Network friend-making system based on audio information hiding and audio information hiding method

Abstract

The invention provides a network friend-making system based on audio information hiding and an audio information hiding method, and belongs to the field of audio information hiding, and the system comprises a secure chat module which is configured to encrypt to-be-embedded data to original digital audio by employing a reversible data hiding algorithm, form first audio data, and carry out communication by employing the first audio data; the hidden help seeking module is configured to set hidden alarm information in the to-be-embedded data, and the network friend-making system monitors the to-be-embedded information in real time and gives an alarm after monitoring the hidden alarm information; the sensitive word detection and automatic alarm module is configured for the network friend-making system to monitor data to be embedded in real time and automatically trigger an alarm mechanism after monitoring that the data to be embedded contains sensitive words; and the multi-channel audio processing module is configured to provide additional information hiding capacity for the network friend-making system. According to the method, safe hiding and accurate extraction of the user information are realized on the premise that the audio quality is not damaged.

Description

Technical Field

[0001] The present invention relates to the technical field of audio information hiding, and in particular to a network friendship system solution based on audio information hiding. Background Art

[0002] With the rapid development of Internet technology, online dating has become an important part of modern people's social life. However, traditional online dating platforms face many challenges in terms of user privacy protection and information security. Although existing audio data hiding technologies (such as digital watermarking, information steganography, etc.) can embed information in audio, they often affect the audio quality or pose security risks. At the same time, with the frequent occurrence of online fraud and personal information leakage, users are increasingly in need of safe and reliable dating platforms.

[0003] In the existing technology, audio hiding technology is easy to be cracked or tampered with, which leads to the easy leakage of personal privacy information. The information hiding capacity is also limited, and it cannot support multi-channel processing and multiple audio formats. In addition, when there is noise in the use environment, the audio quality is easily affected.

[0004] Therefore, finding a method that can both safely hide data while ensuring audio quality and adapt to different usage scenarios is a technical problem that needs to be solved urgently by those skilled in the art. Summary of the invention

[0005] The present invention provides a network friendship system solution based on audio information hiding, which is used to solve the defects in the prior art that data cannot be safely hidden and the audio quality of the hidden information is easily affected, and realizes the safe hiding and accurate extraction of user information without damaging the audio quality, while supporting multiple audio formats and multi-channel processing.

[0006] The present invention provides an online friendship system based on audio information hiding, comprising:

[0007] A secure chat module configured to encrypt the data to be embedded into the original digital audio using a reversible data hiding algorithm to form first audio data, and use the first audio data for communication;

[0008] A hidden help-seeking module is configured to set hidden alarm information in the data to be embedded, and the online dating system monitors the embedded information in real time and issues an alarm after detecting the hidden alarm information;

[0009] The sensitive word detection and automatic alarm module is configured so that the online dating system monitors the data to be embedded in real time and automatically triggers the alarm mechanism after detecting that the data to be embedded contains sensitive words;

[0010] A multi-channel audio processing module is configured to provide additional information hiding capacity for an online dating system.

[0011] According to an online friendship system based on audio information hiding provided by the present invention, the secure chat module specifically includes:

[0012] An audio signal processing unit configured to perform audio signal analysis on the original digital audio to determine a time point to be embedded, a frequency position to be embedded, and a channel to be embedded in the original digital audio; wherein the audio signal analysis includes timestamp analysis, multi-channel analysis, time domain analysis, and frequency domain analysis;

[0013] An information encryption unit configured to convert the data to be embedded into binary and form an encrypted information matrix;

[0014] The information hiding unit is configured to generate a spectrogram from the original digital audio, and use the spectrogram as a reference to embed the encrypted information matrix into the original digital audio based on the time point to be embedded, the frequency position to be embedded, and the channel to be embedded of the original digital audio, so as to obtain a first digital audio; the first audio includes the original digital audio data and the encrypted data;

[0015] The information decryption unit is configured to receive the first digital audio at the receiving end, extract the encrypted data based on the linear code theory, and post-process the encrypted audio data to obtain the data to be embedded.

