Conference information processing method and device, terminal and computer readable storage medium

Abstract

Embodiments of the present application disclose a conference information processing method and device, a terminal and a computer readable storage medium. The method comprises: monitoring a language type of online conference voice data in a target conference; when it is determined that a language type of first real-time voice information in the online conference voice data is a target language type, generating first text corresponding to the first real-time voice information, and displaying the first text on a conference interface; when it is determined that a language type of second real-time voice information in the online conference voice data is a language type to be translated, performing target language type text conversion processing on the second real-time voice information to obtain second text, and displaying the second text on the conference interface. Thus, the technical problem of weak visibility of information transmission in the process of online video conference is solved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, terminal and computer-readable storage medium for processing conference information. Background Technology

[0002] In recent years, the rapid promotion of online conferencing applications has led to its widespread adoption. However, when conducting online meetings, participants often rely on the speaker's online audio to understand the meeting content, resulting in weak visibility of information transmission during the communication process. Summary of the Invention

[0003] This application provides a method, apparatus, terminal, and computer-readable storage medium for processing conference information, which can solve the technical problem of poor visibility of information transmission during online meetings.

[0004] In a first aspect, embodiments of this application provide a method for processing conference information, the method comprising:

[0005] Monitor the language type of online meeting audio data in the target meeting;

[0006] When the language type of the first real-time voice information in the online meeting voice data is determined to be the target language type, the first text corresponding to the first real-time voice information is generated, and the first text is displayed as a subtitle on the meeting interface;

[0007] When the language type of the second real-time voice information in the online meeting voice data is determined to be the language type to be translated, the second real-time voice information is subjected to target language type text conversion processing to obtain the second text, and the second text is displayed as subtitles on the meeting interface.

[0008] Optionally, the language type of the online meeting audio data in the monitored target meeting includes:

[0009] Determine the language translation configuration information of the user terminal for the target meeting, and based on the language translation configuration information, monitor the language type of the current voice information corresponding to the online meeting voice data;

[0010] When the language type of the current voice information successfully matches the preset language type in the language translation configuration information, the current voice information is used as the first real-time voice information, and the language type of the first real-time voice information is determined as the target language type.

[0011] When the language type of the current voice information fails to match the preset language type in the language translation configuration information, the current voice information is used as the second real-time voice information, and the language type of the second real-time voice information is determined as the language type to be translated.

[0012] Optionally, the method further includes:

[0013] The local conference audio data in the target conference is stored, and when the target conference is detected to have ended, the target language type text corresponding to the local conference audio data is generated;

[0014] The first time sequence information corresponding to the first real-time voice information of the first text, the second time sequence information corresponding to the second real-time voice information of the second text, and the third time sequence information corresponding to the offline meeting voice of the target language type text are determined.

[0015] Based on the first time sequence information, the second time sequence information, and the third time sequence information, the first text, the second text, and the target language type text are concatenated in chronological order to obtain the meeting content text information corresponding to the target meeting.

[0016] Optionally, after obtaining the meeting content text information corresponding to the target meeting, the method further includes:

[0017] The meeting content text information is input into the meeting minutes generation model. Based on the meeting minutes generation model, the meeting content text information is processed to generate meeting minutes, and the meeting minutes corresponding to the meeting content text information are obtained and output.

[0018] The meeting minutes generation model is obtained by training the basic model using a sample training set, which includes sample meeting content text information and sample meeting minutes corresponding to the sample meeting content text information.

[0019] Optionally, the method further includes:

[0020] Record the target meeting based on the meeting recording command;

[0021] The meeting recording is terminated and meeting recording data is generated when the target meeting ends; wherein, the meeting recording data includes video recording data, audio recording data, and subtitle data;

[0022] Based on the audio recording data, the subtitle data is processed to eliminate audio subtitle delay, resulting in target subtitle data. Target meeting recording data is then determined based on the video recording data, the audio recording data, and the target subtitle data.

[0023] Optionally, the step of performing audio subtitle delay elimination processing on the subtitle data based on the audio recording data to obtain target subtitle data includes:

[0024] Determine the first timing information corresponding to the first real-time speech information in the audio recording data, find the first text corresponding to the first real-time speech information in the subtitle data, and determine the first text timing information corresponding to the first text in the subtitle data;

[0025] Determine the second time sequence information corresponding to the second real-time voice information in the audio recording data, search for the second text corresponding to the second real-time voice information in the subtitle data, and determine the second text time information corresponding to the second text in the subtitle data;

[0026] The first text time information is updated based on the first time sequence information to obtain the target first text time information; the second text time information is updated based on the second time sequence information to obtain the target second text time information.

[0027] The display time of the first and second texts in the subtitle data is adjusted using the target first text time information and the target second text time information, respectively, to obtain the target subtitle data.

[0028] Optionally, after obtaining the target subtitle data, the method further includes:

[0029] Acquire the local conference voice data corresponding to the user terminal, generate the target language type text corresponding to the local conference voice data, and determine the third time sequence information corresponding to the local conference voice data;

[0030] Based on the third time sequence information, subtitle generation processing is performed on the target language type text to obtain local subtitle data, and the local subtitle data is added to the target meeting recording data to update the target meeting recording data;

[0031] The subtitles corresponding to the target subtitle data are displayed in the first subtitle display area of ​​the conference interface, and the subtitles corresponding to the local subtitle data are displayed in the second subtitle display area of ​​the conference interface.

[0032] Optionally, after determining the target meeting recording data based on the video recording data, the audio recording data, and the target subtitle data, the method further includes:

[0033] The target meeting recording data is stored, and the storage path information of the target meeting recording data is determined;

[0034] Determine the historical meeting information corresponding to the target meeting, and use the historical meeting information and the storage path information to generate a historical meeting interface for the target meeting.

[0035] Optionally, the step of performing target language type text conversion processing on the second real-time speech information to obtain the second text includes:

[0036] The second real-time voice information is converted into text to obtain the second real-time voice information text.

[0037] The second real-time voice information text is translated based on the target language type to obtain the second text.

[0038] Secondly, embodiments of this application provide a conference information processing apparatus, the apparatus comprising:

[0039] The monitoring module is suitable for monitoring the language type of online meeting audio data in the target meeting;

[0040] The first determining module is adapted to generate first text corresponding to the first real-time voice information when it determines that the language type of the first real-time voice information in the online meeting voice data is the target language type, and to display the first text as subtitles on the meeting interface.

[0041] The second determining module is adapted to determine that when the language type of the second real-time voice information in the online conference voice data is the language type to be translated, perform target language type text conversion processing on the second real-time voice information to obtain the second text, and display the second text as subtitles on the conference interface.

[0042] Thirdly, embodiments of this application provide a terminal, the terminal comprising:

[0043] Processor; and

[0044] A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the method according to any one of the preceding descriptions.

[0045] Fourthly, embodiments of this application provide a computer-readable storage medium that stores one or more programs, which, when executed by a processor, implement the method described in any one of the above-described methods.

