Audio leakage protection method, device and electronic equipment based on digital watermark

Through the audio leakage protection method based on digital watermarks, the data scanning frequency and watermark recognition technology are used to solve the problem of difficult-to-control the impact of conference audio leakage, and the rapid detection and identification of audio leakage are achieved, which improves the protection effect.

CN117935819BActive Publication Date: 2024-11-05北京富通亚讯网络信息技术有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410146266.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-02-02
Publication Date
2024-11-05
Estimated Expiration
2044-02-02

AI Technical Summary

Technical Problem

The prior art is difficult to effectively reduce the impact and losses when the conference audio is leaked, and the protection effect is poor, especially when the participants record the audio during the meeting.

Method used

The audio leakage protection method based on digital watermark is adopted to determine the topic discussion degree through data scanning frequency, identify sensitive topics, analyze the matching degree between the discussion content and the target organization meeting content, and use digital watermarks to identify and remind the processor to process.

Benefits of technology

It realizes rapid and accurate detection and identification of audio leakage, improves monitoring efficiency and accuracy, promptly reminds processors, reduces losses and risks, and improves the protection effect of conference content.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117935819B_ABST
    Figure CN117935819B_ABST
Patent Text Reader

Abstract

The present application relates to the field of data processing technology, and in particular to an audio leakage protection method, device and electronic device based on digital watermark. The audio leakage protection method based on digital watermark includes: scanning the data of the monitorable platform according to the data scanning frequency to determine the topic discussion degree of the platform topic; determining the sensitive topics in the platform topic according to the topic discussion degree and several discussion degree thresholds of the target organization; performing data analysis on the discussion content of the sensitive topic to determine the matching degree between the discussion content and several meeting contents of the target organization; if the matching degree is less than or equal to the preset matching degree, then watermark recognition is performed on the discussion content according to the digital watermark of the target organization to obtain the watermark recognition result; if the matching degree is greater than the preset matching degree, or the watermark recognition result corresponds to the digital watermark, it is determined that the target organization has an audio leak, so as to remind the processing personnel to handle it, thereby improving the protection effect on the target organization when the audio leaks.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device and electronic device for audio leakage protection based on digital watermark. Background Art

[0002] For enterprises and senior institutions, there are many internal meetings, some of which may involve confidential content such as internal decision-making, development direction, and technology research and development. If the confidential meeting content audio is leaked, it is likely to cause significant losses to the interests of individuals, the enterprise as a whole, or the country. Therefore, confidentiality during the meeting is particularly important.

[0003] At present, in order to ensure the security of confidential meetings, methods such as controlling the participants, setting meeting disciplines, and encrypting the transmission of the entire meeting audio are usually adopted. However, the leakage of conference audio is not only possible during the transmission process after the meeting, but also the possibility of leakage of the conference audio recorded by the participants. When the audio related to the meeting is leaked, the current leakage protection method is difficult to reduce the impact or loss of the leakage, and the protection effect is poor. Summary of the invention

[0004] In order to reduce the impact or loss of conference content leakage and thus improve the protection effect, the present application provides an audio leakage protection method, device and electronic device based on digital watermark.

[0005] In a first aspect, the present application provides an audio leakage protection method based on digital watermark, comprising:

[0006] Scan the data of the monitorable platform according to the data scanning frequency to determine the topic discussion degree of the platform topic;

[0007] Determine sensitive topics among the platform topics based on the discussion degree of the topic and several discussion degree thresholds of the target organization;

[0008] Performing data analysis on the discussion content of the sensitive topic to determine the matching degree between the discussion content and the content of several meetings of the target organization;

[0009] If the matching degree is less than or equal to the preset matching degree, watermark recognition is performed on the discussion content according to the digital watermark of the target organization to obtain a watermark recognition result;

[0010] If the matching degree is greater than the preset matching degree, or the watermark recognition result corresponds to the digital watermark, it is determined that an audio leak has occurred in the target organization, so as to alert the processing personnel to handle it.

[0011] By adopting the above technical solution, the data of the monitorable platform is scanned through the data scanning frequency, so that the topic discussion degree of the platform topic can be quickly and accurately determined, sensitive topics can be discovered in time, and the efficiency and accuracy of monitoring can be improved. According to the topic discussion degree and several discussion degree thresholds of the target organization, the sensitive topics in the platform topic can be determined, so as to carry out targeted analysis and monitoring, thereby improving the targeted monitoring. Data analysis of the discussion content of sensitive topics can determine the matching degree between the discussion content and several meeting contents of the target organization, so as to determine whether there is an audio leak, thereby improving the accuracy of audio leak detection. According to the digital watermark of the target organization, the discussion content is watermarked, and the watermark recognition result can be obtained to further confirm the audio leak, thereby improving the accuracy of audio leak recognition. When the matching degree is less than or equal to the preset matching degree or the watermark recognition result corresponds to the digital watermark, it can be determined that the target organization has an audio leak, and the processing personnel can be reminded in time to handle it, thereby avoiding possible losses and risks. Through technical means such as data scanning, topic discussion analysis, sensitive topic identification, matching judgment and watermark recognition, we have achieved rapid and accurate detection and identification of audio leaks on the monitoring platform, improved the efficiency and accuracy of audio leak monitoring, and enhanced the protection effect when the meeting content of the target organization is leaked.

[0012] Optionally, the method further includes:

[0013] Obtain the data acquisition authority of the target organization, and determine the corresponding monitorable platform based on the data acquisition authority;

[0014] According to the data flow attributes of the monitorable platform and the organizational properties of the target organization, a data scanning frequency is correspondingly set for the monitorable platform.

[0015] By adopting the above technical solution, by obtaining the data acquisition authority of the target organization, determining the corresponding monitorable platform according to the data acquisition authority, and then setting the data scanning frequency for the monitorable platform according to the data flow attributes of the monitorable platform, data monitoring and audio leakage detection of a specific target organization are realized. Specifically, the data acquisition authority of the target organization can be obtained, so that various data sources of the target organization can be accessed. By using the data acquisition authority, various data of the target organization can also be obtained, and analyzed and processed to detect whether there is an audio leakage. It is possible to realize audio leakage detection and monitoring of a specific target organization, and the data scanning frequency and the selection of the monitorable platform can be flexibly adjusted according to the actual situation, thereby achieving better technical effects.

[0016] Optionally, the method further includes:

[0017] Obtain the historical meeting topics and corresponding meeting contents of the target organization;

[0018] A discussion degree threshold of each historical meeting is determined according to the historical meeting topics, the corresponding meeting content and the current network situation.