[0016] According to an online friendship system based on audio information hiding provided by the present invention, the steps of the information hiding unit are as follows:

[0017] Generate a spectrogram from the original digital audio, and calculate the energy value of each time-frequency unit in the spectrogram according to a noise threshold;

[0018] Detect the intensity of environmental noise, dynamically adjust the information embedding parameters of different scenes according to the noise level, and calculate the embedding energy thresholds of different scenes based on the information embedding parameters of different scenes;

[0019] Determine the energy area to be embedded based on the energy value of each time-frequency unit and the embedding energy threshold of different scenes;

[0020] The encrypted information matrix is ​​secretly scrambled to generate a carrier matrix, and the carrier matrix is ​​information encoded to obtain an encoding matrix;

[0021] Based on the time point to be embedded, the frequency position to be embedded, the embeddable channel and the energy area to be embedded, a reversible data hiding algorithm is used to embed the coding matrix into the energy area to be embedded, and audio format packaging is performed to obtain the first digital audio.

[0022] According to an online dating system based on audio information hiding provided by the present invention, the information embedding parameters based on different scenes calculate the embedding energy thresholds of different scenes, and the calculation formula is:

[0023]

[0024] Among them, α chat represents the information embedding parameter in the chat scenario, α alarm represents the information embedding parameter in the emergency alarm scenario, α monitor represents the information embedding parameter in the supervision scenario, k represents a constant, L represents the chat sensitivity level, m represents a constant, U represents the urgency level, n represents a constant, C represents the privacy protection level, and S represents the monitoring sensitivity level.

[0025] According to an online friendship system based on audio information hiding provided by the present invention, the time domain analysis specifically includes:

[0026] Calculate the average energy of the audio signal in the original digital audio;

[0027] Divide the audio signal of the original data audio into multiple time windows of length t, and calculate the energy of each time window;

[0028] Identify the current scene, set energy adjustment factors according to different scenes, and calculate energy thresholds based on the energy adjustment factors;

[0029] According to the energy and energy threshold of each time window, the position and range of the low-energy segment and the position and range of the silent segment are determined, and the low-energy segment is recorded as the time point to be embedded.

[0030] According to an online friendship system based on audio information hiding provided by the present invention, the energy threshold is calculated based on the energy adjustment factor, specifically including:

[0031] θ chat =0.3×E avg ×(1-β);

[0032] θ alarm =0.3×E avg ×(1+γ);

[0033] θ monitor =0.3×E avg ×(1+ω×S);

[0034] Among them, θ chat represents the energy threshold in the chat scenario, θ alarm represents the energy threshold in the alarm scenario, θ monitor Indicates the energy threshold of the automatically built scene, E avgrepresents the average energy of the audio signal, β represents the adjustment factor, γ represents the enhancement factor, ω represents the real-time adjustment factor, and S represents the monitoring sensitivity level.

[0035] According to an online friendship system based on audio information hiding provided by the present invention, the steps of the information encryption unit are as follows:

[0036] Convert the data to be embedded into ASCII code, and convert the ASCII code into 8-bit binary code to obtain binary data of the data to be embedded;

[0037] Setting an initial matrix, serializing the binary data in sequence, and filling the serially serialized binary data into the initial matrix to obtain an encrypted information matrix;

[0038] or,

[0039] Extract the RGB value of each pixel of the data to be embedded, and convert the RGB value of each pixel into 8-bit binary to obtain binary data of the data to be embedded;

[0040] An initial matrix is ​​set, the binary data is concatenated in sequence, and the binary data is filled into the initial matrix to obtain an encrypted information matrix.

