Audio file anti-eavesdropping processing method and device, equipment and storage medium

Abstract

The embodiment of the invention provides an anti-eavesdropping processing method and device for an audio file, equipment and a storage medium, and the method comprises the steps: obtaining a target audio file; performing voice recognition on the target audio file to obtain voiceprint features and Hertz information of each spokesman included in the target audio file; determining whether a target spokesman exists or not according to the voiceprint features and the Hertz information of each spokesman; and if the target spokesman exists, performing anti-eavesdropping processing on the target audio file in the process of playing the target audio file. Compared with a traditional physical shielding or electromagnetic interference method, the method has the advantages that extra special hardware investment is not needed, recognition and protection of the sound of a specific user can be achieved only through software processing, the anti-eavesdropping efficiency and accuracy are greatly improved, and meanwhile the use cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of communication security technology, and in particular to an anti-eavesdropping processing method for audio files, an anti-eavesdropping processing device for audio files, an electronic device and a computer-readable storage medium. Background Art

[0002] In today's highly developed digital communication, voice calls and recordings have become a common way of communication in people's daily lives. However, with the advancement of technology, recording equipment has become increasingly miniaturized and concealed, making eavesdropping more difficult to detect, posing a serious threat to personal privacy and information security. Especially in sensitive occasions such as business negotiations, political discussions, and personal privacy exchanges, eavesdropping recordings may lead to information leakage, damage to rights and interests, and even cause serious social problems. Existing anti-eavesdropping technologies are mainly concentrated at the hardware level, such as using physical shielding, electromagnetic interference and other means to prevent the normal operation of recording equipment, but these methods are inconvenient to operate, cost-effective, and lack recognition accuracy. Summary of the invention

[0003] In view of the above problems, embodiments of the present invention are proposed to provide an audio file anti-eavesdropping processing method, an audio file anti-eavesdropping processing device, an electronic device and a computer-readable storage medium that overcome the above problems or at least partially solve the above problems.

[0004] In order to solve the above problem, an embodiment of the present invention discloses an anti-eavesdropping processing method for an audio file, comprising:

[0005] Get the target audio file;

[0006] Performing sound recognition on the target audio file to obtain the voiceprint features and Hertz information of each speaker contained in the target audio file;

[0007] Determining whether there is a target speaker based on the voiceprint features and Hertz information of each speaker;

[0008] If the target speaker exists, anti-eavesdropping processing is performed on the target audio file during the process of playing the target audio file.

[0009] Optionally, determining whether there is a target speaker according to the voiceprint features and Hertz information of each speaker includes:

[0010] Obtain the target speaker's voiceprint characteristics and Hertz information;

[0011] Matching the voiceprint features and Hertz information of each speaker with the voiceprint features and Hertz information of the target speaker respectively;

[0012] Based on the matching results, determine whether the target speaker exists.

[0013] Optionally, performing anti-eavesdropping processing on the target audio file during the process of playing the target audio file includes:

[0014] Acquire an anti-eavesdropping strategy, where the anti-eavesdropping strategy is used to indicate target voice information that needs to be processed for anti-eavesdropping;

[0015] Determine the target voice information in the target audio file that complies with the anti-eavesdropping strategy;

[0016] During the process of playing the target audio file, anti-eavesdropping processing is performed on the target voice information in the target audio file.

[0017] Optionally, the anti-eavesdropping strategy is used to indicate that voice information that meets preset conditions is target voice information that needs to be processed for anti-eavesdropping; the preset conditions include: at least one of: voice information recorded within a preset time range, voice information recorded within a preset location range, or voice information containing preset vocabulary.

[0018] Optionally, the anti-eavesdropping processing of the target voice information in the target audio file includes:

[0019] The target voice information in the target audio file is muted or blurred.

[0020] Optionally, it also includes:

[0021] If the target audio file is subjected to anti-eavesdropping processing during the playing of the target audio file, the target audio file contains the target speaker and the target voice information in the target audio file that complies with the anti-eavesdropping strategy.

[0022] Optionally, obtaining the target audio file includes:

[0023] Send a play request to the server;

[0024] Receive a target audio file sent by the server in response to the play request.

[0025] In a second aspect, an embodiment of the present invention provides an anti-eavesdropping processing device for an audio file, comprising:

[0026] An acquisition module is used to acquire a target audio file;

[0027] A recognition module, used to perform sound recognition on the target audio file to obtain the voiceprint features and Hertz information of each speaker contained in the target audio file;

[0028] A determination module, used for determining whether there is a target speaker according to the voiceprint characteristics and Hertz information of each speaker;

[0029] The processing module is used for performing anti-eavesdropping processing on the target audio file during the process of playing the target audio file if the target speaker exists.