[0046] The beneficial effects of the technical solution provided in this application include at least the following: By monitoring the language type of the online meeting audio data in the target meeting, when the language type of the first real-time audio information is the target language type, the first text corresponding to the first real-time audio information is displayed as subtitles on the meeting interface, making the meeting communication information in the target meeting visible, thereby improving meeting communication efficiency and avoiding reliance on the online voice of the speaker to understand the meeting content. Furthermore, when the language type of the second real-time audio information in the online meeting audio data is the language to be translated, the second real-time audio information is automatically converted to the target language type text to obtain the second text, which is then displayed as subtitles on the meeting interface. This avoids the need for professional translation assistance when participants speaking different languages ​​communicate, shortening the meeting time, and similarly making the meeting communication information in the target meeting visible, effectively improving meeting communication efficiency. This solves the technical problem of poor visibility of information transmission during online meetings. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 An exemplary system architecture diagram of a meeting information processing method provided in this application embodiment;

[0049] Figure 2 A flowchart illustrating a meeting information processing method provided in an embodiment of this application;

[0050] Figure 3 This application provides a flowchart illustrating a language type monitoring process.

[0051] Figure 4 This application provides a schematic diagram of a process for generating meeting content text information in an embodiment of the present application.

[0052] Figure 5 This application provides a schematic diagram of a process for generating target meeting recording data.

[0053] Figure 6 This application provides a schematic diagram of a process for determining target subtitle data in an embodiment of the present application.

[0054] Figure 7 This is a schematic diagram illustrating another process for determining target subtitle data provided in an embodiment of this application;

[0055] Figure 8 A flowchart illustrating the process of generating a historical meeting interface for a target meeting, provided as an embodiment of this application;

[0056] Figure 9 A schematic diagram of the structure of a conference information processing device provided in an embodiment of this application;

[0057] Figure 10 This is a schematic diagram of the structure of a terminal provided in an embodiment of this application. Detailed Implementation

[0058] To make the features and advantages of the embodiments of this application more apparent and understandable, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the embodiments of this application.

[0059] In related technologies, the rapid promotion of online conferencing applications has led to the widespread adoption of online meetings. However, when participants communicate in different languages, it can affect the efficiency of the meeting.

[0060] To address related technical issues, this application provides a meeting information processing method. The method includes: monitoring the language type of online meeting audio data in a target meeting; when the language type of a first real-time audio message in the online meeting audio data is determined to be the target language type, generating a first text corresponding to the first real-time audio message and displaying the first text as subtitles on the meeting interface; when the language type of a second real-time audio message in the online meeting audio data is determined to be the language type to be translated, performing target language type text conversion processing on the second real-time audio message to obtain a second text, and displaying the second text as subtitles on the meeting interface. This solves the technical problem of poor visibility of information transmission during online meetings.

[0061] Please see Figure 1 , Figure 1 This is an exemplary system architecture diagram of a meeting information processing method provided in an embodiment of this application.

[0062] like Figure 1As shown, the system architecture may include a terminal 101, a network 102, and a server 103. The network 102 serves as the medium for providing a communication link between the terminal 101 and the server 103. The network 102 may include various types of wired or wireless communication links, such as wired communication links including fiber optic cables, twisted-pair cables, or coaxial cables, and wireless communication links including Bluetooth communication links, Wi-Fi (Wireless Fidelity) communication links, or microwave communication links, etc.

[0063] Terminal 101 can interact with server 103 via network 102 to receive messages from or send messages to server 103. Alternatively, terminal 101 can interact with server 103 via network 102 to receive messages or data sent to server 103 by other users. Terminal 101 can be hardware or software. When terminal 101 is hardware, it can be various terminals, including but not limited to smartwatches, smartphones, tablets, laptops, and desktop computers. When terminal 101 is software, it can be installed in the terminals listed above, and can be implemented as multiple software programs or software modules (e.g., to provide distributed services) or as a single software program or software module; no specific limitation is made here.

[0064] Server 103 can be a business server providing various services. It should be noted that server 103 can be either hardware or software. When server 103 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When server 103 is software, it can be implemented as multiple software programs or software modules (e.g., used to provide distributed services), or as a single software program or software module; no specific limitations are made here.

[0065] In this embodiment, terminal 101 can monitor the language type of online conference audio data in the target conference; when it is determined that the language type of the first real-time audio information in the online conference audio data is the target language type, it generates the first text corresponding to the first real-time audio information and displays the first text as subtitles on the conference interface; when it is determined that the language type of the second real-time audio information in the online conference audio data is the language type to be translated, it performs target language type text conversion processing on the second real-time audio information to obtain the second text and displays the second text as subtitles on the conference interface.

[0066] It should be understood that the number of terminals, networks, and servers mentioned above is merely illustrative, and can be any number of terminals, networks, and servers depending on implementation needs. Of course, in one embodiment provided in this application, the system architecture may not include servers.

[0067] Please see Figure 2 , Figure 2 This is a flowchart illustrating a meeting information processing method provided in an embodiment of this application. The executing entity in this embodiment can be a terminal executing the meeting information processing method, a processor within the terminal executing the meeting information processing method, or a meeting information processing service within the terminal executing the meeting information processing method. For ease of description, the following uses the processor within the terminal as an example to describe the specific execution process of the meeting information processing method.

[0068] like Figure 2 As shown, the meeting information processing methods include:

[0069] S202: Monitor the language type of online meeting audio data in the target meeting.

[0070] The target meetings include both audio and video conferences, each with its own displayable interface. Once the target meeting begins, the language type of the online audio data is monitored in real time. For example, speech recognition technology can be used to analyze the audio signal and determine the language type of the online meeting audio data, or a language type recognition model can be used to identify the language type of the online meeting audio data in the target meeting.

[0071] It should be noted that the language type here includes both the target language type and the language to be translated. The target language type is the preset language type set in the language translation configuration information on the user's device. The target language type can be Chinese, Tibetan, English, Japanese, Korean, German, etc. The language to be translated can be a language other than the target language type. For example, if the target language type is Chinese, the language to be translated can be English, Japanese, Korean, German, etc.

[0072] After determining the language type of the online meeting audio data in the target meeting, the processor executes the corresponding subtitle display strategy based on the language type recognition result. For example, if the target language is Chinese, but the detected online meeting audio data is in English, the processor will activate the translation module to translate the English into Chinese in real time for real-time subtitle display, ensuring that users can understand the meeting content.

[0073] S204: When the language type of the first real-time voice information in the online meeting voice data is determined to be the target language type, the first text corresponding to the first real-time voice information is generated, and the first text is displayed as a subtitle on the meeting interface.

[0074] The online conference audio data refers to the conference audio data received by the current user's terminal from other user terminals or the server. Conversely, the conference audio received by the current user's terminal locally through an audio receiving device is the local conference audio data. Therefore, when displaying subtitles for the first text in the conference interface, subtitles are used to display the online conference audio data sent by other user terminals or the server. This avoids interference with the display of the first text when displaying subtitles for offline conference audio, and also facilitates the user's quick understanding of the online conference audio data sent by other user terminals or the server based on the first text.