[0019] By adopting the above technical solution, by analyzing the historical meeting themes and corresponding meeting contents of the target organization, the focus of the target organization's topics, discussion points and decision-making results can be understood, so as to more accurately determine the discussion threshold, which is helpful to timely discover audio leakage and take corresponding treatment measures, and improve the information security protection capability of the target organization. According to the historical meeting themes, the corresponding meeting contents and the current network situation, personalized discussion thresholds can be set for different meeting scenarios, making audio leakage detection more flexible and accurate, and better meeting the actual needs of different target organizations. Through the analysis of historical meeting themes and corresponding meeting contents, the key topics and discussion hotspots of the target organization can be quickly located, so as to carry out audio monitoring and data scanning in a targeted manner, improve monitoring efficiency, and reduce unnecessary calculations and resource consumption. In the process of determining the discussion threshold, a comprehensive analysis can be carried out according to the current network situation, so that audio leakage detection is more in line with the current network security environment, and can better protect the data security and privacy of the target organization. Overall, accurate detection and timely processing of audio leakage of the target organization can be achieved, and the information security protection capability of the target organization can be improved.

[0020] Optionally, the performing data analysis on the discussion content of the sensitive topic to determine the matching degree between the discussion content and the content of several meetings of the target organization includes:

[0021] Analyze the discussion content of the sensitive topic and determine the content form of the discussion content;

[0022] If the content is in the form of sensitive audio, analyzing the sensitive audio to perform audio analysis and determine audio properties, where the audio properties include sampling rate, signal-to-noise ratio, and distortion;

[0023] According to the audio attribute, audio processing is performed on the sensitive audio, wherein the audio processing method includes audio noise reduction and audio equalization;

[0024] The processed sensitive audio is subjected to content extraction, and the extracted content is compared with the content of several meetings of the target organization to determine the degree of match.

[0025] By adopting the above technical solution, audio properties, including sampling rate, signal-to-noise ratio and distortion, can be obtained by performing audio analysis on sensitive audio. These properties can reflect the quality and integrity of the audio and help improve the accuracy of audio recognition. According to the audio properties, the use of processing methods such as audio noise reduction and audio equalization can effectively remove noise and imbalance in the audio and improve the quality and audibility of the audio. Through the sensitive audio after audio processing, content extraction can be performed on it, and important information in the audio can be extracted to form text or other readable forms of content, making the understanding and analysis of the audio content more convenient and accurate. Comparing the extracted content with the content of several meetings of the target organization can more accurately determine the degree of match, which helps to discover audio leaks and handle and prevent them in a timely manner. In general, by analyzing the discussion content of sensitive topics, the functions of audio processing, content extraction and matching degree identification of sensitive audio are realized, and the accuracy and efficiency of audio processing are improved.

[0026] Optionally, the discussion content includes sensitive audio; and performing watermark recognition on the discussion content according to the digital watermark of the target organization to obtain a watermark recognition result includes:

[0027] Analyze the sensitive audio to determine whether the sensitive audio contains a watermark;

[0028] If included, extracting the watermark in the sensitive audio, and comparing the watermark with the digital watermark of the target organization to determine whether the sensitive audio is the audio leaked from the meeting of the target organization;

[0029] If the audio is leaked from a meeting of the target organization, extract the historical meeting topics of the corresponding meeting content;

[0030] Based on the historical conference topics, segment and sort the sensitive audio so that the segmented and sorted sensitive audio corresponds to the historical conference topics;

[0031] Extracting the sorted continuous watermarks of the sensitive audio, analyzing the distribution of the continuous watermarks, and determining the leakage nodes of the conference content according to the distribution;

[0032] Generate a watermark recognition result according to the meeting content, the historical meeting topics and the leaked node;

[0033] If the matching degree is greater than the preset matching degree, or the watermark recognition result corresponds to the digital watermark, it is determined that an audio leak has occurred in the target organization, so as to remind the processing personnel to handle it, including:

[0034] If the matching degree is greater than the preset matching degree, it is determined that the target organization has audio leakage, so as to remind the processing personnel to handle it;

[0035] If the watermark recognition result shows that the sensitive audio is leaked audio from a meeting of the target organization, the watermark recognition result is sent to the processing personnel so that the processing personnel can locate the leaking node.

[0036] By adopting the above technical solution, by analyzing the sensitive audio, it can be determined whether the audio contains a watermark. If it contains a watermark, the watermark is further extracted and compared with the digital watermark of the target organization to determine whether it is the audio leaked from the meeting, so that the audio leakage can be discovered in time and corresponding treatment measures can be taken to improve the accuracy of audio leakage detection. By segmenting and sorting the sensitive audio and extracting continuous watermarks, the distribution of continuous watermarks can be analyzed, and the leakage nodes of the meeting content can be determined. The leakage nodes can be accurately located, and detailed leakage information and location instructions can be provided to the processing personnel so as to take effective treatment measures. By generating watermark recognition results and reminding the processing personnel to handle, this technical solution can timely remind the processing personnel to pay attention to and handle audio leakage problems, which can enhance the information security protection capabilities of the target organization and reduce possible losses and risks. By extracting watermarks from sensitive audio and locating leakage nodes, detailed leakage information and location instructions are provided to the processing personnel, so that the processing personnel can quickly locate and handle leakage problems and improve processing efficiency. At the same time, the watermark recognition results are only sent to the processing personnel, and only the processing personnel can obtain detailed information about the leakage nodes, protecting the privacy and legal rights of the target organization. In general, it can realize audio leakage detection and leakage node location of the target organization's meeting content, and improve the target organization's information security protection capabilities and processing efficiency.

[0037] Optionally, the method further includes:

[0038] Based on the installation position of the watermark sound wave generator and the speech process of the participant, determine the watermark attachment method of the target participant's position, and the target participant is a participant of the internal meeting corresponding to the conference content;

[0039] The leakage location is determined according to the leakage node and the watermark attachment method.

[0040] By adopting the above technical solution, taking into account the installation position of the watermark sound wave generator and the speaking process of the participants, the watermark attachment method of the target participant's location can be determined more accurately, improving the accuracy of the leakage location determination, and providing more accurate leakage location indications for the processing personnel. The watermark attachment method and the leakage node determination method can be flexibly adjusted according to different conference scenarios and needs, making the positioning of the leakage position more flexible and adaptable, and able to meet the leakage processing needs in different scenarios. By accurately determining the leakage location, a faster and more effective leakage processing method can be provided to the processing personnel, which helps to reduce processing time and cost and improve processing efficiency. Overall, the leakage location is accurately determined, the processing efficiency is improved, and more flexible and effective technical support is provided for audio leakage detection and processing.

