarity, the second abstract or the third abstract is selected as the abstract translation of the first text; wherein,

[0009] The first text, the first abstract and the fourth abstract correspond to a first language; the second text, the second abstract and the third abstract correspond to a second language.

[0010] Among them, in the above scheme, the selecting the second abstract or the third abstract as the abstract translation of the first text based on the first similarity and the second similarity includes:

[0011] determining a larger value between the first similarity and the second similarity;

[0012] When the larger value is the first similarity, selecting the third abstract as the abstract translation;

[0013] When the larger value is the second similarity, the second summary is selected as the summary translation.

[0014] In the above solution, determining the larger value between the first similarity and the second similarity includes:

[0015] determining a difference between the first similarity and the second similarity;

[0016] When the difference is greater than a set threshold, a larger value between the first similarity and the second similarity is determined.

[0017] In the above solution, the selecting the second abstract or the third abstract as the abstract translation of the first text based on the first similarity and the second similarity includes:

[0018] determining a difference between the first similarity and the second similarity;

[0019] When the difference is less than or equal to a set threshold, any one of the second abstract and the third abstract is selected as the abstract translation.

[0020] In the above scheme, the method further comprises:

[0021] The first text is determined, and the second text is determined.

[0022] In the above solution, determining the first text includes:

[0023] Collecting the speech content of the speaker to obtain a first speech;

[0024] Perform speech recognition on the first speech to obtain the first text.

[0025] In the above solution, determining the second text includes:

[0026] The first text is translated to obtain the second text.

[0027] In the above solution, determining the second text includes:

[0028] Performing voice collection on the simultaneous interpretation content corresponding to the speech content to obtain a second voice;

[0029] Perform speech recognition on the second speech to obtain the second text.

[0030] In the above scheme, the method further comprises:

[0031] The summary translation of the first text is displayed while the speaker is speaking; wherein the speech content of the speaker is related to the text content of the first text.

[0032] The present application also provides a summary generation device, the device comprising:

[0033] A generating unit configured to generate a first summary about a first text and a second summary about a second text; the second text is a translation text corresponding to the first text;

[0034] a translation unit configured to translate the first abstract to obtain a third abstract, and to translate the second abstract to obtain a fourth abstract;

[0035] a first determining unit configured to determine a first similarity between the second abstract and the third abstract, and to determine a second similarity between the first abstract and the fourth abstract;

[0036] A second determination unit is configured to select the second abstract or the third abstract as the abstract translation of the first text based on the first similarity and the second similarity; wherein:

[0037] The first text, the first abstract and the fourth abstract correspond to a first language; the second text, the second abstract and the third abstract correspond to a second language.

[0038] An embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of any of the above-mentioned summary generation methods are implemented.

[0039] An embodiment of the present application further provides a storage medium on which computer instructions are stored. When the computer instructions are executed by a processor, the steps of any of the above-mentioned summary generation methods are implemented.

[0040] The summary generation method, device, electronic device and storage medium provided in the embodiments of the present application generate a first summary about a first text, and generate a second summary about a translation text corresponding to the first text, translate the first summary to obtain a third summary, and translate the second summary back to obtain a fourth summary, and then compare the two similarities based on the first similarity between the second summary and the third summary, and the second similarity between the first summary and the fourth summary, and select the second summary or the third summary as the summary translation of the first text according to the comparison result, so that the final determined summary translation can more accurately reflect the content of the first text. In cross-language speech occasions or conference occasions, the scheme of the embodiments of the present application can generate a summary that more accurately reflects the content of the speech, thereby improving the accuracy of information transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1A schematic diagram of a system architecture for applying a summary generation method in related technologies;

[0042] Figure 2 A schematic diagram of a process flow of a summary generation method according to an embodiment of the present application;

[0043] Figure 3 A schematic diagram of a process flow of a summary generation method according to an embodiment of the present application;

[0044] Figure 4 A schematic diagram of a process flow of a summary generation method according to an embodiment of the present application;

[0045] Figure 5 A schematic diagram of the composition structure of a summary generation device according to an embodiment of the present application;

[0046] Figure 6 A schematic diagram of the hardware structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0047] Before describing the technical solution of the embodiment of the present application in detail, a system to which a summary generation method in related art is applied is briefly described first.