[0030] Optionally, the determining module includes:

[0031] The first acquisition submodule is used to acquire the voiceprint characteristics and Hertz information of the target speaker;

[0032] A matching submodule, for matching the voiceprint features and Hertz information of each speaker with the voiceprint features and Hertz information of the target speaker respectively;

[0033] The first determination submodule is used to determine whether a target speaker exists according to the matching result.

[0034] Optionally, the processing module includes:

[0035] A second acquisition submodule is used to acquire an anti-eavesdropping strategy, where the anti-eavesdropping strategy is used to indicate target voice information that needs to be processed for anti-eavesdropping;

[0036] A second determination submodule is used to determine the target voice information in the target audio file that complies with the anti-eavesdropping strategy;

[0037] The processing submodule is used to perform anti-eavesdropping processing on the target voice information in the target audio file during the process of playing the target audio file.

[0038] Optionally, the anti-eavesdropping strategy is used to indicate that voice information that meets preset conditions is target voice information that needs to be processed for anti-eavesdropping; the preset conditions include: at least one of: voice information recorded within a preset time range, voice information recorded within a preset location range, or voice information containing preset vocabulary.

[0039] Optionally, the processing submodule includes:

[0040] The processing unit is used to perform mute processing or fuzzy sound effect processing on the target voice information in the target audio file.

[0041] Optionally, the processing module further includes:

[0042] The recording submodule is used to record the target audio file including the target speaker and the target voice information in the target audio file that complies with the anti-eavesdropping strategy if the target audio file is subjected to anti-eavesdropping processing during the playing of the target audio file.

[0043] Optionally, the acquisition module includes:

[0044] The sending submodule is used to send a playback request to the server;

[0045] The receiving submodule is used to receive the target audio file sent by the server in response to the play request.

[0046] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein when the computer program is executed by the processor, the steps of the anti-eavesdropping processing method for audio files as described above are implemented.

[0047] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the anti-eavesdropping processing method for audio files as described above are implemented.

[0048] The embodiments of the present invention include the following advantages:

[0049] The embodiment of the present invention discloses an anti-eavesdropping processing method for audio files, including obtaining a target audio file; performing sound recognition on the target audio file to obtain the voiceprint characteristics and Hertz information of each speaker contained in the target audio file; determining whether there is a target speaker based on the voiceprint characteristics and Hertz information of each speaker; if there is a target speaker, performing anti-eavesdropping processing on the target audio file during the process of playing the target audio file. Compared with traditional physical shielding or electromagnetic interference methods, this method does not require additional special hardware investment, and can achieve recognition and protection of specific user voices only through software processing, greatly improving the efficiency and accuracy of anti-eavesdropping, while reducing the cost of use. This method is applicable to various recording devices and cloud servers, is not limited by platforms or operating systems, and has the advantage of strong compatibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a flowchart of a method for anti-eavesdropping processing of an audio file provided by an embodiment of the present invention;

[0051] Figure 2 is a flowchart of another method for anti-eavesdropping processing of audio files provided by an embodiment of the present invention;

[0052] Figure 3 This is an architecture diagram of an anti-eavesdropping system for audio files provided by an embodiment of the present invention;

[0053] Figure 4 The present invention provides a structural block diagram of an anti-eavesdropping processing device for audio files. DETAILED DESCRIPTION

[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0055] One of the core concepts of the embodiments of the present invention is to propose a dual recognition mechanism that combines the user's voiceprint features and Hertz information, allowing the user to record the sound sample in advance, the system to extract the individual's unique voiceprint features and Hertz information, and the user to independently set the anti-eavesdropping permissions. When a recording audio is uploaded to the cloud server and a playback request is obtained, the system will automatically compare and confirm whether the audio contains the user's voice with the anti-eavesdropping permission set. If a matching voice is recognized, the recording will be muted when it is played, which can effectively identify and prevent the spread of illegal recordings, thereby effectively preventing privacy leaks.

[0056] Reference Figure 1 , shows a flowchart of a method for anti-eavesdropping processing of an audio file provided by an embodiment of the present invention, and the method may specifically include the following steps:

[0057] Anti-eavesdropping is an important measure to protect sensitive information from being obtained by unauthorized third parties, and it plays a vital role in enterprises, government agencies and personal life. With the advancement of technology, eavesdropping methods are becoming increasingly diverse and complex, so it is necessary to adopt a multi-level and multi-dimensional security protection strategy to deal with potential threats. Anti-eavesdropping is not just dependent on a certain technology or product, but a comprehensive project involving hardware deployment, software support, personnel management and legal compliance. According to the specific application scenario and needs, the most suitable anti-eavesdropping solution is selected to ensure the security and privacy of information exchange. The present invention performs a series of anti-eavesdropping processing on the audio files that have been secretly recorded to protect information security.