[0041] In a second aspect, the present application provides an audio leakage protection device based on digital watermark, comprising:

[0042] A topic discussion degree determination module is used to scan the data of the monitorable platform according to the data scanning frequency to determine the topic discussion degree of the platform topic;

[0043] A sensitive topic determination module, used to determine sensitive topics in the platform topics according to the discussion degree of the topic and several discussion degree thresholds of the target organization;

[0044] A matching degree determination module, used to perform data analysis on the discussion content of the sensitive topic to determine the matching degree between the discussion content and the content of several meetings of the target organization;

[0045] A watermark recognition result determination module, configured to perform watermark recognition on the discussion content according to the digital watermark of the target organization to obtain a watermark recognition result if the matching degree is less than or equal to a preset matching degree;

[0046] The reminder module is used to determine that an audio leak has occurred in the target organization if the matching degree is greater than the preset matching degree, or the watermark recognition result corresponds to the digital watermark, so as to remind the processing personnel to handle it.

[0047] Optionally, the digital watermark-based audio leakage protection device further includes a data scanning frequency determination module, which is used to:

[0048] Obtain the data acquisition authority of the target organization, and determine the corresponding monitorable platform based on the data acquisition authority;

[0049] According to the data flow attributes of the monitorable platform and the organizational properties of the target organization, a data scanning frequency is correspondingly set for the monitorable platform.

[0050] Optionally, the digital watermark-based audio leakage protection device further includes a discussion threshold determination module, which is used to:

[0051] Obtain the historical meeting topics and corresponding meeting contents of the target organization;

[0052] A discussion degree threshold of each historical meeting is determined according to the historical meeting topics, the corresponding meeting content and the current network situation.

[0053] Optionally, the matching degree determination module is specifically used to:

[0054] Analyze the discussion content of the sensitive topic and determine the content form of the discussion content;

[0055] If the content is in the form of sensitive audio, analyzing the sensitive audio to perform audio analysis and determine audio properties, where the audio properties include sampling rate, signal-to-noise ratio, and distortion;

[0056] According to the audio attribute, audio processing is performed on the sensitive audio, wherein the audio processing method includes audio noise reduction and audio equalization;

[0057] The processed sensitive audio is subjected to content extraction, and the extracted content is compared with the content of several meetings of the target organization to determine the degree of match.

[0058] Optionally, the discussion content includes sensitive audio; when the watermark recognition result determination module performs watermark recognition on the discussion content according to the digital watermark of the target organization and obtains the watermark recognition result, it is specifically used to

[0059] Analyze the sensitive audio to determine whether the sensitive audio contains a watermark;

[0060] If included, extracting the watermark in the sensitive audio, and comparing the watermark with the digital watermark of the target organization to determine whether the sensitive audio is the audio leaked from the meeting of the target organization;

[0061] If the audio is leaked from a meeting of the target organization, extract the historical meeting topics of the corresponding meeting content;

[0062] Based on the historical conference topics, segment and sort the sensitive audio so that the segmented and sorted sensitive audio corresponds to the historical conference topics;

[0063] Extracting the sorted continuous watermarks of the sensitive audio, analyzing the distribution of the continuous watermarks, and determining the leakage nodes of the conference content according to the distribution;

[0064] Generate a watermark recognition result according to the meeting content, the historical meeting topics and the leaked node;

[0065] The reminder module is specifically used for:

[0066] If the matching degree is greater than the preset matching degree, it is determined that the target organization has audio leakage, so as to remind the processing personnel to handle it;

[0067] If the watermark recognition result shows that the sensitive audio is leaked audio from a meeting of the target organization, the watermark recognition result is sent to the processing personnel so that the processing personnel can locate the leaking node.

[0068] Optionally, the audio leakage protection device based on digital watermark further includes a leakage position determination module, which is used to:

[0069] Based on the installation position of the watermark sound wave generator and the speech process of the participant, determine the watermark attachment method of the target participant's position, and the target participant is a participant of the internal meeting corresponding to the conference content;

[0070] The leakage location is determined according to the leakage node and the watermark attachment method.

[0071] In a third aspect, the present application provides an electronic device, comprising: a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the method of the first aspect.

[0072] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute the method of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0074] Figure 1 A schematic diagram of an application scenario provided for an embodiment of the present application;

[0075] Figure 2 A flowchart of an audio leakage protection method based on digital watermarking provided in an embodiment of the present application;

[0076] Figure 3 A schematic diagram of the installation position of a watermark sound wave generator provided in a top view according to an embodiment of the present application;

[0077] Figure 4A schematic diagram of the structure of an audio leakage protection device based on digital watermark provided in one embodiment of the present application;

[0078] Figure 5 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application. Implementation

[0079] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0080] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0081] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0082] For many companies and senior institutions, internal meetings on improving their competitiveness and future development are very confidential. If they are leaked, it is likely to cause damage to personal reputation and cause significant losses to the interests of the company or the country. At present, many companies and senior institutions will ensure the security of confidential meetings by controlling the participants, formulating meeting disciplines, and encrypting the audio of the entire meeting. Among them, the method of encrypting the audio of the entire meeting is the most common, but the leakage of meeting content does not only occur when the complete audio is transmitted after the entire meeting, but also during the meeting. During the meeting, there may be participants who use recording equipment to record the content of the meeting. Then, when the secretly recorded audio is shared and discussed, it will be discovered only when it causes a high level of heat. At this time, handling may have caused a great impact or loss. Therefore, the current leakage protection method is less effective in protecting the meeting content of companies and senior institutions when the audio is leaked.

[0083] Based on this, the present application provides an audio leakage protection method, device and electronic device based on digital watermark. The data of the monitorable platform can be scanned according to the data scanning frequency to determine the topic discussion degree of the platform topic. Then, based on the topic discussion degree and several discussion degree thresholds of the target organization, the sensitive topics in the platform topics are determined. Data analysis is performed on the discussion content of the sensitive topic to determine the matching degree between the discussion content and several meeting contents of the target organization. If the matching degree is greater than or equal to the preset matching degree, the discussion content is watermarked according to the digital watermark of the target organization to obtain a watermark recognition result; if the matching degree is less than the preset matching degree, or the watermark recognition result corresponds to the digital watermark, it is determined that an audio leak has occurred in the target organization to remind the processing personnel to handle it.

[0084] Figure 1 This is a schematic diagram of an application scenario provided by this application. Different enterprises and senior institutions may have different data monitoring permissions, and correspondingly different platforms may be used for monitoring. Figure 1 In the application scenario shown, the audio leakage protection method based on digital watermark can be built on an audio leakage analysis server. The audio leakage analysis server can be associated with a platform (monitorable platform) that enterprises and senior institutions have the authority to monitor and perform data monitoring. By analyzing the discussion of platform topics on the platform, sensitive topics can be determined, and then the content contained in the sensitive topics can be analyzed to see whether it is the meeting content leaked by the enterprise or senior institution, so as to deal with it in time.

[0085] For specific implementation methods, please refer to the following embodiments.