[0048] Figure 1 Schematic diagram of the system architecture for the application of the summary generation method in the related art. Figure 1 As shown, the system may include: a machine simultaneous interpretation server, a speech recognition server, a translation server, a mobile terminal sending server, an audience mobile terminal, a computer (PC, Personal Computer) client, and a display screen.

[0049] In actual applications, speakers can make conference speeches through PC clients, and project the displayed documents, such as documents of presentation software (PPT, PowerPoint), onto the display screen, and display them to users through the display screen. During the conference speech, the PC client collects the speaker's audio and sends the collected audio to the machine simultaneous interpretation server. The machine simultaneous interpretation server recognizes the audio data through the speech recognition server to obtain the recognized text, and then translates the recognized text through the translation server to obtain the translation result; the machine simultaneous interpretation server sends the translation result to the PC client, and sends the translation result to the audience's mobile terminal through the mobile terminal sending server, and displays the translation result to the user, thereby translating the speaker's speech content into the language required by the user and displaying it.

[0050] In addition, in the solution of the related technology, the machine simultaneous interpretation server can also generate a summary translation of the speech content based on the translation results of the translation server, and send the summary translation to the audience mobile terminal through the mobile terminal distribution server to help users quickly understand the speaker's speech content. However, taking cross-language speech occasions or cross-language conference occasions as an example, each speaker speaks for a long time and has a lot of speech content. When the speech content is large, the translation server performs machine translation on the speech content, and the translation results obtained will have more or less translation errors. In this case, the generated summary translation cannot accurately reflect the speech content, which has a negative effect on the user's accurate understanding of the speech content and reduces the accuracy of information transmission.

[0051] Based on this, in various embodiments of the present application, two different methods are used to obtain two summary translations about the speech content, and by back-translating, comparing the similarities between different summary originals and comparing the similarities between different summary translations, one of the two summary translations is finally selected as the summary translation about the speech content, thereby selecting a summary translation that can more accurately reflect the speech content and accurately realizing information transmission.

[0052] The present application is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0053] An embodiment of the present application provides a summary generation method, which is applied to an electronic device. Figure 2 FIG. 1 is a flow chart of a method for generating a summary according to an embodiment of the present application. Figure 2 As shown, the method includes:

[0054] Step 201: Generate a first summary about a first text, and generate a second summary about a second text; the second text is a translation text corresponding to the first text.

[0055] The second text is a translation text corresponding to the first text, that is, the first text is the original text and the second text is the corresponding translation. Here, based on the first text and the second text, a first summary and a second summary are generated respectively. When generating the summary, the methods that can be used may include at least one of the following:

[0056] Generate summaries in an extractive way;

[0057] Generate a summary in an abstractive way.

[0058] Among them, generating a summary in an extractive way means extracting some representative text fragments from the text to form a summary. These fragments can be words, sentences, paragraphs or sections in the entire text. In the extraction process, the fragments in the text are first sorted, the top-ranked fragments are selected, and then the redundant information in the selected fragments is removed to finally generate a summary. Generating a summary in a generative way means forming a summary by establishing an abstract semantic representation and using natural language generation technology. For an extractive summary, the similarity with the text is high, but the readability is poor and the amount of information is low. For a generative summary, it has high readability and rich information, but the similarity with the text is low. In practical applications, the extractive method and the generative method can also be combined. For example, the text is first processed in an extractive way to obtain an extractive summary, and then the obtained extractive summary is processed in a generative way, so as to obtain a summary that can take into account the similarity of the text and the readability and information content.

[0059] In one embodiment, the method further comprises:

[0060] The first text is determined, and the second text is determined.

[0061] Before generating the first summary and the second summary, the first text and the second text are first determined. Here, the first text is the original text, and the first text can be the speech manuscript of the speaker. In addition, in one embodiment, the determining of the first text includes:

[0062] Collecting the speech content of the speaker to obtain a first speech;

[0063] Perform speech recognition on the first speech to obtain the first text.

[0064] Here, when the speaker is speaking, the speaker's voice can be collected through a voice collection module, such as a microphone, to obtain a first voice, and then the first voice is converted into a first text through voice recognition, thereby obtaining the original text corresponding to the actual speech content of the speaker.