[0058] Step 101, obtaining a target audio file;

[0059] For audio files that have been recorded, especially when dealing with sensitive information or conducting critical communications, a variety of technical and management measures can be taken to ensure the security and integrity of the recordings and prevent unauthorized access, tampering or leakage.

[0060] First, from the perspective of physical security: access rights can be restricted to ensure that only authorized personnel can access recording equipment and storage media. Access control systems, password locks, and other means can be used to protect the rooms or cabinets where recording equipment is stored. Environmental control can also be performed, and noise shielding and other physical isolation measures can be implemented in the recording environment to prevent external monitoring devices from capturing audio signals.

[0061] Technical protection: encryption can be performed, such as end-to-end encryption (E2EE), which encrypts the entire transmission process of the recording data from the collection point to the final storage location to ensure that the content cannot be interpreted even if it is intercepted during transmission on the network; or static data encryption, when the recording file is stored on a hard disk or other media, it is encrypted using a strong encryption algorithm (such as AES-256) to prevent unauthorized access. Digital watermarks can also be used, such as embedded watermarks, to embed invisible but verifiable digital watermarks in the recording to track any illegal copying or distribution and as evidence to prove the authenticity of the recording. Integrity verification can also be performed, such as hash verification, to generate a unique hash value for each recording file and save it with the file. Each time the file is read, the hash value is recalculated and compared with the original value to confirm whether the file has been tampered with. You can also start with access logs, such as audit trails, to record all operation logs that access recording files, including who accessed which files and when, and what operations were performed, which helps to discover potential security threats and provide a basis for post-review.

[0062] Dedicated hardware and software: You can use professional-grade recording equipment and choose professional recording equipment with built-in security features, such as products that support hardware encryption, remote wipe, etc. Or secure recording software, use certified secure recording applications, which usually provide higher levels of encryption, access control and management features.

[0063] There are also management and policy aspects: formulate clear security policies, define which recordings need special protection, how to handle these recordings, and what penalties will be imposed for violations. Ensure that all relevant personnel understand and abide by these rules; regularly provide security awareness education and technical training to employees involved in recording work so that they understand how to properly operate the equipment and management system; third-party assessments can also be conducted, inviting independent security experts or institutions to regularly check the security of the recording system, find loopholes and make improvement suggestions.

[0064] Step 102, performing sound recognition on the target audio file to obtain the voiceprint features and Hertz information of each speaker contained in the target audio file;

[0065] Voiceprint features refer to information that can reflect the unique identity of the speaker. They can be extracted from voice signals and used to distinguish different individuals. Hertz information (frequency information) refers to the frequency component in the audio signal, which is one of the important parameters for describing sound characteristics.

[0066] Identifying the voiceprint features and Hertz information (i.e., frequency information) of people in audio files has a wide range of applications in many fields. Voiceprint recognition, as a biometric technology, confirms the identity of the speaker by extracting unique features from voice signals. This technology has a high recognition accuracy, and because each person's pronunciation organ structure is different, the voiceprint features are unique to different speakers. Compared with traditional password or card authentication methods, voiceprint recognition does not require users to remember complex passwords or carry additional identity documents, and is more convenient and quick to use.

[0067] Identifying the voiceprint features and hertz information (i.e. frequency information) of people in audio files is an important part of modern speech processing technology. This process involves multiple steps and technologies, including but not limited to preprocessing, feature extraction, pattern matching, etc.

[0068] Before starting the analysis, it is usually necessary to perform a series of preprocessing operations on the original audio signal to ensure the effectiveness and accuracy of subsequent processing. These steps may include denoising, framing, windowing, etc. Denoising refers to removing background noise and other non-target sounds in order to capture the speaker's voice more clearly; framing refers to dividing the continuous audio stream into short time periods (frames), each frame is usually 20-30 milliseconds, and there is a certain overlap between frames; windowing is to reduce the discontinuity between adjacent frames, and a window function (such as Hamming window or Hanning window) is applied to each frame, so that the sound signal outside the window gradually decays to zero.

[0069] Common voiceprint feature extraction methods include MFCC (Mel-Frequency Cepstral Coefficients), PLP (Perceptual Linear Prediction), etc. MFCC is a widely used voiceprint feature extraction method that simulates the working principle of the human auditory system, converts the time domain signal to the frequency domain through a series of transformations (such as Fourier transform, Mel filter bank, discrete cosine transform), and finally obtains a set of coefficients that can represent the characteristics of the speaker. PLP is based on the human ear perception model and takes into account the perceptual characteristics of the sound, such as loudness, pitch, etc., thereby improving the robustness of the feature.