[0086] Figure 2 This is a flowchart of a method for protecting audio leakage based on digital watermarks provided in an embodiment of the present application. The method of this embodiment can be applied to the audio leakage analysis server in the above scenario. Figure 2 As shown, the method includes:

[0087] S201. Scan the data of the monitorable platform according to the data scanning frequency to determine the topic discussion degree of the platform topic.

[0088] The audio leakage analysis server can be associated with the platforms that the corresponding enterprise or senior organization has the authority to monitor. These platforms can be used as monitorable platforms, and there can be multiple monitorable platforms. Since the users of different monitorable platforms are of different age groups, their browsing preferences and attention to different types of topics are also different, so the attention to topics consistent with the meeting direction of the enterprise or senior organization (collectively referred to as organization) is also different, and the data scanning frequency can be set based on this.

[0089] The data scanning frequency can be used to indicate how often the data on the monitorable platform is identified and processed, so as to determine which platform topics the users of the monitorable platform have paid attention to during the period between the current scan and the previous scan, as well as the degree of discussion of each platform topic. The degree of discussion of a topic can be specifically reflected in the number of views, comments, user sharing, and specific content of user comments on the platform topic. The more views, comments, and sharing, and the more intense the specific content of the comments, the higher the degree of discussion of the topic.

[0090] Different values ​​of topic discussion degrees can be set for different views, comments, user sharing, and specific content of user comments. The topic discussion degree can be matched according to the summary of the views, comments, user sharing, and specific content of user comments of the corresponding platform topic.

[0091] S202. Determine sensitive topics in the platform topics based on the topic discussion degree and several discussion degree thresholds of the target organization.

[0092] The target organization can be an enterprise, high-level institution or other organization that needs to protect against audio leakage at the moment. There are many meetings held in the target organization, and the main content and theme of each meeting may be different. If different themes and content are leaked on the monitored platform, the discussion level may be different. Since the content discussed in different time periods in a meeting may be different, the corresponding possible discussion level threshold can be set for the meeting based on the content of each time period of each meeting, so each meeting may correspond to multiple discussion level thresholds. Sensitive topics can be used to indicate topics that may be generated after the content of the target organization's meeting is leaked, which may be audio or text.

[0093] Specifically, the topic discussion degree may be compared with several discussion degree thresholds of the target organization. When the topic discussion degree is greater than or equal to any discussion degree threshold, the platform topic corresponding to the topic discussion degree may be determined as a sensitive topic.

[0094] S203: Perform data analysis on the discussion content of sensitive topics to determine the matching degree between the discussion content and the content of several meetings of the target organization.

[0095] The discussion content of sensitive topics can include the target of the platform topic that triggered the discussion among platform users and what kind of speech the users made regarding the target, etc. The target can be an audio clip or a text clip. The main content discussed in each meeting held by the target organization can be used as the meeting content, which can include the speeches of each participant.

[0096] Specifically, different data analysis can be performed on the discussion content of different types of targets, such as audio and text, and combined with the speeches of the target platform users, the theme of the sensitive topic and the content related to the theme can be analyzed and determined. The theme and the content related to the theme can be matched with the content of several meetings of the target organization to determine the matching degree.

[0097] In some implementations, it can be determined whether the subject of the sensitive topic or the content related to the subject is the same as the meeting content of the target organization. If not, the matching degree is 0. If the same content is found, the total audio content corresponding to each position corresponding to the content of the sensitive topic is extracted from each meeting content of the target organization. The content that is the same as the total audio content is divided by the total audio content, and the quotient is used to determine the matching degree.

[0098] If it is audio, divide the duration of the same audio content by the duration of the total audio content; if it is text, divide the number of the same text content by the number of related texts corresponding to the total audio content. The relevant text here may refer to the text after removing the text that is not related to the meeting topic.

[0099] S204: If the matching degree is less than or equal to the preset matching degree, watermark recognition is performed on the discussion content according to the digital watermark of the target organization to obtain a watermark recognition result.

[0100] For each discussion section in the meeting content, calculate the ratio of the more important words in the discussion, such as words related to scientific research, funding, and future development, and the words involving internal confidentiality in the discussion to the overall vocabulary or recording duration of the entire discussion section, and use this value as the preset matching degree.

[0101] When the target organization holds a meeting in a conference room, the entire meeting process can be recorded. A watermark sound wave generator can be installed in the meeting to emit sound waves in real time during the meeting, so that the sound waves can be recorded together with the meeting content to generate a digital watermark. Various styles of digital watermarks in the recorded audio when secretly recording at various locations in the conference room can be simulated and stored in advance. The watermark recognition result can include whether it corresponds to the digital watermark of the target organization, which meeting it corresponds to, and which part of the meeting.

[0102] Specifically, if the matching degree is less than or equal to the preset matching degree, the discussion content can be identified using a watermark recognition tool, and the identified watermark is compared with the pre-stored digital watermark of the target organization to obtain a watermark recognition result.

[0103] In some implementations, if no watermark is recognized, then it is determined not to be relevant to the target organization's meeting.

[0104] S205. If the matching degree is greater than a preset matching degree, or the watermark recognition result corresponds to the digital watermark, it is determined that an audio leak has occurred in the target organization, so as to alert the processing personnel to handle it.

[0105] The case where the watermark recognition result corresponds to the digital watermark may include that the watermark of the content discussed in the watermark recognition result is partially identical or completely identical to the digital watermark of the target organization.

[0106] Specifically, if the matching degree is greater than the preset matching degree, it can be directly determined that the relevant audio of the meeting content of the target organization has been leaked. If it is determined in step S204 that the watermark recognition result corresponds to the digital watermark, it can also indicate that audio leakage has occurred in the target organization. At this time, a reminder can be sent to the relevant processing personnel to handle the leaked audio in time.

[0107] This embodiment scans the data of the monitorable platform through the data scanning frequency, which can quickly and accurately determine the topic discussion degree of the platform topic, timely discover sensitive topics, and improve the efficiency and accuracy of monitoring. According to the topic discussion degree and several discussion degree thresholds of the target organization, the sensitive topics in the platform topic can be determined, so as to carry out targeted analysis and monitoring, thereby improving the targeted monitoring. Data analysis of the discussion content of sensitive topics can determine the matching degree between the discussion content and several meeting contents of the target organization, so as to determine whether there is an audio leak, thereby improving the accuracy of audio leak detection. According to the digital watermark of the target organization, the discussion content is watermarked, and the watermark recognition result can be obtained to further confirm the audio leak, thereby improving the accuracy of audio leak recognition. When the matching degree is less than or equal to the preset matching degree or the watermark recognition result corresponds to the digital watermark, it can be determined that an audio leak has occurred in the target organization, and the processing personnel can be reminded in time to handle it, thereby avoiding possible losses and risks. Through technical means such as data scanning, topic discussion analysis, sensitive topic identification, matching judgment and watermark recognition, we have achieved rapid and accurate detection and identification of audio leaks on the monitoring platform, improved the efficiency and accuracy of audio leak monitoring, and enhanced the protection effect when the meeting content of the target organization is leaked.