[0065] In addition, after obtaining the first text, the first text can be post-processed by adding punctuation and text smoothing. Adding punctuation can enable the first text to be reasonably segmented, and text smoothing can adjust the word order of the text recognized by speech, or add conjunctions, auxiliary words, etc. to the text recognized by speech, so as to overcome the defect of spoken language of the text recognized by speech and obtain the first text that conforms to grammatical expression.

[0066] The second text is a translation corresponding to the first text. In one embodiment, determining the second text includes:

[0067] The first text is translated to obtain the second text.

[0068] The second text is generated based on the first text, and the translation corresponding to the first text is obtained by translating the determined first text as the second text. Here, the translation quality of the second text for the speech content is largely related to the accuracy of the first text's description of the speech content.

[0069] In one embodiment, determining the second text includes:

[0070] Performing voice collection on the simultaneous interpretation content corresponding to the speech content to obtain a second voice;

[0071] Perform speech recognition on the second speech to obtain the second text.

[0072] When a speaker is speaking, if the speech is interpreted simultaneously, the voice of the simultaneous interpreter can be collected through a voice collection module, such as a microphone, to obtain a second voice, and then the second voice can be converted into a second text through voice recognition, thereby obtaining a translation corresponding to the actual speech content of the speaker.

[0073] Here, the translation quality of the second text for the speech content is no longer related to the accuracy of the first text's description of the speech content, but is related to the translation quality of simultaneous interpretation.

[0074] In addition, after obtaining the second text by performing speech recognition on the simultaneous interpretation content, it is also necessary to perform post-processing on the second text such as adding punctuation marks and text smoothing to obtain a second text that conforms to grammatical expression.

[0075] In actual applications, the second text can be obtained in a corresponding manner according to the application scenario. For example, if the speaker provides a speech manuscript in advance, the second text can be obtained by translating the first text; for another example, if the speaker is speaking impromptu, the second text can be determined by performing speech recognition on the content of the simultaneous interpretation.

[0076] Step 202: Translate the first abstract to obtain a third abstract, and translate the second abstract to obtain a fourth abstract.

[0077] Here, the first abstract is an abstract of the first text, which is the original abstract, and the second abstract is an abstract of the second text, which is the translated abstract. Therefore, the third abstract obtained by translating the first abstract is the translated abstract of the first text, and the fourth abstract obtained by translating the second abstract is the original abstract of the second text.

[0078] For the sake of convenience, the above-mentioned first text, first summary and fourth summary all correspond to the first language, that is, the language used by the speaker when speaking, and the above-mentioned second text, second summary and third summary all correspond to the second language, that is, the translated language.

[0079] In practical applications, machine translation technology can be used to translate from a first language to a second language, and back from a second language to a first language. Specifically, the server can use rule-based machine translation technology or corpus-based machine translation technology for translation.

[0080] Step 203: Determine a first similarity between the second abstract and the third abstract, and determine a second similarity between the first abstract and the fourth abstract.

[0081] Since the first summary, the second summary, the third summary and the fourth summary are all texts, in actual application, the bilingual translation quality auxiliary tool (BLEU, Bilingual Evaluation Understudy) or ROUGE (Recall-oriented Understanding for Gisting Evaluation) can be used to calculate the similarity between the second summary and the third summary, or between the first summary and the fourth summary. Among them, BLEU is an indicator used to measure the similarity between two texts. The final similarity value ranges from 0 to 1. The closer the value is to 1, the more semantically similar the two texts are; ROUGE compares and calculates the two texts to obtain the corresponding score to measure the similarity between the two texts.

[0082] Step 204: Based on the first similarity and the second similarity, select the second abstract or the third abstract as the abstract translation of the first text.

[0083] In one embodiment, Figure 3 As shown, the selecting the second abstract or the third abstract as the abstract translation of the first text based on the first similarity and the second similarity includes:

[0084] Step 2041: Determine a larger value between the first similarity and the second similarity;

[0085] Step 2042: When the larger value is the first similarity, select the third abstract as the abstract translation; when the larger value is the second similarity, select the second abstract as the abstract translation.