[0075] When the language type of the first real-time voice information in the online meeting audio data is detected to be the target language type, real-time speech-to-text processing is performed on the first real-time voice information to generate the corresponding first text, with the language type of the first text being the target language type. During the real-time speech-to-text processing of the first real-time voice information, the semantics of all previously identified first texts can be combined to improve the accuracy of the first text recognition.

[0076] Next, the first text is displayed as subtitles on the meeting interface corresponding to the target meeting. The meeting interface has a pre-defined subtitle display area, which can be located near the bottom of the interface. This area is used to display subtitles of the real-time recognized first text. It should be understood that to prevent some text from being unreadable due to exceeding the subtitle display area, a maximum number of characters can be set for the first subtitle display area. When the number of characters displayed in the first subtitle display area exceeds the maximum, the displayed text will be cleared to display the real-time recognized text. Here, the maximum number of characters includes punctuation marks.

[0077] The first real-time audio information can be audio information within a preset time period corresponding to the current time point, such as 0.3 seconds, 0.5 seconds, 1 second, etc. When the language type of the first real-time audio information in the online meeting audio data is determined to be the target language type, text recognition is performed on the first real-time audio information within the first preset time period to obtain the corresponding first text. Then, the first text is displayed as subtitles in the first subtitle display area. Next, text recognition is performed on the first real-time audio information within the next first preset time period to obtain the first text for that time period.

[0078] When it is detected that the number of words displayed in the first subtitle display area does not exceed the maximum number of subtitle display words, in the first subtitle display area, the current first text is displayed after the previously displayed first text; when it is detected that the number of words displayed in the first subtitle display area exceeds the maximum number of subtitle display words, the text displayed in the first subtitle display area is cleared to display the current first text.

[0079] Before performing subtitle display on the first text in the meeting interface, to improve the readability and accuracy of subtitle display, embodiments of this application can perform natural language processing on the recognized first text to determine the sentence boundaries in the text. By analyzing the grammatical structure and semantic content of all the recognized first text, the start and end positions of sentences are further recognized, and punctuation marks such as full stops and commas are inserted at appropriate positions.

[0080] In addition, to adapt to the punctuation habits of different languages, this method can also adjust the use of punctuation marks according to the target language type. For example, in Chinese, a semicolon can be used to separate juxtaposed words, while in English, a comma is used. Through this method of intelligent sentence segmentation and punctuation addition, subtitle display can be more fluent.

[0081] S206: When it is determined that the language type of the second real-time voice information in the online meeting voice data is the target language type to be translated, perform text conversion processing on the second real-time voice information to obtain a second text, and perform subtitle display on the second text in the meeting interface.

[0082] Among them, when it is monitored that the language type of the second real-time voice information in the online meeting voice data is the target language type to be translated, perform text conversion processing on the second real-time voice information to obtain the second text corresponding to the second real-time voice information.

[0083] Exemplarily, text conversion processing can be performed on the second real-time voice information to obtain the second real-time voice information text; then, text translation processing is performed on the second real-time voice information text based on the target language type to obtain the second text. Here, the text translation processing can be specifically implemented based on speech-to-text technology.

[0084] The second real-time voice information can be the voice information within a preset time period corresponding to the current time point. For example, the preset time period can be set to 0.3 seconds, 0.5 seconds, 1 second, etc. Therefore, when it is determined that the language type of the second real-time voice information in the online meeting voice data is the target language type to be translated, text recognition can be performed on the second real-time voice information within the second preset time period to obtain the corresponding second text.

[0085] The process of converting the second real-time speech information into text is actually a process of speech-to-text recognition of the second real-time speech information. Therefore, the language of the second real-time speech information text is the same as that of the second real-time speech information. Then, the second real-time speech information text is translated into a second text in the target language.

[0086] When the language type of the second real-time voice information in the online meeting voice data is determined to be the language type to be translated, text recognition is performed on the second real-time voice information within the second preset time period to obtain the corresponding second text. Then, the second text is displayed as subtitles in the first subtitle display area. Then, text recognition is performed on the second real-time voice information within the next second preset time period to obtain the second text for that time period.

[0087] When it is detected that the number of characters displayed in the first subtitle display area does not exceed the maximum number of characters displayed in the subtitle display area, the current second text is displayed after the previously displayed second text in the first subtitle display area; when it is detected that the number of characters displayed in the first subtitle display area exceeds the maximum number of characters displayed in the subtitle display area, the text displayed in the first subtitle display area is cleared in order to display the current second text.

[0088] Before displaying subtitles on the second text in the meeting interface, to improve the readability and accuracy of the subtitle display, this embodiment of the application can perform natural language processing on the identified second text to determine the sentence boundaries in the text. By analyzing the grammatical structure and semantic content of all identified second text, the start and end positions of sentences can be identified, thereby inserting punctuation marks such as periods and commas at appropriate positions.

[0089] This application provides a meeting information processing method. By monitoring the language type of online meeting audio data in a target meeting, when the language type of the first real-time audio information is the target language type, the corresponding first text is displayed as subtitles on the meeting interface. This allows for the visualization of meeting communication information in the target meeting, thereby improving meeting communication efficiency and avoiding reliance on the speaker's online voice to understand the meeting content. Furthermore, when the language type of the second real-time audio information in the online meeting audio data is the language to be translated, the second real-time audio information is automatically converted to the target language type text to obtain the second text, which is then displayed as subtitles on the meeting interface. This avoids the need for professional translation assistance when participants speak different languages, shortening the meeting time and again making the meeting communication information in the target meeting visible, effectively improving meeting communication efficiency. This solves the technical problem of poor visibility of information transmission during online meetings.

[0090] Please see Figure 3 , Figure 3 This is a flowchart illustrating a language type monitoring method provided in an embodiment of this application. Figure 3 As shown in S202, the language types of the online meeting audio data in the target meeting being monitored include:

[0091] S302: Determine the language translation configuration information of the user terminal for the target meeting, and based on the language translation configuration information, monitor the language type of the current voice information corresponding to the online meeting voice data.

[0092] In the online conferencing software on the user's end, corresponding language translation configuration information can be set for different meetings. This language translation configuration information is used to set the language type of the subtitles displayed on the corresponding meeting interface on the user's end. The preset language types in the language translation configuration information can be flexibly set in advance based on user needs.

[0093] After obtaining the language translation configuration information for the target meeting from the user's end, the language translation configuration information is used to monitor the language type of the current voice information corresponding to the online meeting voice data in real time, and then the corresponding subtitle display strategy is determined based on the monitoring results.

[0094] S304: When the language type of the current voice information successfully matches the preset language type in the language translation configuration information, the current voice information is used as the first real-time voice information, and the language type of the first real-time voice information is determined as the target language type.

[0095] After determining the language type of the current voice information, this language type is matched against the preset language types in the language translation configuration information. When a successful match is found between the language type and the preset language types in the language translation configuration information, it indicates that the language type is the target language type, and the current voice information is used as the first real-time voice information.