[0070] For feature extraction of Hertz information, the sampling rate determines the highest frequency that can be captured from the signal, which is called the Nyquist limit, which is exactly half of the sampling rate. For example, the sampling rate of CD quality audio is 44,100Hz, which means that frequencies up to 22,050Hz can be accurately represented. Spectral analysis is performed by performing a fast Fourier transform (FFT) on the audio signal, and its spectrum can be obtained, showing the energy distribution of each frequency component. This is very helpful for understanding the audio content, especially in music or speech analysis. Considering that the human ear is more sensitive to low frequency bands, the linear frequency scale is first converted to the Mel frequency scale when calculating MFCC, so that the features can better match the human hearing characteristics.

[0071] Once enough voiceprint features are extracted, various algorithms can be used for pattern matching to determine the identity of the speaker. Common algorithms include but are not limited to support vector machines (SVM), Gaussian mixture models (GMM), deep neural networks (DNN), etc. 1:1 recognition refers to confirming whether a speech to be recognized comes from the target speaker it claims to be. This type of recognition is usually used in electronic payment, smart hardware and other fields; 1:N recognition aims to find one or several candidates that are most similar to the test sample from a large number of known speakers.

[0072] In the embodiment of the present invention, the audio file is identified to obtain the voiceprint characteristics and Hertz information of each speaker, so as to determine the uniqueness of the task corresponding to the voice information in the audio file, accurately lock sensitive persons, and ensure that important information is not omitted and protected.

[0073] Step 103, determining whether there is a target speaker according to the voiceprint features and Hertz information of each speaker;

[0074] Voiceprint recognition is a biometric technology that confirms the identity of a speaker by extracting unique features from voice signals. The principle of this technology is based on the physiological and physical process of human speech production. When a person speaks, the differences in size and shape of the vocal organs such as the tongue, teeth, larynx, lungs, and nasal cavity make each person's voiceprint unique. This uniqueness is mainly reflected in the spectral characteristics of the sound, including fundamental frequency, formant frequency, timbre and other parameters.

[0075] The reason why voiceprint features are unique to different speakers is that the vocal organs used by each person when speaking have individual differences in size and shape, and these differences lead to each person's voice waveform having its own unique pattern. In addition, factors such as age, personality, and language habits will also affect a person's pronunciation, further increasing the uniqueness of voiceprints. Even when imitating others, the imitator cannot completely copy all the pronunciation details of the target object, so they can still be distinguished from the voiceprint feature analysis.

[0076] Moreover, voiceprint recognition technology has achieved a very high accuracy rate. In some application scenarios, if the user reads out the content according to the pre-specified text, the accuracy rate of voiceprint recognition can reach 99.8%; even if the voice is spoken randomly, the accuracy rate can reach 99.1%. Such high accuracy makes voiceprint recognition a reliable means of identity authentication, especially when remote identity authentication is required. For example, in the financial field, banks can use voiceprint recognition technology to provide customers with a more convenient way of identity authentication and improve transaction security.

[0077] Compared with traditional password or card authentication methods, voiceprint recognition does not require users to remember complex passwords or carry additional identity documents, which not only improves the convenience of use, but also reduces the trouble caused by forgetting passwords or losing cards. Only a microphone or a phone and mobile phone can collect user voiceprint feature information to complete identity authentication.

[0078] In this embodiment, the voiceprint features and Hertz information of each speaker in the audio file can ensure high accuracy of the matching result and accurately verify whether the target speaker exists.

[0079] Step 104: If the target speaker exists, anti-eavesdropping processing is performed on the target audio file during the process of playing the target audio file.

[0080] If the voiceprint features and Hertz information identified in the acquired audio file successfully match those of the target speaker, that is, it is determined that the audio file contains the speech of the target speaker, then the sensitive condition is triggered, indicating that the content recorded in the audio file is sensitive content that is not allowed to be recorded. In this case, the audio file will be processed for anti-eavesdropping, so that even if the secret recorder records the sensitive information, he will not be able to play it again or obtain the important content in the recording, thus cutting off the possibility of secondary dissemination of sensitive information.

[0081] In some embodiments of the present invention, the anti-eavesdropping system can be installed on the communication software of all users. By turning on the anti-eavesdropping system, the recorded files can be managed and controlled. Once a sensitive specific user pre-set on the server is matched, the audio file will be processed, making it impossible to read useful content in the secretly recorded audio file and making it meaningless, thereby achieving the purpose of preventing eavesdropping.

[0082] The embodiment of the present invention discloses an anti-eavesdropping processing method for audio files, including obtaining a target audio file; performing sound recognition on the target audio file to obtain the voiceprint characteristics and Hertz information of each speaker contained in the target audio file; determining whether there is a target speaker based on the voiceprint characteristics and Hertz information of each speaker; if there is a target speaker, performing anti-eavesdropping processing on the target audio file during the process of playing the target audio file. Compared with traditional physical shielding or electromagnetic interference methods, this method does not require additional special hardware investment, and can achieve recognition and protection of specific user voices only through software processing, greatly improving the efficiency and accuracy of anti-eavesdropping, while reducing the cost of use. This method is applicable to various recording devices and cloud servers, is not limited by platforms or operating systems, and has the advantage of strong compatibility.