[0108] In some embodiments, the monitorable platform that can be associated can be determined based on the data acquisition authority of the target organization, and then the data scanning frequency can be set for the monitorable platform. Specifically, the data acquisition authority of the target organization is obtained, and the corresponding monitorable platform is determined according to the data acquisition authority; according to the data flow attribute of the monitorable platform, the data scanning frequency is set for the monitorable platform.

[0109] For some high-level organizations, they can monitor more data and manage many platforms, while for general enterprises, they may only be able to monitor data on ordinary public platforms. The permissions brought by these inherent attributes of the target organization can be used as data acquisition permissions. Data acquisition permissions can be permissions predetermined by the target organization.

[0110] The data types, information, audio, etc. promoted on different monitorable platforms may be different, and the age range of users of this platform is different. People of different ages discuss different content information to different degrees. In addition, the degree of data cleaning of different monitorable platforms is different. For example, some confidential information or national information will be automatically filtered and not displayed. These characteristics of the monitorable platform can be used as data flow attributes, and the corresponding data flow attributes can be determined by combining the historical traffic of different monitorable platforms. Due to the differences in the main business and business of the target organization, the direction and content of internal meetings are different, which can be expressed as different organizational natures.

[0111] Specifically, the data acquisition authority of the target organization stored in advance can be obtained first, and then based on the data acquisition authority, the corresponding monitorable platforms within the authority range can be found. The data flow attributes of the monitorable platform and the organizational nature of the target organization are extracted for analysis. The greater the correlation between the platform topic pushed by the monitorable platform and the organizational nature shown by the data flow attribute, the faster the corresponding data scanning frequency is set.

[0112] This embodiment achieves data monitoring and audio leakage detection for a specific target organization by obtaining the data acquisition authority of the target organization, determining the corresponding monitorable platform based on the data acquisition authority, and then setting the data scanning frequency for the monitorable platform based on the data flow attributes of the monitorable platform. Specifically, the data acquisition authority of the target organization can be obtained, so that various data sources of the target organization can be accessed. By using the data acquisition authority, various data of the target organization can also be obtained, and analyzed and processed to detect whether there is an audio leakage. It is possible to achieve audio leakage detection and monitoring of a specific target organization, and the data scanning frequency and the selection of the monitorable platform can be flexibly adjusted according to the actual situation, thereby achieving better technical effects.

[0113] In some embodiments, the discussion threshold can be set based on the historical meeting topics and corresponding meeting contents of the target organization and the current network situation. Specifically, the historical meeting topics and corresponding meeting contents of the target organization are obtained; based on the historical meeting topics, corresponding meeting contents and the current network situation, the discussion threshold of each historical meeting is determined.

[0114] Before the current analysis, all the meeting topics of the target organization can be called historical meeting topics, and each historical meeting topic has its own corresponding meeting content. Every meeting held by the target organization can be recorded and stored in real time.

[0115] As time goes by, things happening in real life are constantly updated, so the hotly discussed content on the Internet in different time periods is different. At this time, other content will be ignored to a certain extent. This situation can be regarded as the current network situation. Since the current network situation is changing, the discussion threshold of the same historical meeting at different times is also different. A discussion threshold determination model to be trained can be established in advance, and the historical meeting themes of the target organization, the corresponding meeting contents, different network situations, and the discussion threshold of each historical meeting under the network situation can be used to construct a training set. Then, training samples consisting of historical meeting themes, corresponding meeting contents, and different network situations can be obtained from the training set, and input into the discussion threshold determination model to be trained to obtain the output result, that is, the discussion threshold of different historical meetings under different network situations. The discussion threshold obtained by training is compared with the actual corresponding discussion threshold in the training set, and the loss function is calculated. The model parameters of the discussion threshold determination model to be trained are adjusted using the gradient descent method to reduce the loss function, and the training is stopped until the loss function is no longer reduced, and the discussion threshold determination model is obtained.

[0116] Specifically, we can first obtain the historical meeting topics and corresponding meeting contents of the target organization, and then input the historical meeting topics, corresponding meeting contents and current network situation into the discussion threshold determination model to directly output the discussion threshold of each historical meeting.

[0117] By analyzing the historical meeting themes and corresponding meeting contents of the target organization, this embodiment can understand the topic focus direction, discussion focus and decision-making results of the target organization, so as to more accurately determine the discussion threshold, which is helpful to timely discover audio leakage and take corresponding treatment measures, and improve the information security protection capability of the target organization. According to the historical meeting themes, the corresponding meeting contents and the current network situation, personalized discussion thresholds can be set for different meeting scenarios, making audio leakage detection more flexible and accurate, and better meeting the actual needs of different target organizations. Through the analysis of historical meeting themes and corresponding meeting contents, the key topics and discussion hotspots of the target organization can be quickly located, so as to carry out audio monitoring and data scanning in a targeted manner, improve monitoring efficiency, and reduce unnecessary calculations and resource consumption. In the process of determining the discussion threshold, a comprehensive analysis can be carried out according to the current network situation, so that audio leakage detection is more in line with the current network security environment, and the data security and privacy of the target organization can be better protected. Overall, accurate detection and timely processing of audio leakage of the target organization can be achieved, and the information security protection capability of the target organization can be improved.

[0118] In some embodiments, when the content of the discussion is in the form of sensitive audio, the degree of match can be determined in combination with the audio attributes. Specifically, the discussion content of sensitive topics is analyzed to determine the content form of the discussion content; if the content form is sensitive audio, the sensitive audio is analyzed for audio analysis to determine the audio attributes, which include sampling rate, signal-to-noise ratio, and distortion; based on the audio attributes, the sensitive audio is processed, and the audio processing methods include audio noise reduction and audio equalization; the processed sensitive audio is subjected to content extraction, and the extracted content is compared with the content of several meetings of the target organization to determine the degree of match.

[0119] The discussion content may include audio, audio and text, etc. Since the audio leakage analysis server analyzes the leaked audio, the discussion content needs to include audio. The audio of the discussion content corresponding to sensitive topics can be used as sensitive audio. Different audio processing methods can be pre-set for sensitive audio with different audio attributes.