[0096] The current voice information can be the voice information within a preset time period corresponding to the current time point. The preset time period can be adjusted based on the pronunciation attributes and grammatical complexity coefficient of the target language type. The current audio recognition difficulty value is calculated by weighted summation of the pronunciation attribute value and the grammatical complexity coefficient. The pronunciation attribute value is used to characterize the length of pronunciation in common languages ​​under the corresponding language type, and the grammatical complexity coefficient is used to characterize the grammatical complexity of common sentences under the corresponding language type.

[0097] For example, the weights of the pronunciation attribute value and the grammatical complexity coefficient can each be set to 0.5. Of course, the weights of the pronunciation attribute value and the grammatical complexity coefficient can also be flexibly set based on the recognition characteristics of the speech-to-text conversion program currently in use. For instance, if the conversion program is accurate in pronunciation recognition, the weight of the pronunciation attribute value can be lowered; if the conversion program is accurate in grammar recognition, the weight of the grammatical complexity coefficient can be lowered.

[0098] In one feasible implementation, when the current audio recognition difficulty value exceeds the preset difficulty range, the duration of the preset time period can be extended; when the current audio recognition difficulty value is less than each value in the preset difficulty range, the duration of the preset time period can be shortened; when the current audio recognition difficulty value is within the preset difficulty range, the duration of the preset time period can remain unchanged.

[0099] S306: When the language type of the current voice information fails to match the preset language type in the language translation configuration information, the current voice information is used as the second real-time voice information, and the language type of the second real-time voice information is determined as the language type to be translated.

[0100] After determining the language type of the current voice information, the language type is matched with the preset language type in the language translation configuration information. When the language type fails to match the preset language type in the language translation configuration information, it indicates that the language type is the language type to be translated, and the current voice information is used as the second real-time voice information.

[0101] The current voice information can be the voice information within a preset time period corresponding to the current time point. The duration of the preset time period can be adjusted based on the pronunciation attributes and grammatical complexity coefficient of the language type of the current voice information. Here, the current audio recognition difficulty value can also be calculated by weighted summation of the pronunciation attribute value and the grammatical complexity coefficient.

[0102] Similarly, when the current audio recognition difficulty value exceeds the preset difficulty range, the duration of the preset time period can be extended; when the current audio recognition difficulty value is less than any value in the preset difficulty range, the duration of the preset time period can be shortened; when the current audio recognition difficulty value is within the preset difficulty range, the duration of the preset time period can remain unchanged.

[0103] The embodiments provided in this application monitor the language type of the current audio information corresponding to the audio data of online meetings through language translation configuration information, thereby enabling users to flexibly adjust the language translation configuration information based on actual subtitle language display needs, thereby improving meeting efficiency.

[0104] Please see Figure 4 , Figure 4This is a schematic diagram illustrating a process for generating meeting content text information, provided as an embodiment of this application. For example... Figure 4 As shown, the method includes:

[0105] S402: Store the local conference audio data in the target conference, and when the target conference ends, generate the target language text corresponding to the local conference audio data.

[0106] During the target meeting, local meeting audio data is continuously acquired and stored. When the target meeting is detected to have ended, all acquired local meeting audio data is processed into text in the target language to obtain the corresponding text. It should be noted that the local meeting audio data carries the recording time period of the corresponding audio, i.e., third-order time information.

[0107] S404: Determine the first timing information of the first text corresponding to the first real-time voice information, the second timing information of the second text corresponding to the second real-time voice information, and the third timing information of the target language type text corresponding to the offline meeting voice.

[0108] In the target meeting, there are generally multiple first texts, each corresponding to a first real-time audio message. Each first real-time audio message carries its corresponding recording time period information, such as 13:30:11 to 13:30:11.3. Therefore, the recording time period information carried by the first real-time audio message can be used as the first time sequence information of the first real-time audio message, the recording time period information carried by the second real-time audio message can be used as the second time sequence information of the second real-time audio message, and the recording time period information carried by the third real-time audio message can be used as the third time sequence information of the third real-time audio message.

[0109] S406: Based on the first time sequence information, the second time sequence information, and the third time sequence information, perform text concatenation processing on the first text, the second text, and the target language type text in chronological order to obtain the meeting content text information corresponding to the target meeting.

[0110] In this system, the first text corresponds to the first temporal information, the second text corresponds to the second temporal information, and the target language text corresponds to the third temporal information. Therefore, the first, second, and third temporal information can be used to sort the first, second, and target language texts. Natural language processing is then performed on the sorted texts to determine sentence boundaries. By analyzing the grammatical structure and semantic content of the identified texts, the start and end positions of sentences are identified, and punctuation marks such as periods and commas are inserted at appropriate positions to complete the text splicing process and obtain the meeting content text information corresponding to the target meeting.

[0111] In one feasible implementation, after obtaining the meeting content text information corresponding to the target meeting, the meeting content text information can be input into the meeting minutes generation model. Based on the meeting minutes generation model, the meeting content text information is processed to generate meeting minutes, and the meeting minutes corresponding to the meeting content text information are obtained and output. The meeting minutes generation model is obtained by training the basic model through a sample training set. The sample training set includes sample meeting content text information and sample meeting minutes corresponding to the sample meeting content text information.

[0112] For example, when training a base model, sample meeting minutes text information can be input into the base model. The base model then processes this information to generate meeting minutes, and outputs reference meeting minutes. The model loss function is then constructed using the reference and sample meeting minutes. For instance, in each training round, the reference score for the reference meeting minutes and the sample meeting minutes for the sample meeting minutes can be determined. The parameters corresponding to these scores are then used to construct a model loss function for the base model. The model loss value is then determined based on these scores, and the model parameters are adjusted using this loss value. Finally, the model training is completed, resulting in a large-scale meeting minutes generation model.

[0113] In the embodiments provided in this application, the first text, the second text, and the target language text are concatenated in chronological order using first time-series information, second time-series information, and third time-series information to obtain the meeting content text information. This effectively utilizes the subtitles displayed during the meeting to generate the meeting content text information, avoiding repeated text recognition of the speech during the meeting. Furthermore, in this embodiment, after the target meeting concludes, corresponding meeting minutes can be automatically generated using the meeting content text information, enabling seamless recording of the target meeting and improving the user experience.

[0114] Please see Figure 5 , Figure 5 This is a schematic diagram illustrating a process for generating target meeting recording data, provided as an embodiment of this application. Figure 5 As shown, the method includes:

[0115] S502: Record the target meeting based on the meeting recording command.

[0116] During the meeting recording process, the original meeting interface and corresponding audio of the target meeting are recorded, and the generated subtitle data is saved. The original meeting interface does not include subtitles; the subtitle data includes the data corresponding to the first and second texts. It should be noted that the target meeting interface is obtained by synthesizing the original meeting interface and the subtitle data.

[0117] In one feasible implementation, the meeting data corresponding to the target meeting received by the user terminal can be saved to record the original meeting interface and corresponding audio in the target meeting. The meeting data includes video data and audio data. After the meeting recording is received, the video data is used as video recording data and the audio data is used as audio recording data.

[0118] In another feasible implementation, the original meeting interface and corresponding audio in the target meeting can be recorded directly. At the same time, the clarity of the video and audio recordings can be selected during the recording process.