[0041] The present invention also provides an audio information hiding method, which is applied to the above-mentioned online dating system, and comprises the following steps:

[0042] S1. Perform timestamp analysis on the original audio signal, select feature words for timestamp marking according to different security requirements, and obtain keyword timestamps;

[0043] S2, performing multi-channel analysis on the original audio signal, and determining the embeddable channels according to the total number of channels in the original digital audio and the amount of data that can be hidden;

[0044] S3, performing time domain analysis on the original audio data, calculating the energy of the original audio data according to the energy adjustment factor, and determining the energy area to be embedded;

[0045] S4, performing frequency domain analysis on the original audio data, and determining the frequency position to be embedded according to the information embedding parameter;

[0046] S5, converting the data to be embedded into binary and forming an encrypted information matrix;

[0047] S6. Based on the keyword timestamp, the embeddable channel, the energy area to be embedded, and the frequency position to be embedded, a reversible data hiding algorithm is used to add the encrypted information matrix to the energy area to be embedded to obtain the first audio data.

[0048] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the audio information hiding methods described above is implemented.

[0049] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the audio information hiding method described in any one of the above methods is implemented.

[0050] The network dating system solution based on audio information hiding provided by the present invention combines reversible audio data hiding technology with the network dating system, thereby realizing the safe hiding and accurate extraction of user information without damaging the audio quality, and supports multiple audio formats and multi-channel processing, providing a safe and reliable new solution for network dating, which not only protects user privacy but also improves the security and fun of social interaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 It is a flow chart of the network friendship system provided by the present invention;

[0053] Figure 2 It is a flow chart of the data embedding and extraction steps of the online dating system provided by the present invention;

[0054] Figure 3 It is a spectrogram of an embodiment of the network friendship system provided by the present invention;

[0055] Figure 4 It is a functional diagram of the online friendship system provided by the present invention;

[0056] Figure 5 This is one of the flow charts of the audio information hiding method provided by the present invention;

[0057] Figure 6 This is the second flow chart of the audio information hiding method provided by the present invention;

[0058] Figure 7 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in combination with the drawings and embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] like Figure 1 As shown, an embodiment of the present invention provides an online friendship system based on audio information hiding, comprising:

[0061] A secure chat module configured to encrypt the data to be embedded into the original digital audio using a reversible data hiding algorithm to form first audio data, and use the first audio data for communication;

[0062] A hidden help-seeking module is configured to set hidden alarm information in the data to be embedded, and the online dating system monitors the embedded information in real time and issues an alarm after detecting the hidden alarm information;

[0063] The sensitive word detection and automatic alarm module is configured so that the online dating system monitors the data to be embedded in real time and automatically triggers the alarm mechanism after detecting that the data to be embedded contains sensitive words;

[0064] A multi-channel audio processing module is configured to provide additional information hiding capacity for an online dating system.

[0065] like Figure 2 As shown, in one embodiment of the present invention, the secure chat module specifically includes:

[0066] An audio signal processing unit configured to perform audio signal analysis on the original digital audio to determine the time point to be embedded, the frequency position to be embedded, and the channel to be embedded in the original digital audio; wherein the audio signal analysis includes timestamp analysis, multi-channel analysis, time domain analysis, and frequency domain analysis;

[0067] An information encryption unit configured to convert the data to be embedded into binary and form an encrypted information matrix;

[0068] The information hiding unit is configured to generate a spectrogram from the original digital audio, and use the spectrogram as a reference to embed the encrypted information matrix into the original digital audio based on the time point to be embedded, the frequency position to be embedded, and the channel to be embedded of the original digital audio, so as to obtain a first digital audio; the first audio includes the original digital audio data and the encrypted data;

[0069] The information decryption unit is configured to receive the first digital audio at the receiving end, extract the encrypted data based on the linear code theory, and post-process the encrypted audio data to obtain the data to be embedded.

[0070] It can be understood that the secure chat module first modulates the original audio separately into WAV audio, searches for low-energy or silent segments of the audio, determines the location and range of data points based on frequency domain analysis, time domain analysis, and spectrogram, and finds the corresponding binary idle code;

[0071] Performing key-retaining binary encoding on the secret information, merging the binary encoding of the secret information with the binary idle encoding in the original audio, and decompiling the result into a wav audio containing the secret information, namely, the first audio data;

[0072] The first audio data is decoded to separate different coding information, so that the original audio coding and the secret information coding can be obtained. The secret information coding is extracted based on the linear code theory to obtain the secret information.