[0086] Here, the first similarity is compared with the second similarity. When the comparison result is that the first similarity is larger, that is, the similarity between the second abstract and the third abstract in the same translation language (second language) is higher, it means that the quality of the generated abstract original text (first abstract) about the first text is higher, and the translation quality of the abstract translation (third abstract) obtained based on the translation of the abstract original text is also higher, and the third abstract reflects that the content of the first text is more accurate, so the third abstract is used as the abstract translation of the first text; when the comparison result is that the second similarity is larger, that is, the similarity between the first abstract and the fourth abstract in the same original language (first language) is higher, it means that the translation quality of the second text obtained based on the translation of the first text is higher, and the quality of the second abstract generated about the second text is also higher, and the second abstract reflects that the content of the first text is more accurate, so the second abstract is used as the abstract translation of the first text.

[0087] In one embodiment, determining the larger value between the first similarity and the second similarity includes:

[0088] determining a difference between the first similarity and the second similarity;

[0089] When the difference is greater than a set threshold, a larger value between the first similarity and the second similarity is determined.

[0090] Here, when it is determined that the difference between the first similarity and the second similarity is greater than the set threshold, it is considered that the semantic similarity between the second abstract and the third abstract is not large enough, and then the larger value between the first similarity and the second similarity is further determined to select the second abstract or the third abstract as the abstract translation of the first text. When the difference between the first similarity and the second similarity is less than the set threshold, it is considered that the semantic similarity between the second abstract and the third abstract is high. In this case, either the second abstract or the third abstract can be selected as the abstract translation of the first text. Figure 4 As shown, the selecting the second abstract or the third abstract as the abstract translation of the first text based on the first similarity and the second similarity includes:

[0091] Step 2043: Determine a difference between the first similarity and the second similarity;

[0092] Step 2044: When the difference is less than or equal to a set threshold, select any one of the second summary and the third summary as the summary translation.

[0093] In one embodiment, the method further comprises:

[0094] The summary translation of the first text is displayed while the speaker is speaking; wherein the speech content of the speaker is related to the text content of the first text.

[0095] Here, the text content of the first text is the speech content of the speaker. The first text can be the speech manuscript of the speaker, or it can be the recognized text obtained by real-time voice collection of the speaker's audio and then voice recognition. While the speaker is speaking, a summary translation of the first text is generated, so that the participants in the meeting can better understand the speaker's speech content in real time and grasp the key points of the speech, thereby realizing accurate and rapid transmission of information.

[0096] The summary generation method provided by the embodiment of the present application generates a first summary about the first text, and generates a second summary about the translation text corresponding to the first text, translates the first summary to obtain a third summary, and flips back the second summary to obtain a fourth summary, and then, based on the first similarity between the second summary and the third summary, and the second similarity between the first summary and the fourth summary, compares the two similarities, and selects the second summary or the third summary as the summary translation of the first text according to the comparison result, so that the final determined summary translation can more accurately reflect the content of the first text. In cross-language speech occasions or conference occasions, the scheme of the embodiment of the present application can generate a summary that more accurately reflects the content of the speech, and improves the accuracy of information transmission. In the scene of simultaneous interpretation, the scheme of the embodiment of the present application can avoid information transmission errors caused by translation errors of machine translation to a certain extent, and more accurately generate a summary about the content of the speech.

[0097] In order to implement the translation method of the embodiment of the present application, the embodiment of the present application also provides a summary generation device, which can be set in an electronic device. Figure 5 Schematic diagram of the composition structure of the summary generation device in the embodiment of the present application. Figure 5 As shown, the summary generating device includes:

[0098] A generating unit 51 is configured to generate a first summary about a first text and a second summary about a second text; the second text is a translation text corresponding to the first text;

[0099] a translation unit 52, configured to translate the first abstract to obtain a third abstract, and translate the second abstract to obtain a fourth abstract;

[0100] a first determining unit 53, configured to determine a first similarity between the second abstract and the third abstract, and to determine a second similarity between the first abstract and the fourth abstract;

[0101] The second determining unit 54 is configured to select the second abstract or the third abstract as the abstract translation of the first text based on the first similarity and the second similarity; wherein

[0102] The first text, the first abstract and the fourth abstract correspond to a first language; the second text, the second abstract and the third abstract correspond to a second language.