[0083] Reference Figure 2 , shows a flowchart of another method for anti-eavesdropping processing of audio files provided by an embodiment of the present invention, the method may specifically include the following steps:

[0084] Step 201, obtaining a target audio file;

[0085] In some embodiments of the present invention, step 201 includes:

[0086] Sub-step S11, sending a play request to the server;

[0087] The audio files and the target speaker's voiceprint features and Hertz information are not necessarily retrieved and used immediately after they are acquired. Instead, they can be uploaded to a cloud server or other storage media for storage. When they are retrieved and used, the terminal device sends a play request to the server to obtain the audio file. This saves the storage space of the terminal device, extends the storage time, and is not restricted by the terminal device. Even if a terminal device is changed, its functionality is not affected.

[0088] Sub-step S12, receiving the target audio file sent by the server in response to the play request.

[0089] Since the server stores multiple audio files and voiceprint features and Hertz information of multiple sound samples, after sending a play request to the server, the server needs to call out the corresponding target audio file for the play request and send it, and at the same time receive the anti-eavesdropping processing strategy required for the audio file for subsequent anti-eavesdropping specific processing.

[0090] Step 202, performing sound recognition on the target audio file to obtain the voiceprint features and Hertz information of each speaker contained in the target audio file;

[0091] Hertz is a unit of measurement for frequency, which indicates the number of times a periodic event occurs per second. Specifically for sound waves, it refers to the number of complete waveforms that pass a given point per second. The frequencies of sound waves within the audible range of humans are approximately between 20Hz and 20,000Hz. Different sounds have different frequency components, and these frequency characteristics are very important for distinguishing various types of sounds and performing speech recognition.

[0092] In sound recognition technology, frequency information is widely used in many aspects:

[0093] Speech feature extraction: In order to effectively process speech signals, it is first necessary to divide the continuous audio stream into short time periods and perform Fourier transform (such as short-time Fourier transform STFT) on them to obtain the spectrum information in each time period. This step helps to reveal the frequency composition of the audio and its changes over time. After that, the Mel filter bank will be further applied to simulate the changes in the sensitivity of the human ear to different frequencies and generate a feature representation that is more in line with human auditory characteristics - Mel frequency cepstral coefficients (MFCC). This method can not only capture the key frequency characteristics of speech, but also reduce computational complexity.

[0094] Noise suppression: Environmental noise often contains a large number of non-target frequency components, which may interfere with the accuracy of the speech recognition system. Therefore, in practical applications, methods such as adaptive filtering are often used to filter out frequency components that do not belong to the speaker's voice, thereby improving the recognition effect. In addition, Hertz information can even be used to analyze people's emotional state, cross-language recognition, etc.

[0095] In this embodiment, sound recognition is performed on the audio file to obtain the voiceprint features and Hertz information of each speaker contained therein, which can facilitate subsequent comparison and matching with high accuracy.

[0096] Step 203, obtaining the voiceprint features and Hertz information of the target speaker; matching the voiceprint features and Hertz information of each speaker with the voiceprint features and Hertz information of the target speaker; and determining whether there is a target speaker based on the matching results;

[0097] The user can upload a sensitive target speaker's voice sample in advance, perform voice recognition on the voice sample, obtain the target speaker's voiceprint features and hertz information and store them for subsequent comparison. If the voiceprint features and hertz information of each speaker identified in the audio file match the pre-stored target speaker's voiceprint features and hertz information, it is determined that the target speaker exists in the recording information of the audio file.

[0098] Step 204: If the target speaker exists, anti-eavesdropping processing is performed on the target audio file during the process of playing the target audio file.

[0099] In some embodiments of the present invention, step 204 includes:

[0100] Sub-step S21, obtaining an anti-eavesdropping strategy, wherein the anti-eavesdropping strategy is used to indicate target voice information that needs to be processed for anti-eavesdropping;

[0101] If the target speaker is matched, it means that the audio file contains sensitive information and is not allowed to be played or transmitted. The audio file name needs to be processed for anti-eavesdropping to prevent the leakage of the recording content. The specific anti-eavesdropping processing method is executed according to the anti-eavesdropping strategy received from the server.

[0102] In some embodiments of the present invention, it further comprises:

[0103] The anti-eavesdropping strategy is used to indicate that voice information that meets preset conditions is the target voice information that needs to be processed for anti-eavesdropping; the preset conditions include: at least one of: voice information recorded within a preset time range, voice information recorded within a preset location range, or voice information containing preset vocabulary.

[0104] The anti-eavesdropping strategy can be set by the user in advance and uploaded to the server for storage. The user can independently choose how to handle sensitive audio information to prevent sensitive content leakage. The implementation object of the anti-eavesdropping strategy is the target voice information in the audio file, and the target voice is selected based on whether it meets the preset conditions. If the preset conditions are met, it means that the voice information is the target voice information and needs to be processed for anti-eavesdropping.