[0073] The present invention combines reversible audio data hiding technology with an online dating system to achieve secure hiding and accurate extraction of user information without compromising audio quality. It also supports multiple audio formats and multi-channel processing, providing a safe and reliable new solution for online dating, which not only protects user privacy but also improves the security and fun of social interaction.

[0074] Specifically, in one embodiment of the present invention, the audio signal analysis specifically includes:

[0075] Perform timestamp analysis on the original audio signal, select feature words for timestamp marking according to different security requirements, and obtain keyword timestamps;

[0076] Perform multi-channel analysis on the original audio signal, and determine the embeddable channels according to the total number of channels in the original digital audio and the amount of data that can be hidden;

[0077] Performing time domain analysis on the original audio data, calculating the energy of the original audio data according to the energy adjustment factor, and determining the energy area to be embedded;

[0078] Perform frequency domain analysis on the original audio data and determine the frequency position to be embedded according to the information embedding parameters.

[0079] In the embodiments of the present invention, customized requirements are imposed on audio carriers in different scenarios, and characteristic vocabulary is marked according to different security requirements. For daily chats, in Chinese audio, words such as "de", "ni", and "wo" are used as marks. The appearance of these words is not easily noticed, and at the same time, sufficient marking density can be ensured. For the fast alarm encryption scenarios (119, 120, 110), the timestamp marking vocabulary selects words that may appear in emergency situations, such as "save", "fire", "danger", etc. These words do not often appear under normal circumstances, but their use in emergency situations will not arouse suspicion. For background real-time supervision, words that may appear in the supervision scenario are selected, such as "supervise", "manage", "audit", etc. These words do not often appear in normal conversations, but their use in the supervision environment conforms to the context.

[0080] The present invention adopts a timestamp marking method based on scenario customization, selects corresponding marking vocabulary for different scenarios (such as daily chats, alarm encryption, background supervision), making the information hiding more natural and less noticeable, while ensuring sufficient marking density, and improving the concealment and practicality of information hiding.

[0081] In an embodiment of the present invention, when using timestamps to hide information in an audio file, the accuracy of the timestamp is not less than 50 milliseconds.

[0082] In an embodiment of the present invention, after performing timestamp marking analysis on the original audio signal, information is hidden using timestamps, which specifically includes:

[0083] Select a suitable audio file as the carrier, such as a file in WAV format;

[0084] Ensure that the accuracy of the timestamp is not less than 50 milliseconds to facilitate the hiding and extraction of the data to be embedded;

[0085] Select suitable keywords for marking according to different scenarios. For example, common words such as "de", "ni", and "wo" are used in daily chats; words such as "save", "fire", "danger", etc. are used in emergency situations; words such as "supervise", "manage", "audit", etc. are used in the supervision scenario;

[0086] Perform time chunking on the original audio data and perform audio information hiding. For example, if the timestamps are T1 and T2, the word "save" can be inserted between T1 and T2. By adjusting the interval of the timestamps, ensure that the appearance of each keyword does not arouse attention, while maintaining sufficient marking density. Facilitate parsing the audio file, reading the timestamps and identifying the keywords when extracting the data to be embedded. Restore the hidden information according to the order and position of the keywords.

[0087] Furthermore, the time domain analysis specifically includes:

[0088] Calculate the average energy of the audio signal in the original digital audio. The calculation formula is:

[0089]

[0090] Among them, E avg represents the average energy of the audio signal, N represents the total number of samples in the audio signal, n represents the sample index, x[n] represents the audio signal, and N is a non-zero natural number;

[0091] The audio signal of the original data audio is divided into multiple time windows of length t, and the energy of each time window is calculated. The calculation formula is:

[0092]

[0093] Among them, E i represents the energy of the i-th time window, i represents the time window index, and t represents the length of the time window; i is the time window index, counting from zero, so the first sample of the i-th window is at index (i-1)t, and because the length of the window is t, the index of the last sample is (i-1)t+t-1=it-1;