[0120] Specifically, audio and text extraction can be performed on the discussion content of sensitive topics to determine the content format of the discussion content. If the content format is sensitive audio, use software such as Adobe Audition to perform audio analysis on the sensitive audio to determine the various audio attributes. Then find the audio processing method corresponding to the audio attribute, and perform noise reduction, equalization and other processing on the sensitive audio. Perform voiceprint extraction, text conversion and audio content on the processed sensitive audio, and compare it with the voiceprints, converted text and audio content corresponding to several meeting contents of the target organization to determine the degree of match.

[0121] In some implementations, if the discussion content of sensitive topics includes text, the text can be directly compared with the text converted from several meeting contents of the target organization, and then the sensitive audio can be combined with the comparison results of the hot audio in several meetings to determine the match.

[0122] This embodiment can obtain audio properties, including sampling rate, signal-to-noise ratio, and distortion, by performing audio analysis on sensitive audio. These properties can reflect the quality and integrity of the audio and help improve the accuracy of audio recognition. According to the audio properties, the use of processing methods such as audio noise reduction and audio equalization can effectively remove noise and imbalance in the audio and improve the quality and audibility of the audio. Through the sensitive audio after audio processing, content extraction can be performed on it, and important information in the audio can be extracted to form text or other readable forms of content, making the understanding and analysis of the audio content more convenient and accurate. Comparing the extracted content with the content of several meetings of the target organization can more accurately determine the degree of match, which helps to discover audio leaks and handle and prevent them in a timely manner. In general, by analyzing the discussion content of sensitive topics, the functions of audio processing, content extraction, and matching degree identification of sensitive audio are realized, and the accuracy and efficiency of audio processing are improved.

[0123] In some embodiments, the leakage node of the target organization can be determined, and then a watermark recognition result can be generated. Specifically, the sensitive audio is analyzed to determine whether the sensitive audio contains a watermark; if it does, the watermark in the sensitive audio is extracted, and the watermark is compared with the digital watermark of the target organization to determine whether the sensitive audio is the audio leaked from the meeting of the target organization; if it is the audio leaked from the meeting of the target organization, the historical meeting topics of the corresponding meeting content are extracted; based on the historical meeting topics, the sensitive audio is segmented and sorted so that the segmented and sorted sensitive audio corresponds to the historical meeting topics; the continuous watermarks of the sorted sensitive audio are extracted, the distribution of the continuous watermarks is analyzed, and the leakage nodes of the meeting content are determined according to the distribution; the watermark recognition result is generated according to the meeting content, historical meeting topics and leakage nodes.

[0124] Correspondingly, if the matching degree is greater than the preset matching degree, it is determined that an audio leak has occurred in the target organization, so as to remind the processing personnel to handle it; if the watermark recognition result shows that the sensitive audio is the audio leaked from the meeting of the target organization, the watermark recognition result will be sent to the processing personnel to enable the processing personnel to locate the leakage node.

[0125] Since a watermark sound wave generator is pre-installed in the target organization's meeting room, the distribution of the watermark of the overall audio during the target organization's meeting and the distribution of the watermarks corresponding to each audio segment in an overall audio can be determined.

[0126] Specifically, watermark detectors and other tools can be used to detect watermarks in sensitive audio. If a watermark is included, the audio content corresponding to the watermark is extracted and compared with the digital watermark of the target organization. If the same watermark exists, the sensitive audio is determined to be the audio leaked from the meeting of the target organization. The same watermark comes from which meeting content, and then the historical meeting topics corresponding to these meeting contents are extracted to determine what the meeting mainly involves. Then, based on this direction, the sensitive audio is segmented and sorted so that the segmented and sorted sensitive audio corresponds to the historical meeting topic, avoiding analysis errors caused by the disrupted upload of sensitive audio. If there is no identical watermark, it means that it is not leaked by the target organization. The sorted sensitive audio is processed using watermark extraction software to obtain a continuous watermark, and the distribution mode of the continuous watermark is compared with the distribution mode of the digital watermark of the meeting content to determine the leakage node of the meeting content. The meeting content, historical meeting topic and leakage node are taken as the watermark recognition result as a whole. When the matching degree is greater than the preset matching degree, it is directly determined that the target organization has an audio leak, so as to remind the processing personnel to handle it. If the watermark recognition result shows that the sensitive audio is the audio leaked from the meeting of the target organization, the watermark recognition result is sent to the processing personnel so that the processing personnel can locate the leaking node.

[0127] This embodiment can determine whether the audio contains a watermark by analyzing the sensitive audio. If it contains a watermark, the watermark is further extracted and compared with the digital watermark of the target organization to determine whether it is the audio leaked from the meeting. The audio leakage can be discovered in time and corresponding treatment measures can be taken to improve the accuracy of audio leakage detection. By segmenting and sorting the sensitive audio and extracting continuous watermarks, the distribution of continuous watermarks can be analyzed, and the leakage nodes of the conference content can be determined. The leakage nodes can be accurately located, and detailed leakage information and location instructions can be provided to the processing personnel so that effective treatment measures can be taken. By generating watermark recognition results and reminding the processing personnel to handle, this technical solution can timely remind the processing personnel to pay attention to and handle audio leakage problems, which can enhance the information security protection capabilities of the target organization and reduce possible losses and risks. By extracting watermarks from sensitive audio and locating leakage nodes, detailed leakage information and location instructions are provided to the processing personnel, so that the processing personnel can quickly locate and handle leakage problems and improve processing efficiency. At the same time, the watermark recognition results are only sent to the processing personnel, and only the processing personnel can obtain detailed information about the leakage nodes, protecting the privacy and legal rights of the target organization. In general, it can realize audio leakage detection and leakage node location of the target organization's meeting content, and improve the target organization's information security protection capabilities and processing efficiency.

[0128] In some embodiments, the watermark attachment method can be determined in combination with the installation location of the watermark sound wave generator and the speech process, and then the leakage location can be determined. Specifically, based on the installation location of the watermark sound wave generator and the speech process of the participant, the watermark attachment method of the target participant's location is determined. The target participant is the participant of the internal meeting corresponding to the conference content; the leakage location is determined according to the leakage node and the watermark attachment method.

[0129] The watermark sound wave generator can be a device that can generate sound waves of special frequencies. The installation of the watermark sound wave generator in the conference room can refer to the following: Figure 3 A schematic diagram of the installation position of a watermark sound wave generator from a bird's-eye view is shown. A watermark sound wave generator can be installed at each of the four corners of the ceiling of the conference room. A number of seats are set up next to the conference table, and then watermark sound wave generators are installed on both sides of each seat to ensure that there are watermark sound wave generators on the left and right sides of each seat. For the watermark sound wave generator on the ceiling, it can be a functional module integrated in other devices, and in some implementations, it can be integrated into the surveillance camera of the conference room. Since the watermark sound wave generator is installed in various positions, and when participants in different seats make speeches or discussions are held in the meeting, the propagation process of the sound is blocked differently. Therefore, the sound waves emitted by the watermark sound wave generator recorded at different locations of the secretly recorded equipment are different, and the corresponding watermarks generated are also different, which can be used as different watermark attachment methods.