[0105] The preset conditions can be set by the user. It can be that all audio files recorded within a certain time period are judged as target voice files, or all audio files recorded at a certain location are judged as target voice files, or audio files containing certain predetermined sensitive words are judged as target voice files and need to be anti-eavesdropping processed. The specific preset conditions can also be set according to user needs, and multiple preset conditions can be set to improve the level of confidentiality.

[0106] In some embodiments of the present invention, for a specific location, it can be preset that when the recording location of the audio file is in a place where confidential content is discussed, such as a conference room or a certain office; or the recording location is a relatively private place where there may be a strong sense of personal privacy, such as a private club, hotel room, high-end restaurant box, private residence bedroom, etc., the content of the audio file will be determined as the target voice information for anti-eavesdropping processing.

[0107] For a specific time, for example, if a company holds a highly confidential business meeting from 9:00 to 11:00 in the morning, 9:00 to 11:00 can be set as the preset condition, and the audio files recorded during this period are determined as target voice information; or for a hotel, after 10:00 p.m. is a more private time, and personal privacy does not want to be exposed, the time from 10:00 p.m. to 6:00 a.m. can be set as the preset condition, and the audio files during this period are determined as target voice information to be processed for anti-eavesdropping.

[0108] For specific words, users can also make preset settings. For example, words such as "strategy", "company", "document", "plan", and "technology" can be set as sensitive words. Audio files involving preset words will be identified as target voice information, and anti-eavesdropping processing will be performed later.

[0109] Or you can set multiple preset conditions at the same time, for example, from 9 to 10 in the morning, all audio files related to "technology" and "plan" in a certain company will be protected from eavesdropping. The specific preset conditions can be freely set by the user according to their needs.

[0110] Sub-step S22, determining the target voice information in the target audio file that complies with the anti-eavesdropping strategy;

[0111] According to the anti-eavesdropping strategy determined by the preset conditions, the target voice information to be processed in the target audio file is determined, and the part to be processed for anti-eavesdropping is determined.

[0112] Sub-step S23, during the process of playing the target audio file, anti-eavesdropping processing is performed on the target voice information in the target audio file.

[0113] In some embodiments of the present invention, the sub-step S23 includes:

[0114] The target voice information in the target audio file is muted or blurred.

[0115] After determining the target voice information portion to be processed in the audio file, the target voice information portion is muted directly or replaced with a fuzzy sound effect, so that the played audio file cannot obtain useful information to achieve the purpose of preventing eavesdropping. Muting means directly silencing the target voice information portion, eliminating this voice information, deleting and discarding it; fuzzy sound effect replacement means completely replacing this portion of the target voice information with a sound effect such as "dong dong dong", and deleting and discarding the original target voice information.

[0116] In some embodiments of the present invention, it further comprises:

[0117] If the target audio file is subjected to anti-eavesdropping processing during the playing of the target audio file, the target audio file contains the target speaker and the target voice information in the target audio file that complies with the anti-eavesdropping strategy.

[0118] If the audio file is processed for anti-eavesdropping, it means that the audio file is a sensitive file and recording is not allowed. It also indicates that important information has been secretly recorded and there is a risk of information leakage. Therefore, the target speaker, recording time, recording location and other target voice information that meets the anti-eavesdropping strategy matched in this sensitive audio file are recorded to facilitate subsequent analysis and management, including but not limited to stopping the discussion of sensitive topics, notifying relevant departments, and taking technical means to eliminate hidden dangers.

[0119] The embodiment of the present invention discloses an anti-eavesdropping processing method for audio files, including obtaining a target audio file; performing sound recognition on the target audio file to obtain the voiceprint characteristics and Hertz information of each speaker contained in the target audio file; determining whether there is a target speaker based on the voiceprint characteristics and Hertz information of each speaker; if there is a target speaker, performing anti-eavesdropping processing on the target audio file during the process of playing the target audio file. Compared with traditional physical shielding or electromagnetic interference methods, this method does not require additional special hardware investment, and can achieve recognition and protection of specific user voices only through software processing, greatly improving the efficiency and accuracy of anti-eavesdropping, while reducing the cost of use. This method is applicable to various recording devices and cloud servers, is not limited by platforms or operating systems, and has the advantage of strong compatibility.