[0103] In one embodiment, the second determining unit 54 is further configured to:

[0104] determining a larger value between the first similarity and the second similarity;

[0105] When the larger value is the first similarity, selecting the third abstract as the abstract translation;

[0106] When the larger value is the second similarity, the second summary is selected as the summary translation.

[0107] In one embodiment, when the second determination unit 54 determines the larger value between the first similarity and the second similarity, it is further configured as follows:

[0108] determining a difference between the first similarity and the second similarity;

[0109] When the difference is greater than a set threshold, a larger value between the first similarity and the second similarity is determined.

[0110] In one embodiment, the second determining unit 54 is further configured to:

[0111] determining a difference between the first similarity and the second similarity;

[0112] When the difference is less than or equal to a set threshold, any one of the second abstract and the third abstract is selected as the abstract translation.

[0113] In one embodiment, the device further comprises:

[0114] The third determining unit is configured to determine the first text and the second text.

[0115] In one embodiment, the third determining unit is further configured to:

[0116] Collecting the speech content of the speaker to obtain a first speech;

[0117] Perform speech recognition on the first speech to obtain the first text.

[0118] In one embodiment, the third determining unit is further configured to:

[0119] The first text is translated to obtain the second text.

[0120] In one embodiment, the third determining unit is further configured to:

[0121] Performing voice collection on the simultaneous interpretation content corresponding to the speech content to obtain a second voice;

[0122] Perform speech recognition on the second speech to obtain the second text.

[0123] In one embodiment, the device further comprises:

[0124] The display unit is configured to display a summary translation of the first text while the speaker is speaking; wherein the speech content of the speaker is related to the text content of the first text.

[0125] In actual application, the generation unit 51, the translation unit 52, the first determination unit 53, the second determination unit 54, and the third determination unit can be implemented by a processor in an electronic device, such as a central processing unit (CPU), a digital signal processor (DSP), a microcontroller unit (MCU) or a programmable gate array (FPGA); the display unit can be implemented through a communication interface in the electronic device.

[0126] It should be noted that: the summary generation device provided in the above embodiment only uses the division of the above program modules as an example to illustrate when performing summary generation. In actual applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the electronic device is divided into different program modules to complete all or part of the above-described processing. In addition, the summary generation device provided in the above embodiment and the summary generation method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0127] The summary generating device in the above embodiment may be correspondingly arranged at Figure 1 In the system shown in FIG. 1 , that is, the corresponding Figure 1 The summary generation method of the embodiment of the present application is applied to the machine simultaneous interpretation server of the system. Figure 1 in the system shown.

[0128] Based on the hardware implementation of the above-mentioned device, an embodiment of the present application also provides an electronic device. Figure 6 Schematic diagram of the hardware structure of the electronic device of the embodiment of the present application. Figure 6 As shown, the electronic device 60 includes a memory 63, a processor 62, and a computer program stored in the memory 63 and executable on the processor 62; when the processor 62 executes the program, the method provided by one or more of the above technical solutions is implemented.

[0129] Specifically, when the processor 62 in the electronic device 60 executes the program, it is achieved: generating a first summary about the first text, and generating a second summary about the second text; the second text is a translated text corresponding to the first text; translating the first summary to obtain a third summary, and translating the second summary to obtain a fourth summary; determining a first similarity between the second summary and the third summary, and determining a second similarity between the first summary and the fourth summary; based on the first similarity and the second similarity, selecting the second summary or the third summary as the summary translation of the first text; wherein the first text, the first summary and the fourth summary correspond to a first language; and the second text, the second summary and the third summary correspond to a second language.

[0130] It should be noted that the specific steps implemented when the processor 62 executes the program have been described in detail in the above embodiment and will not be repeated here.

[0131] It is understood that the electronic device 60 also includes a communication interface 61; the various components in the electronic device 60 are coupled together via a bus system 64. It is understood that the bus system 64 is configured to achieve connection and communication between these components. In addition to the data bus, the bus system 64 also includes a power bus, a control bus, and a status signal bus.

[0132] Understandably, the above Figure 6The memory in the embodiment may be a volatile memory or a non-volatile memory, and may also include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a ferromagnetic random access memory, a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory may be a disk memory or a tape memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), direct memory bus random access memory (DRRAM). The memory described in the embodiments of the present application is intended to include but is not limited to these and any other suitable types of memory.