[0130] The actual scene model can be established in advance to simulate the situation when recording audio at different locations in the conference room scene, and obtain the watermark attachment methods at different secretly recorded locations. Then, several watermark attachment methods corresponding to the speech process of the participants in the meeting are extracted, and combined with the leakage time corresponding to the leakage node, the location of the secretly recorded audio, that is, the leakage location, is quickly located, and the secretly recorded participant is further determined.

[0131] This embodiment takes into account the installation location of the watermark sound wave generator and the speaking process of the participants, and can more accurately determine the watermark attachment method of the target participant's location, improve the accuracy of determining the leakage location, and provide more accurate leakage location indications for processing personnel. The watermark attachment method and the leakage node determination method can be flexibly adjusted according to different conference scenarios and needs, making the positioning of the leakage position more flexible and adaptable, and able to meet the leakage processing needs in different scenarios. By accurately determining the leakage location, a faster and more effective leakage processing method can be provided to the processing personnel, which helps to reduce processing time and cost and improve processing efficiency. Overall, the leakage location is accurately determined, the processing efficiency is improved, and more flexible and effective technical support is provided for audio leakage detection and processing.

[0132] Figure 4 A schematic diagram of the structure of an audio leakage protection device based on digital watermarking provided in an embodiment of the present application is shown in FIG. Figure 4 As shown, the digital watermark-based audio leakage protection device 400 of this embodiment includes: a topic discussion degree determination module 401, a sensitive topic determination module 402, a matching degree determination module 403, a watermark recognition result determination module 404 and a reminder module 405.

[0133] The topic discussion degree determination module 401 is used to scan the data of the monitorable platform according to the data scanning frequency to determine the topic discussion degree of the platform topic;

[0134] A sensitive topic determination module 402, for determining sensitive topics in platform topics according to the topic discussion degree and several discussion degree thresholds of the target organization;

[0135] A matching degree determination module 403 is used to perform data analysis on the discussion content of sensitive topics and determine the matching degree between the discussion content and the content of several meetings of the target organization;

[0136] The watermark recognition result determination module 404 is used to perform watermark recognition on the discussion content according to the digital watermark of the target organization to obtain a watermark recognition result if the matching degree is less than or equal to the preset matching degree;

[0137] The reminder module 405 is used to determine that an audio leak has occurred in the target organization if the matching degree is greater than a preset matching degree, or the watermark recognition result corresponds to the digital watermark, so as to remind the processing personnel to handle it.

[0138] Optionally, the digital watermark-based audio leakage protection device 400 further includes a data scanning frequency determination module 406, which is used to:

[0139] Obtain the target organization's data access permissions, and determine the corresponding monitorable platform based on the data access permissions;

[0140] According to the data flow attributes of the monitored platform and the organizational nature of the target organization, the data scanning frequency is set accordingly for the monitored platform.

[0141] Optionally, the digital watermark-based audio leakage protection device 400 further includes a discussion threshold determination module 407, which is used to:

[0142] Obtain the historical meeting topics and corresponding meeting contents of the target organization;

[0143] The discussion threshold of each historical meeting is determined based on the historical meeting topics, corresponding meeting content and current network situation.

[0144] Optionally, the matching degree determination module 403 is specifically used for:

[0145] Analyze the discussion content of sensitive topics and determine the content format of the discussion content;

[0146] If the content is in the form of sensitive audio, the sensitive audio is analyzed to perform audio analysis and determine audio properties, including sampling rate, signal-to-noise ratio, and distortion;

[0147] Perform audio processing on sensitive audio according to audio properties. The audio processing methods include audio noise reduction and audio equalization.

[0148] The processed sensitive audio is then used for content extraction, and the extracted content is compared with several meeting contents of the target organization to determine the degree of match.

[0149] Optionally, the discussion content includes sensitive audio; when the watermark recognition result determination module 404 performs watermark recognition on the discussion content according to the digital watermark of the target organization and obtains the watermark recognition result, it is specifically used for

[0150] Analyze sensitive audio to determine whether the sensitive audio contains watermarks;

[0151] If it is included, the watermark in the sensitive audio is extracted and compared with the digital watermark of the target organization to determine whether the sensitive audio is the audio leaked from the meeting of the target organization;

[0152] If the audio is leaked from a meeting of the target organization, extract the historical meeting topics of the corresponding meeting content;

[0153] Based on the historical meeting topics, the sensitive audio is segmented and sorted so that the segmented and sorted sensitive audio corresponds to the historical meeting topics;

[0154] Extract the continuous watermarks of the sorted sensitive audio, analyze the distribution of the continuous watermarks, and determine the leakage nodes of the conference content based on the distribution;

[0155] Generate watermark recognition results based on meeting content, historical meeting topics and leaked nodes;

[0156] The reminder module 405 is specifically used for:

[0157] If the matching degree is greater than the preset matching degree, it is determined that the target organization has experienced an audio leak, so as to alert the processing personnel to handle it;

[0158] If the watermark recognition result shows that the sensitive audio is the audio leaked from the meeting of the target organization, the watermark recognition result is sent to the processing personnel so that the processing personnel can locate the leaking node.

[0159] Optionally, the digital watermark-based audio leakage protection device 400 further includes a leakage position determination module 408, which is used to:

[0160] Based on the installation position of the watermark sound wave generator and the speech process of the participant, the watermark attachment method of the target participant's position is determined. The target participant is the participant of the internal meeting corresponding to the meeting content;

[0161] Determine the leakage location based on the leakage node and watermark attachment method.

[0162] The device of this embodiment can be used to execute the method of any of the above embodiments. Its implementation principle and technical effects are similar and will not be described in detail here.

[0163] Figure 5 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application is shown in FIG. Figure 5 As shown, the electronic device 500 of this embodiment may include: a memory 501 and a processor 502 .

[0164] The memory 501 stores a computer program that can be loaded by the processor 502 and execute the method in the above embodiment.

[0165] The processor 502 and the memory 501 are connected, for example, via a bus.

[0166] Optionally, the electronic device 500 may further include a transceiver. It should be noted that in actual applications, the number of transceivers is not limited to one, and the structure of the electronic device 500 does not constitute a limitation on the embodiments of the present application.

[0167] The processor 502 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 502 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0168] The bus may include a path to transmit information between the above components. The bus may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0169] The memory 501 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0170] The memory 501 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 502. The processor 502 is used to execute the application code stored in the memory 501 to implement the contents shown in the above method embodiment.