[0120] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0121] Reference Figure 3 , shows an architecture diagram of an anti-eavesdropping system for audio files provided by an embodiment of the present invention, which is as follows:

[0122] The system includes a cloud server and at least one terminal device. At least one virtual machine runs on the cloud server, and the terminal device is connected to the corresponding virtual machine. The terminal device includes an acquisition module, a processing module and a receiving module, wherein the acquisition module and the receiving module are both connected to the processing module, and the receiving module is connected to the corresponding virtual machine in the cloud server for communication. When the receiving module of the terminal device obtains a play request, it sends the play request to the virtual machine. The virtual machine parses the play request and sends an audio file to the receiving module of the terminal device. The receiving module then sends the received audio file to the processing module. When the processing module receives the audio file, it starts the anti-eavesdropping process. It is mainly divided into the following steps:

[0123] Specific user voice sample recording and uploading: Users (can be single or multiple) upload their voices to the cloud or other storage media through user terminals (such as mobile phones, computers, etc.). This is the basis for subsequent voiceprint recognition and anti-eavesdropping processing, and this step is completed by the acquisition module.

[0124] Voiceprint feature and Hertz information recognition: The system uses advanced voiceprint recognition technology to extract and store the personal voiceprint features of a specific user. At the same time, the Hertz information of the sound is also recorded to provide more sophisticated recognition capabilities. This step is completed by the processing module.

[0125] Anti-leakage requirements: Users can set specific anti-eavesdropping permissions according to their needs, such as protecting only recordings of personal voice messages within a specific time range, location range, or containing specific vocabulary and sentences. When a terminal device (such as a smartphone, smart TV, etc.) receives a request to play a certain audio / video, the system will respond according to the anti-leakage requirements raised by the specific user. This step is set by the user on the mobile device and uploaded to the virtual machine.

[0126] Get a play request: The play request is used to request to play an audio file. When the user wants to listen to an audio file, he can select the audio file he wants to listen to, and then click the play button, so that the terminal device obtains the play request. This step is received by the receiving module.

[0127] Comparison of voiceprint and Hz information and silencing: When the terminal device receives a playback request, the system will compare the sound information in the audio / video to be played with the voiceprint characteristics and Hz information of the specific user. If matching sound information is found, a certain section of the sound in the file will be muted according to the user's anti-leakage requirements to protect the user's privacy and security. For example, for sensitive personal voice information, the system can replace it with silent or blurred sound effects. If no matching sound information is found, the system does not need to perform any processing, and the audio can be played normally. This step is completed by the processing module. At the same time, the system can also record detailed information of the anti-eavesdropping incident, such as time, location, users involved, etc., for subsequent analysis and management.

[0128] For example, in this embodiment, corporate executive A discussed sensitive business strategies at an internal company meeting. To maintain confidentiality, he enabled the anti-eavesdropping recording system, which recorded the voice characteristics of the main speaker and set anti-eavesdropping permissions for sensitive words at a certain time period and location. An employee tried to record illegally, but the system detected sensitive words at the time period and location and automatically muted them to ensure the security of all meeting content. The anti-eavesdropping function can also be integrated into existing communication software to improve user convenience and usage experience. This method is applicable to various recording devices and cloud servers, is not limited by the platform or operating system, and has the advantage of strong compatibility.

[0129] Reference Figure 4 , shows a structural block diagram of an anti-eavesdropping processing device for audio files provided by an embodiment of the present invention, which may specifically include the following modules:

[0130] An acquisition module 401 is used to acquire a target audio file;

[0131] The recognition module 402 is used to perform sound recognition on the target audio file to obtain the voiceprint features and Hertz information of each speaker contained in the target audio file;

[0132] A determination module 403 is used to determine whether there is a target speaker according to the voiceprint features and Hertz information of each speaker;

[0133] The processing module 404 is configured to perform anti-eavesdropping processing on the target audio file during the process of playing the target audio file if the target speaker exists.

[0134] In some embodiments of the present invention, the determining module includes:

[0135] The first acquisition submodule is used to acquire the voiceprint characteristics and Hertz information of the target speaker;

[0136] A matching submodule, for matching the voiceprint features and Hertz information of each speaker with the voiceprint features and Hertz information of the target speaker respectively;

[0137] The first determination submodule is used to determine whether a target speaker exists according to the matching result.

[0138] In some embodiments of the present invention, the processing module includes:

[0139] A second acquisition submodule is used to acquire an anti-eavesdropping strategy, where the anti-eavesdropping strategy is used to indicate target voice information that needs to be processed for anti-eavesdropping;

[0140] A second determination submodule is used to determine the target voice information in the target audio file that complies with the anti-eavesdropping strategy;

[0141] The processing submodule is used to perform anti-eavesdropping processing on the target voice information in the target audio file during the process of playing the target audio file.

[0142] In some embodiments of the present invention, the anti-eavesdropping strategy is used to indicate that voice information that meets preset conditions is target voice information that needs to be processed for anti-eavesdropping; the preset conditions include: at least one of: voice information recorded within a preset time range, voice information recorded within a preset location range, or voice information containing preset vocabulary.