[0133] The method disclosed in the above embodiment of the present application can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor may be a general-purpose processor, a DSP, or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or any conventional processor, etc. In combination with the steps of the method disclosed in the embodiment of the present application, it can be directly embodied as a hardware decoding processor to execute, or it can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium, which is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0134] The present application also provides a storage medium, specifically a computer storage medium, more specifically a computer readable storage medium, on which computer instructions, namely a computer program, are stored, and when the computer instructions are executed by a processor, the method provided by one or more of the above technical solutions is implemented.

[0135] In the several embodiments provided in the present application, it should be understood that the disclosed methods and intelligent devices can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0136] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0137] In addition, all functional units in the embodiments of the present application may be integrated into a second processing unit, or each unit may be a separate unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0138] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium, which, when executed, executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, disks or optical disks.

[0139] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiment of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0140] It should be noted that: "first", "second", etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0141] In addition, the technical solutions described in the embodiments of the present application can be combined arbitrarily without conflict.

[0142] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for generating a summary, comprising: generating a first summary about the first text, and generating a second summary about the second text; The second text is a translation text corresponding to the first text; translating the first abstract to obtain a third abstract, and translating the second abstract to obtain a fourth abstract; determining a first similarity between the second abstract and the third abstract, and determining a second similarity between the first abstract and the fourth abstract; Based on the comparison result of the first similarity and the second similarity, the second abstract or the third abstract is selected as the abstract translation of the first text; wherein, The first text, the first abstract and the fourth abstract correspond to a first language; the second text, the second abstract and the third abstract correspond to a second language.

2. The method according to claim 1, wherein: The selecting the second abstract or the third abstract as the abstract translation of the first text based on the comparison result of the first similarity and the second similarity includes: determining a larger value between the first similarity and the second similarity; When the larger value is the first similarity, selecting the third abstract as the abstract translation; When the larger value is the second similarity, the second summary is selected as the summary translation.

3. The method according to claim 2, wherein: The determining a larger value between the first similarity and the second similarity includes: determining a difference between the first similarity and the second similarity; When the difference is greater than a set threshold, a larger value between the first similarity and the second similarity is determined.

4. The method according to claim 1, wherein: The selecting the second abstract or the third abstract as the abstract translation of the first text based on the comparison result of the first similarity and the second similarity includes: determining a difference between the first similarity and the second similarity; When the difference is less than or equal to a set threshold, any one of the second abstract and the third abstract is selected as the abstract translation.

5. The method according to claim 1, wherein: The method further comprises: The first text is determined, and the second text is determined.

6. The method according to claim 5, wherein: The determining the first text includes: Collecting the speech content of the speaker to obtain a first speech; Perform speech recognition on the first speech to obtain the first text.

7. The method according to claim 5, wherein: The determining the second text comprises: The first text is translated to obtain the second text.

8. The method according to claim 6, wherein: The determining the second text comprises: Performing voice collection on the simultaneous interpretation content corresponding to the speech content to obtain a second voice; Perform speech recognition on the second speech to obtain the second text.

9. The method according to claim 1, wherein: The method further comprises: The summary translation of the first text is displayed while the speaker is speaking; wherein the speech content of the speaker is related to the text content of the first text.

10. A summary generation device, comprising: a generating unit configured to generate a first summary about the first text and generate a second summary about the second text; The second text is a translation text corresponding to the first text; a translation unit configured to translate the first abstract to obtain a third abstract, and to translate the second abstract to obtain a fourth abstract; a first determining unit configured to determine a first similarity between the second abstract and the third abstract, and to determine a second similarity between the first abstract and the fourth abstract; A second determination unit is configured to select the second abstract or the third abstract as the abstract translation of the first text based on the comparison result of the first similarity and the second similarity; wherein, The first text, the first abstract and the fourth abstract correspond to a first language; the second text, the second abstract and the third abstract correspond to a second language.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

12. A storage medium having computer instructions stored thereon, wherein the computer instructions, when executed by a processor, implement the steps of the method according to any one of claims 1 to 9.

Images