[0119] S504: Determine when the target meeting ends and terminate the meeting recording and generate meeting recording data; wherein, the meeting recording data includes video recording data, audio recording data and subtitle data.

[0120] Specifically, when the target meeting is detected to be over, the meeting recording is terminated, and meeting recording data is generated based on the data acquired during the meeting recording process. The meeting recording data includes video recording data, audio recording data, and subtitle data.

[0121] The video recording data and audio recording data are interconnected, which makes the original meeting interface synchronized with the user's voice. However, due to the speech-to-text and cross-language translation processes involved in the generation of subtitle data, the display time of the subtitle data often lags behind the user's voice, that is, there is a delay between the display of subtitles and the corresponding user voice.

[0122] S506: Perform audio subtitle delay elimination processing on the subtitle data based on the audio recording data to obtain the target subtitle data, and determine the target meeting recording data based on the video recording data, audio recording data, and target subtitle data.

[0123] To eliminate the delay between subtitle display and corresponding user speech, audio recording data can be used to perform audio-subtitle delay elimination processing on the subtitle data. The subtitle data typically records the display time period for each subtitle. To ensure that the subtitle display time period and the corresponding user speech correspond, the time periods of each speech segment in the audio recording data can be used to synchronize the display time periods of the corresponding subtitles in the subtitle data to eliminate the delay, thus obtaining the target subtitle data. Subsequently, the video recording data, audio recording data, and target subtitle data are correlated to determine the target meeting recording data.

[0124] The embodiments provided in this application utilize meeting recording data to perform corresponding audio subtitle delay elimination processing to obtain target meeting recording data. This enables synchronization between the original meeting interface, user voice, and subtitles during the playback of the target meeting recording data. Furthermore, even if the participants in the meeting communicate in the language to be translated, synchronization between the original meeting interface, user voice, and subtitles can still be achieved during subsequent playback.

[0125] Please see Figure 6 , Figure 6 This is a schematic diagram illustrating a process for determining target subtitle data, provided as an embodiment of this application. Figure 6 As shown, in S506, audio subtitle delay elimination processing is performed on the subtitle data based on the audio recording data to obtain the target subtitle data, including:

[0126] S602: Determine the first timing information corresponding to the first real-time voice information in the audio recording data, search for the first text corresponding to the first real-time voice information in the subtitle data, and determine the first text timing information corresponding to the first text in the subtitle data.

[0127] The audio recording data can contain multiple first real-time voice information pieces, each with corresponding first time sequence information. For a description of the first time sequence information, please refer to the relevant description in S404, which will not be repeated here.

[0128] Since the first real-time voice information corresponds to the first text, when generating the first text based on the first real-time voice information, the first text can be marked based on the identification features corresponding to the first real-time voice information. Each piece of first real-time voice information has its own corresponding identification features, and different pieces of first real-time voice information have different identification features.

[0129] Therefore, the first text corresponding to the first real-time voice information can be found from the subtitle data based on the identification features corresponding to the first real-time voice information. The first text will be displayed as subtitles in the target conference, and the subtitle display time period corresponding to the first text is the time information of the first text.

[0130] S604: Determine the second timing information corresponding to the second real-time speech information in the audio recording data, find the second text corresponding to the second real-time speech information in the subtitle data, and determine the second text timing information corresponding to the second text in the subtitle data.

[0131] The audio recording data may contain multiple second real-time voice information, each of which has corresponding second time sequence information. For a description of the second time sequence information, please refer to the relevant description in S404, which will not be repeated here.

[0132] Since the second real-time voice information corresponds to the second text, when generating the second text based on the second real-time voice information, the second text can be marked based on the identification features corresponding to the second real-time voice information. Each second real-time voice information has its own corresponding identification features, and different second real-time voice information have different identification features.

[0133] Therefore, the second text corresponding to the second real-time voice information can be found from the subtitle data based on the identification features corresponding to the second real-time voice information. The second text will be displayed as subtitles in the target meeting, and the subtitle display time period corresponding to the second text is the time information of the second text.

[0134] S606: Update the first text time information based on the first time sequence information to obtain the target first text time information.

[0135] Since the first real-time voice information and the first text correspond to each other, and there is a delay between the first time sequence information and the first text time information, the first time sequence information can be used to update the first text time information. This makes the start time in the target first text time information the same as the start time in the first time sequence information, such as 13:11:11.

[0136] Since the end time in the first text time information depends on the time when the number of characters displayed in its corresponding first subtitle display area exceeds the maximum number of characters displayed in the subtitle, when the number of characters corresponding to the subtitle in the conference interface reaches the maximum, the first or second text subtitles corresponding to the subtitles displayed in this batch have the same end time. Subtitles with the same end time are considered to be in the same batch.

[0137] Therefore, it is possible to obtain the subtitles in the same batch corresponding to the first text, query the subtitles in the next batch corresponding to the same batch, determine the first target start time of the first real-time voice information or the second real-time voice information corresponding to the first subtitle in the next batch, and use the first target start time as the end time in the target first text time information.

[0138] S608: Update the second text time information based on the second time sequence information to obtain the target second text time information.

[0139] Since the second real-time voice information and the second text correspond to each other, and there is a delay between the second time sequence information and the second text time information, the second time sequence information can be used to update the second text time information. This makes the start time in the target second text time information the same as the start time in the first time sequence information, such as 13:11:11.

[0140] Since the end time in the second text's time information depends on the time when the number of characters displayed in its corresponding first subtitle display area exceeds the maximum number of characters displayed in the subtitle, when the number of characters corresponding to the subtitle in the conference interface reaches the maximum, the first or second text corresponding to the subtitles displayed in this batch have the same end time. Subtitles with the same end time are considered to be in the same batch.

[0141] Therefore, it is possible to obtain the subtitles in the same batch corresponding to the second text, query the subtitles in the next batch corresponding to the same batch, determine the second target start time of the first real-time voice information or the second real-time voice information corresponding to the first subtitle in the next batch, and use the second target start time as the end time in the target second text time information.

[0142] S610: Adjust the display time of the first and second texts in the subtitle data using the target first text time information and the target second text time information respectively to obtain the target subtitle data.

[0143] Each first text corresponds to a first text time information, and each first text time information corresponds to a target first text time information. Therefore, the display time of the first text corresponding to the target first text can be adjusted using the target first text time information, and the display time of the second text corresponding to the target second text can be adjusted using the target second text time information, thereby obtaining the target subtitle data.

[0144] In the embodiments provided in this application, the display time of the first text and the second text in the subtitle data are adjusted using the target first text time information and the target second text time information, respectively, so that the time period of each voice in the audio recording data is synchronized with the display time period of the corresponding subtitle in the target subtitle data, thereby effectively avoiding the phenomenon of subtitles and audio being out of sync.

[0145] Please see Figure 7 , Figure 7 This is a schematic diagram illustrating another process for determining target subtitle data provided in an embodiment of this application. For example... Figure 7 As shown, after obtaining the target subtitle data in S610, the method further includes:

[0146] S702: Obtain the local conference voice data corresponding to the user terminal, generate the target language type text corresponding to the local conference voice data, and determine the third time sequence information corresponding to the local conference voice data.