[0143] In some embodiments of the present invention, the processing submodule includes:

[0144] The processing unit is used to perform mute processing or fuzzy sound effect processing on the target voice information in the target audio file.

[0145] In some embodiments of the present invention, the processing module further includes:

[0146] The recording submodule is used to record the target audio file including the target speaker and the target voice information in the target audio file that complies with the anti-eavesdropping strategy if the target audio file is subjected to anti-eavesdropping processing during the playing of the target audio file.

[0147] In some embodiments of the present invention, the acquisition module includes:

[0148] The sending submodule is used to send a playback request to the server;

[0149] The receiving submodule is used to receive the target audio file sent by the server in response to the play request.

[0150] The embodiment of the present invention obtains a target audio file; performs sound recognition on the target audio file to obtain the voiceprint characteristics and Hertz information of each speaker contained in the target audio file; determines whether a target speaker exists based on the voiceprint characteristics and Hertz information of each speaker; if a target speaker exists, performs anti-eavesdropping processing on the target audio file during the playback of the target audio file. Compared with traditional physical shielding or electromagnetic interference methods, this method does not require additional special hardware investment, and can achieve recognition and protection of specific user voices only through software processing, greatly improving the efficiency and accuracy of anti-eavesdropping while reducing the cost of use. This method is applicable to various recording devices and cloud servers, is not limited by the platform or operating system, and has the advantage of strong compatibility.

[0151] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0152] An embodiment of the present invention further provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, the various processes of the above-mentioned anti-eavesdropping processing method embodiment of the audio file are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0153] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, each process of the above-mentioned anti-eavesdropping processing method embodiment of the audio file is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0154] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0155] It will be appreciated by those skilled in the art that the embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the embodiments of the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0156] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0157] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0158] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0159] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0160] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0161] The above is a detailed introduction to an audio file anti-eavesdropping processing method and an audio file anti-eavesdropping processing device provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for a person skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for preventing eavesdropping of audio files, characterized in that: include: Get the target audio file; Performing sound recognition on the target audio file to obtain the voiceprint features and Hertz information of each speaker contained in the target audio file; Determining whether there is a target speaker based on the voiceprint features and Hertz information of each speaker; If the target speaker exists, anti-eavesdropping processing is performed on the target audio file during the process of playing the target audio file.

2. The anti-eavesdropping processing method for audio files according to claim 1, characterized in that: The determining whether there is a target speaker according to the voiceprint features and Hertz information of each speaker includes: Obtain the target speaker's voiceprint characteristics and Hertz information; Matching the voiceprint features and Hertz information of each speaker with the voiceprint features and Hertz information of the target speaker respectively; Based on the matching results, determine whether the target speaker exists.

3. The anti-eavesdropping processing method for audio files according to claim 1, characterized in that: The anti-eavesdropping process on the target audio file during the process of playing the target audio file includes: Acquire an anti-eavesdropping strategy, where the anti-eavesdropping strategy is used to indicate target voice information that needs to be processed for anti-eavesdropping; Determine the target voice information in the target audio file that complies with the anti-eavesdropping strategy; During the process of playing the target audio file, anti-eavesdropping processing is performed on the target voice information in the target audio file.

4. The method for anti-eavesdropping of audio files according to claim 3, characterized in that: The anti-eavesdropping strategy is used to indicate that the voice information that meets the preset conditions is the target voice information that needs to be processed for anti-eavesdropping; The preset condition includes at least one of the following conditions: voice information recorded within a preset time range, voice information recorded within a preset location range, or voice information containing preset vocabulary.

5. The anti-eavesdropping processing method for audio files according to claim 3, characterized in that: The anti-eavesdropping processing of the target voice information in the target audio file includes: The target voice information in the target audio file is muted or blurred.

6. The anti-eavesdropping processing method for audio files according to claim 3, characterized in that: Also includes: If the target audio file is subjected to anti-eavesdropping processing during the playing of the target audio file, the target audio file contains the target speaker and the target voice information in the target audio file that complies with the anti-eavesdropping strategy.

7. The method for anti-eavesdropping of audio files according to claim 1, characterized in that: The step of obtaining the target audio file comprises: Send a play request to the server; Receive a target audio file sent by the server in response to the play request.

8. An anti-eavesdropping processing device for audio files, characterized in that: include: An acquisition module is used to acquire a target audio file; A recognition module, used to perform sound recognition on the target audio file to obtain the voiceprint features and Hertz information of each speaker contained in the target audio file; A determination module, used for determining whether there is a target speaker according to the voiceprint characteristics and Hertz information of each speaker; The processing module is used for performing anti-eavesdropping processing on the target audio file during the process of playing the target audio file if the target speaker exists.

9. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein when the computer program is executed by the processor, the steps of the anti-eavesdropping processing method for audio files according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the anti-eavesdropping processing method for audio files according to any one of claims 1 to 7 are implemented.

Images