[0171] The electronic devices include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and fixed terminals such as digital TVs, desktop computers, etc. It can also be a server, etc. Figure 5 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0172] The electronic device of this embodiment can be used to execute the method of any of the above embodiments, and its implementation principle and technical effect are similar, which will not be described in detail here.

[0173] The present application also provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute the method in the above embodiment.

[0174] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

Claims

1. A method for protecting audio from leakage based on digital watermark, characterized in that: include: Scan the data of the monitorable platform according to the data scanning frequency to determine the topic discussion degree of the platform topic; Determine sensitive topics among the platform topics based on the discussion degree of the topic and several discussion degree thresholds of the target organization; Performing data analysis on the discussion content of the sensitive topic to determine the matching degree between the discussion content and the content of several meetings of the target organization; If the matching degree is less than or equal to the preset matching degree, watermark recognition is performed on the discussion content according to the digital watermark of the target organization, and the recognized watermark is compared with the pre-stored digital watermark of the target organization to obtain a watermark recognition result; If the matching degree is greater than the preset matching degree, or the watermark recognition result corresponds to the digital watermark, it is determined that an audio leak has occurred in the target organization to alert the processing personnel to handle it; wherein, the situation where the watermark recognition result corresponds to the digital watermark includes that the watermark of the content discussed in the watermark recognition result is partially identical or completely identical to the digital watermark of the target organization.

2. The method for protecting audio leakage based on digital watermark according to claim 1, characterized in that: Also includes: Obtain the data acquisition authority of the target organization, and determine the corresponding monitorable platform based on the data acquisition authority; According to the data flow attributes of the monitorable platform and the organizational properties of the target organization, a data scanning frequency is correspondingly set for the monitorable platform.

3. The method for audio leakage protection based on digital watermark according to claim 1, characterized in that: Also includes: Obtain the historical meeting topics and corresponding meeting contents of the target organization; The discussion threshold of each historical meeting is determined based on the historical meeting topics, the corresponding meeting content and the current network situation; wherein, as time goes by, things happening in real life are constantly updated, so the hotly discussed content on the Internet in different time periods is different, and other content will be ignored to a certain extent at this time. This situation is taken as the current network situation.

4. The method for protecting audio leakage based on digital watermark according to any one of claims 1 to 3, characterized in that: The data analysis of the discussion content of the sensitive topic to determine the matching degree between the discussion content and the content of several meetings of the target organization includes: Analyze the discussion content of the sensitive topic and determine the content form of the discussion content; If the content is in the form of sensitive audio, performing audio analysis on the sensitive audio to determine audio properties, where the audio properties include sampling rate, signal-to-noise ratio, and distortion; According to the audio attribute, audio processing is performed on the sensitive audio, wherein the audio processing method includes audio noise reduction and audio equalization; The processed sensitive audio is subjected to content extraction, and the extracted content is compared with the content of several meetings of the target organization to determine the degree of match.

5. The method for audio leakage protection based on digital watermark according to claim 2, characterized in that: Said discussion included sensitive audio; The step of performing watermark recognition on the discussion content according to the digital watermark of the target organization to obtain a watermark recognition result includes: Analyze the sensitive audio to determine whether the sensitive audio contains a watermark; If included, extracting the watermark in the sensitive audio, and comparing the watermark with the digital watermark of the target organization to determine whether the sensitive audio is the audio leaked from the meeting of the target organization; If the audio is leaked from a meeting of the target organization, extract the historical meeting topics of the corresponding meeting content; Based on the historical conference topics, segment and sort the sensitive audio so that the segmented and sorted sensitive audio corresponds to the historical conference topics; Extracting the sorted continuous watermarks of the sensitive audio, analyzing the distribution of the continuous watermarks, and determining the leakage nodes of the conference content according to the distribution; Generate a watermark recognition result according to the meeting content, the historical meeting topics and the leaked node; If the matching degree is greater than the preset matching degree, or the watermark recognition result corresponds to the digital watermark, it is determined that an audio leak has occurred in the target organization, so as to remind the processing personnel to handle it, including: If the matching degree is greater than the preset matching degree, it is determined that the target organization has audio leakage, so as to remind the processing personnel to handle it; If the watermark recognition result shows that the sensitive audio is leaked audio from a meeting of the target organization, the watermark recognition result is sent to the processing personnel so that the processing personnel can locate the leaking node.

6. The method for protecting audio leakage based on digital watermark according to claim 5, characterized in that: Also includes: Based on the installation position of the watermark sound wave generator and the speech process of the participant, determine the watermark attachment method of the target participant's position, and the target participant is a participant of the internal meeting corresponding to the conference content; The leakage location is determined according to the leakage node and the watermark attachment method.

7. An audio leakage protection device based on digital watermark, characterized in that: include: A topic discussion degree determination module is used to scan the data of the monitorable platform according to the data scanning frequency to determine the topic discussion degree of the platform topic; A sensitive topic determination module, used to determine sensitive topics in the platform topics according to the discussion degree of the topic and several discussion degree thresholds of the target organization; A matching degree determination module, used to perform data analysis on the discussion content of the sensitive topic to determine the matching degree between the discussion content and the content of several meetings of the target organization; A watermark recognition result determination module, configured to perform watermark recognition on the discussion content according to the digital watermark of the target organization if the matching degree is less than or equal to a preset matching degree, and compare the recognized watermark with the pre-stored digital watermark of the target organization to obtain a watermark recognition result; A reminder module is used to determine that an audio leak has occurred in the target organization if the matching degree is greater than the preset matching degree, or the watermark recognition result corresponds to the digital watermark, so as to remind the processing personnel to handle it, wherein the situation where the watermark recognition result corresponds to the digital watermark includes that the watermark of the content discussed in the watermark recognition result is partially identical or completely identical to the digital watermark of the target organization.

8. The audio leakage protection device based on digital watermark according to claim 7 is characterized in that: Also included is a data scanning frequency determination module, which is used to: Obtain the data acquisition authority of the target organization, and determine the corresponding monitorable platform based on the data acquisition authority; According to the data flow attributes of the monitorable platform and the organizational properties of the target organization, a data scanning frequency is correspondingly set for the monitorable platform.

9. The audio leakage protection device based on digital watermark according to claim 7, characterized in that: It also includes a discussion threshold determination module for: Obtain the historical meeting topics and corresponding meeting contents of the target organization; The discussion threshold of each historical meeting is determined based on the historical meeting topics, the corresponding meeting content and the current network situation; wherein, as time goes by, things happening in real life are constantly updated, so the hotly discussed content on the Internet in different time periods is different, and other content will be ignored to a certain extent at this time. This situation is taken as the current network situation.

10. An electronic device, characterized in that: include: Memory and processor; The memory is used to store program instructions; The processor is used to call and execute the program instructions in the memory to perform the audio leakage protection method based on digital watermark as described in any one of claims 1-6.