[0147] In order to display subtitles corresponding to the offline meeting audio during subsequent playback, the corresponding offline meeting audio can be obtained from the user's terminal, the offline meeting audio can be processed into language-to-text, and based on the language type of the offline meeting audio, it can be determined whether to translate the converted text corresponding to the offline meeting audio into the corresponding target language type text, thereby obtaining the target language type text corresponding to the offline meeting audio.

[0148] The third temporal information corresponding to the local conference audio data can be one or more, and each third temporal information corresponds to a word or character in the local conference audio data, that is, each third temporal information can be the audio pronunciation time of the corresponding word or character. Similarly, each character in the target language text carries the identification features corresponding to the pronunciation of the corresponding word or character, thereby associating the character with its corresponding word or character pronunciation.

[0149] S704: Based on the third time sequence information, perform subtitle generation processing on the target language type text to obtain local subtitle data, and add the local subtitle data to the target meeting recording data to update the target meeting recording data.

[0150] The subtitles corresponding to the target subtitle data are displayed in the first subtitle display area of ​​the conference interface, while the subtitles corresponding to the local subtitle data are displayed in the second subtitle display area of ​​the conference interface.

[0151] The first and second subtitle display areas can be set adjacent to each other without overlapping, thereby avoiding mutual interference when the first and second subtitle display areas display subtitles respectively. This allows for effective differentiation between subtitles corresponding to local meeting users and subtitles corresponding to online meeting users during subsequent playback, and ensures that the insertion of local subtitle data does not interfere with the original display method of the prior target subtitle data.

[0152] Similarly, when it is detected that the number of characters displayed in the second subtitle display area does not exceed the maximum number of characters displayed in the subtitle display area, the text corresponding to the current audio will continue to be displayed in the second subtitle display area after the text that was previously displayed; when it is detected that the number of characters displayed in the second subtitle display area exceeds the maximum number of characters displayed in the subtitle display area, the text displayed in the second subtitle display area will be cleared in order to display the text corresponding to the current audio.

[0153] During the subtitle generation process for the target language text, the maximum number of characters displayed in the second subtitle display area is obtained. Based on the maximum number of characters displayed in the second subtitle display area, the target language text is divided into batches. Each word or character in the same batch of text has the same subtitle display end time. The subtitle display end time of each word or character in the same batch of text can be used as the subtitle display start time of the next batch of text. The subtitle display start time of each word or character in the same batch of text is the audio start time corresponding to that word or character, which is the start time of the corresponding third timing information.

[0154] In the embodiments provided in this application, the subtitles corresponding to the target subtitle data are displayed based on the first subtitle display area, and the subtitles corresponding to the local subtitle data are displayed based on the second subtitle display area. This enables the effective display of subtitles corresponding to local meeting users and online meeting users during subsequent playback, and ensures that the insertion of local subtitle data does not interfere with the original display method of the prior target subtitle data.

[0155] Please see Figure 8 , Figure 8 This is a flowchart illustrating a process for generating a historical meeting interface for a target meeting, as provided in an embodiment of this application. Figure 8 As shown, after S506 determines the target meeting recording data based on video recording data, audio recording data, and target caption data, the method further includes:

[0156] S802: Store the target meeting recording data and determine the storage path information of the target meeting recording data.

[0157] One approach is to store the target meeting recording data in the cloud and then obtain the cloud storage path information corresponding to the target meeting recording data; another approach is to store the target meeting recording data in local storage and then obtain the local storage path information corresponding to the target meeting recording data.

[0158] S804: Determine the historical meeting information corresponding to the target meeting, and generate a historical meeting interface for the target meeting using the historical meeting information and storage path information.

[0159] The historical meeting information corresponding to the target meeting includes, but is not limited to, the meeting date, time, list of participants, meeting topic, and meeting minutes. Then, based on the historical meeting information and storage path information, a historical meeting interface for the target meeting is generated to facilitate subsequent users in accessing historical meeting information and viewing meeting replay videos.

[0160] In the embodiments provided in this application, a historical meeting interface for the target meeting is generated using historical meeting information and storage path information, thereby facilitating subsequent users to view historical meeting information and watch meeting playback videos.

[0161] Please see Figure 9 , Figure 9 This is a schematic diagram of a conference information processing device provided in an embodiment of this application. Figure 9 As shown, the conference information processing device 900 includes:

[0162] Monitoring module 910 is suitable for monitoring the language type of online meeting audio data in the target meeting;

[0163] The first determining module 920 is adapted to generate the first text corresponding to the first real-time voice information when determining that the language type of the first real-time voice information in the online meeting voice data is the target language type, and to display the first text as subtitles on the meeting interface.

[0164] The second determining module 930 is adapted to determine that when the language type of the second real-time voice information in the online meeting voice data is the language type to be translated, perform target language type text conversion processing on the second real-time voice information to obtain the second text, and display the second text as subtitles on the meeting interface.

[0165] Optionally, the monitoring module 910 includes:

[0166] The unit is identified, and the language translation configuration information of the user terminal for the target meeting is determined. Based on the language translation configuration information, the language type of the current voice information corresponding to the online meeting voice data is monitored.

[0167] The first matching unit is adapted to use the current voice information as the first real-time voice information and determine the language type of the first real-time voice information as the target language type when the language type of the current voice information successfully matches the preset language type in the language translation configuration information.

[0168] The second matching unit is adapted to use the current voice information as the second real-time voice information and determine the language type of the second real-time voice information as the language type to be translated when the language type of the current voice information fails to match the preset language type in the language translation configuration information.

[0169] Optionally, the conference information processing device 900 also includes:

[0170] The generation module is suitable for storing local conference audio data in the target conference, and generating text in the target language type corresponding to the local conference audio data when the target conference is detected to be over.

[0171] The determining module is adapted to determine the first time sequence information of the first text corresponding to the first real-time voice information, the second time sequence information of the second text corresponding to the second real-time voice information, and the third time sequence information of the target language type text corresponding to the offline conference voice.

[0172] The splicing module is suitable for splicing the first text, the second text, and the target language type text in chronological order based on the first time sequence information, the second time sequence information, and the third time sequence information to obtain the meeting content text information corresponding to the target meeting.

[0173] Optionally, the conference information processing device 900 also includes:

[0174] The meeting minutes generation module is suitable for inputting the meeting content text information into the meeting minutes generation model, processing the meeting content text information based on the meeting minutes generation model, and obtaining and outputting the meeting minutes corresponding to the meeting content text information.

[0175] Among them, the large model for generating meeting minutes is obtained by training the basic large model through a sample training set, which includes sample meeting content text information and sample meeting minutes corresponding to the sample meeting content text information.

[0176] Optionally, the conference information processing device 900 also includes:

[0177] The recording module is suitable for recording target meetings based on meeting recording commands;

[0178] The meeting recording data generation module is suitable for determining when the target meeting ends, ending the meeting recording, and generating meeting recording data; wherein, the meeting recording data includes video recording data, audio recording data, and subtitle data;

[0179] The delay elimination processing module is suitable for performing audio subtitle delay elimination processing on subtitle data based on audio recording data to obtain target subtitle data, and determining target meeting recording data based on video recording data, audio recording data, and target subtitle data.

[0180] Optionally, the delay elimination processing module includes:

[0181] The first text timing information determining unit is adapted to determine the first timing information corresponding to the first real-time speech information in the audio recording data, search for the first text corresponding to the first real-time speech information in the subtitle data, and determine the first text timing information corresponding to the first text in the subtitle data.

[0182] The second text timing information determination unit is adapted to determine the second timing information corresponding to the second real-time speech information in the audio recording data, search for the second text corresponding to the second real-time speech information in the subtitle data, and determine the second text timing information corresponding to the second text in the subtitle data.

[0183] The first update unit is adapted to update the first text time information based on the first time sequence information to obtain the target first text time information;

[0184] The second update unit is adapted to update the second text time information based on the second time sequence information to obtain the target second text time information;

[0185] The display time adjustment unit is adapted to adjust the display time of the first text and the second text in the subtitle data by using the target first text time information and the target second text time information respectively, so as to obtain the target subtitle data.

[0186] Optionally, the conference information processing device 900 also includes:

[0187] The third timing information determination module is suitable for acquiring local conference voice data corresponding to the user terminal, generating target language type text corresponding to the local conference voice data, and determining the third timing information corresponding to the local conference voice data.

[0188] The subtitle generation module is suitable for generating subtitles for target language text based on third-order time information, obtaining local subtitle data, and adding the local subtitle data to the target meeting recording data to update the target meeting recording data.

[0189] The subtitles corresponding to the target subtitle data are displayed in the first subtitle display area of ​​the conference interface, while the subtitles corresponding to the local subtitle data are displayed in the second subtitle display area of ​​the conference interface.

[0190] Optionally, the conference information processing device 900 also includes:

[0191] The storage module is suitable for storing the target meeting recording data and determines the storage path information of the target meeting recording data;

[0192] The historical meeting interface generation module is suitable for determining the historical meeting information corresponding to the target meeting, and generating a historical meeting interface for the target meeting using the historical meeting information and storage path information.

[0193] Optionally, the second determining module 930 includes:

[0194] The text conversion unit is adapted to perform text conversion processing on the second real-time speech information to obtain the second real-time speech information text.

[0195] The text translation unit is suitable for performing text translation processing on the second real-time speech information text based on the target language type to obtain the second text.

[0196] Please see Figure 10 , Figure 10 This is a schematic diagram of the structure of a terminal provided in an embodiment of this application. Figure 10 As shown, terminal 1000 may include: at least one processor 1001, at least one network interface 1004, user interface 1003, memory 1005, and at least one communication bus 1002.

[0197] The communication bus 1002 is used to realize the connection and communication between these components.

[0198] The user interface 1003 may include a display screen and a camera. Optionally, the user interface 1003 may also include a standard wired interface and a wireless interface.

[0199] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0200] The processor 1001 may include one or more processing cores. The processor 1001 connects to various parts within the terminal 1000 using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 1005, and by calling data stored in the memory 1005. Optionally, the processor 1001 may be implemented using at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), or PLA (Programmable Logic Array). The processor 1001 may integrate one or more of the following: CPU (Central Processing Unit), GPU (Graphics Processing Unit), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content to be displayed on the screen; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 1001 and may be implemented as a separate chip.

[0201] The memory 1005 may include RAM (Random Access Memory) or ROM (Read-Only Memory). Optionally, the memory 1005 may include a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 1005 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 1005 may also be at least one storage device located remotely from the aforementioned processor 1001. Figure 10 As shown, the memory 1005, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a conference information processing program.

[0202] exist Figure 10 In the terminal 1000 shown, the user interface 1003 is mainly used to provide an input interface for the user and to obtain the user's input data; while the processor 1001 can be used to call the conference information processing program stored in the memory 1005 and specifically perform the following operations:

[0203] Monitor the language type of online meeting audio data in the target meeting;

[0204] When the language type of the first real-time voice information in the online meeting voice data is determined to be the target language type, the first text corresponding to the first real-time voice information is generated, and the first text is displayed as a subtitle on the meeting interface;

[0205] When the language type of the second real-time voice information in the online meeting audio data is determined to be the language type to be translated, the second real-time voice information is processed by target language type text conversion to obtain the second text, and the second text is displayed as subtitles on the meeting interface.

[0206] Optionally, when the processor 1001 executes the language type of the online meeting audio data in the target meeting, it specifically performs the following:

[0207] Determine the language translation configuration information for the target meeting on the user end, and based on the language translation configuration information, monitor the language type of the current voice information corresponding to the online meeting voice data;

[0208] When the language type of the current voice information successfully matches the preset language type in the language translation configuration information, the current voice information is used as the first real-time voice information, and the language type of the first real-time voice information is determined as the target language type.

[0209] When the language type of the current voice information fails to match the preset language type in the language translation configuration information, the current voice information is used as the second real-time voice information, and the language type of the second real-time voice information is determined as the language type to be translated.

[0210] Alternatively, the processor 1001 is also adapted to perform:

[0211] Store the local conference audio data in the target conference, and when the target conference ends, generate the target language text corresponding to the local conference audio data;

[0212] Determine the first time sequence information corresponding to the first real-time voice information of the first text, the second time sequence information corresponding to the second real-time voice information of the second text, and the third time sequence information corresponding to the offline meeting voice of the target language type text;

[0213] Based on the first time-series information, the second time-series information, and the third time-series information, the first text, the second text, and the target language type text are concatenated in chronological order to obtain the meeting content text information corresponding to the target meeting.

[0214] Optionally, after obtaining the meeting content text information corresponding to the target meeting, the processor 1001 further executes the following:

[0215] Input the meeting content text information into the meeting minutes generation model, process the meeting content text information to generate meeting minutes based on the meeting minutes generation model, and obtain and output the meeting minutes corresponding to the meeting content text information.

[0216] Among them, the large model for generating meeting minutes is obtained by training the basic large model through a sample training set, which includes sample meeting content text information and sample meeting minutes corresponding to the sample meeting content text information.

[0217] Alternatively, the processor 1001 is also adapted to perform:

[0218] Record the target meeting based on the meeting recording command;

[0219] The meeting recording ends and the meeting recording data is generated when the target meeting ends; the meeting recording data includes video recording data, audio recording data, and subtitle data;

[0220] The audio subtitle data is processed by eliminating audio subtitle delay based on the audio recording data to obtain the target subtitle data. The target meeting recording data is then determined based on the video recording data, audio recording data, and target subtitle data.

[0221] Optionally, when processor 1001 performs audio subtitle delay elimination processing on subtitle data based on audio recording data to obtain target subtitle data, it specifically executes:

[0222] Determine the first timing information corresponding to the first real-time speech information in the audio recording data, find the first text corresponding to the first real-time speech information in the subtitle data, and determine the first text timing information corresponding to the first text in the subtitle data.

[0223] Determine the second time sequence information corresponding to the second real-time speech information in the audio recording data, find the second text corresponding to the second real-time speech information in the subtitle data, and determine the second text time information corresponding to the second text in the subtitle data.

[0224] The first text time information is updated based on the first time series information to obtain the target first text time information; the second text time information is updated based on the second time series information to obtain the target second text time information.

[0225] The display time of the first and second texts in the subtitle data is adjusted using the target first text time information and the target second text time information, respectively, to obtain the target subtitle data.

[0226] Optionally, after obtaining the target subtitle data, the processor 1001 further executes the following:

[0227] Acquire the local conference audio data corresponding to the user terminal, generate text in the target language type corresponding to the local conference audio data, and determine the third time sequence information corresponding to the local conference audio data;

[0228] Based on the third time sequence information, subtitles are generated for the target language text to obtain local subtitle data. The local subtitle data is then added to the target meeting recording data to update the target meeting recording data.

[0229] The subtitles corresponding to the target subtitle data are displayed in the first subtitle display area of ​​the conference interface, while the subtitles corresponding to the local subtitle data are displayed in the second subtitle display area of ​​the conference interface.

[0230] Optionally, after determining the target meeting recording data based on the video recording data, audio recording data, and target caption data, the processor 1001 further executes the following:

[0231] Store the target meeting recording data and determine the storage path information for the target meeting recording data;

[0232] Determine the historical meeting information corresponding to the target meeting, and use the historical meeting information and storage path information to generate a historical meeting interface for the target meeting.

[0233] Optionally, when the processor 1001 performs target language type text conversion processing on the second real-time speech information to obtain the second text, it specifically performs the following:

[0234] The second real-time voice information is converted into text to obtain the second real-time voice information text.

[0235] The second real-time speech information text is translated based on the target language type to obtain the second text.

[0236] This application also provides a computer-readable storage medium that stores one or more programs, which, when executed by a processor, implement any of the methods described above.

[0237] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0238] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0239] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0240] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application embodiment, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0241] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the embodiments of this application.

[0242] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0243] The above is a description of a conference information processing method, apparatus, terminal, and computer-readable storage medium provided in the embodiments of this application. For those skilled in the art, based on the ideas of the embodiments of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation on the embodiments of this application.

Claims

1. A method for processing meeting information, wherein, The method includes: Monitor the language type of online meeting audio data in the target meeting; When the language type of the first real-time voice information in the online meeting voice data is determined to be the target language type, the first text corresponding to the first real-time voice information is generated, and the first text is displayed as a subtitle on the meeting interface; When the language type of the second real-time voice information in the online meeting voice data is determined to be the language type to be translated, the second real-time voice information is subjected to target language type text conversion processing to obtain the second text, and the second text is displayed as subtitles on the meeting interface.

2. The method according to claim 1, wherein, The language types of the online meeting audio data in the monitored target meetings include: Determine the language translation configuration information of the user terminal for the target meeting, and based on the language translation configuration information, monitor the language type of the current voice information corresponding to the online meeting voice data; When the language type of the current voice information successfully matches the preset language type in the language translation configuration information, the current voice information is used as the first real-time voice information, and the language type of the first real-time voice information is determined as the target language type. When the language type of the current voice information fails to match the preset language type in the language translation configuration information, the current voice information is used as the second real-time voice information, and the language type of the second real-time voice information is determined as the language type to be translated.

3. The method according to claim 1, wherein, The method further includes: The local conference audio data in the target conference is stored, and when the target conference is detected to have ended, the target language type text corresponding to the local conference audio data is generated; The first time sequence information corresponding to the first real-time voice information of the first text, the second time sequence information corresponding to the second real-time voice information of the second text, and the third time sequence information corresponding to the offline meeting voice of the target language type text are determined. Based on the first time sequence information, the second time sequence information, and the third time sequence information, the first text, the second text, and the target language type text are concatenated in chronological order to obtain the meeting content text information corresponding to the target meeting.

4. The method according to claim 3, wherein, After obtaining the meeting content text information corresponding to the target meeting, the method further includes: The meeting content text information is input into the meeting minutes generation model. Based on the meeting minutes generation model, the meeting content text information is processed to generate meeting minutes, and the meeting minutes corresponding to the meeting content text information are obtained and output. The meeting minutes generation model is obtained by training the basic model using a sample training set, which includes sample meeting content text information and sample meeting minutes corresponding to the sample meeting content text information.

5. The method according to claim 1, wherein, The method further includes: Record the target meeting based on the meeting recording command; The meeting recording is terminated and meeting recording data is generated when the target meeting ends; wherein, the meeting recording data includes video recording data, audio recording data, and subtitle data; Based on the audio recording data, the subtitle data is processed to eliminate audio subtitle delay, resulting in target subtitle data. Target meeting recording data is then determined based on the video recording data, the audio recording data, and the target subtitle data.

6. The method according to claim 5, wherein, The step of performing audio subtitle delay elimination processing on the subtitle data based on the audio recording data to obtain target subtitle data includes: Determine the first timing information corresponding to the first real-time speech information in the audio recording data, find the first text corresponding to the first real-time speech information in the subtitle data, and determine the first text timing information corresponding to the first text in the subtitle data; Determine the second time sequence information corresponding to the second real-time voice information in the audio recording data, search for the second text corresponding to the second real-time voice information in the subtitle data, and determine the second text time information corresponding to the second text in the subtitle data; The first text time information is updated based on the first time sequence information to obtain the target first text time information; the second text time information is updated based on the second time sequence information to obtain the target second text time information. The display time of the first and second texts in the subtitle data is adjusted using the target first text time information and the target second text time information, respectively, to obtain the target subtitle data.

7. The method according to claim 6, wherein, After obtaining the target subtitle data, the method further includes: Acquire the local conference voice data corresponding to the user terminal, generate the target language type text corresponding to the local conference voice data, and determine the third time sequence information corresponding to the local conference voice data; Based on the third time sequence information, subtitle generation processing is performed on the target language type text to obtain local subtitle data, and the local subtitle data is added to the target meeting recording data to update the target meeting recording data; The subtitles corresponding to the target subtitle data are displayed in the first subtitle display area of ​​the conference interface, and the subtitles corresponding to the local subtitle data are displayed in the second subtitle display area of ​​the conference interface.

8. A conference information processing device, wherein, The device includes: The monitoring module is suitable for monitoring the language type of online meeting audio data in the target meeting; The first determining module is adapted to generate first text corresponding to the first real-time voice information when determining that the language type of the first real-time voice information in the online meeting voice data is the target language type, and to display the first text as subtitles on the meeting interface. The second determining module is adapted to determine that when the language type of the second real-time voice information in the online meeting voice data is the language type to be translated, perform target language type text conversion processing on the second real-time voice information to obtain the second text, and display the second text as subtitles on the meeting interface.

9. A terminal, wherein, The terminal includes: Processor; and A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, wherein, The computer-readable storage medium stores one or more programs that, when executed by a processor, implement the method of any one of claims 1 to 7.

Images