[0094] Identify the current scene, set energy adjustment factors according to different scenes, and calculate energy thresholds based on the energy adjustment factors;

[0095] The calculating of the energy threshold based on the energy adjustment factor specifically includes:

[0096] θ chat =0.3×E avg ×(1-β);

[0097] θ alarm =0.3×E avg ×(1+γ);

[0098] θ monitor =0.3×E avg ×(1+ω×S);

[0099] Among them, θ chat represents the energy threshold in the chat scenario, θ alarm represents the energy threshold in the alarm scenario, θ monitor Indicates the energy threshold of the automatically built scene, E avg represents the average energy of the audio signal, β represents the adjustment factor, γ represents the enhancement factor, ω represents the real-time adjustment factor, and S represents the monitoring sensitivity level;

[0100] According to the energy and energy threshold of each time window, the position and range of the low-energy segment and the position and range of the silent segment are determined, and the low-energy segment is recorded as the time point to be embedded.

[0101] The sensitivity of chat is set through the background of the online dating system. When discussing daily chat, the sensitivity is 0.3, and the sensitivity of military and political chat is set according to actual usage. γ is set according to the expected attack type and interference level. For example, if the attacker attempts to destroy the hidden information by increasing the background noise, the value of γ can be increased to ensure that the information can still be extracted under higher noise levels.

[0102] ω can be set according to actual environmental monitoring data. For example, if abnormal sounds or known sensitive words are detected, the online dating system can increase the value of ω to improve the detection probability of these events.

[0103] S can be set based on a variety of factors, such as the type of sound, volume, frequency content, or known threat level. If the ambient sound contains known sensitive words or abnormal patterns, the value of S can be increased accordingly.

[0104] It can be understood that in the low energy segment of the audio signal, the silent segment is an ideal location to embed data because the embedded data is less likely to be perceived. The threshold θ is set to 1% of the average energy. For each time window, if its energy E is less than or equal to 0.01E avg , then the time period can be considered as a silent period, and the formula is as follows:

[0105]

[0106] Among them, S i Represents the energy segment of the original digital audio. When S i =1, indicating that the energy segment corresponding to the current time window is a silent segment. i =0, indicating that the energy segment corresponding to the current time window is not a silent segment and is not suitable as a time point to be embedded.

[0107] Furthermore, the frequency domain analysis specifically includes:

[0108] Perform Fourier transform on the audio signal in the original digital audio to obtain the frequency domain representation of the original digital audio. The calculation formula is:

[0109]

[0110] Where X[k] represents the frequency domain representation, k represents the frequency index, and j represents the imaginary unit;

[0111] Calculate the energy of each frequency component according to the frequency domain representation to obtain the energy distribution of each frequency component; calculation formula:

[0112] E[k]=|X[k]| 2 ;

[0113] Where E[k] represents the energy of the kth frequency;

[0114] The key frequency components and non-key frequency components are determined according to the energy distribution of each frequency component, and the non-key frequency components are recorded as the frequency positions to be embedded.

[0115] Specifically, an embedding threshold θ is set, which is based on the auditory sensitivity of the human ear and the characteristics of the audio signal. If E[k] is less than θ, the frequency component is considered to be non-critical and suitable for embedding data. If there is no non-critical frequency in the audio, data can be embedded in some critical frequency components.

[0116] In one embodiment of the present invention, by using a phase embedding method, by changing the phase embedding data of X[k], the calculation formula is:

[0117]

[0118] in, represents a phase change according to the data to be embedded;

[0119] The modified frequency domain representation X′[k] is inversely Fourier transformed to obtain the time domain signal x′[n] after embedding the data. The calculation formula is:

[0120]

[0121] It can be understood that the formula for the optimal embedding strategy is:

[0122] In one embodiment of the present invention, the multi-channel analysis specifically includes:

[0123] Analyze the frequency characteristics and volume characteristics of each channel in the original digital audio;

[0124] The total number of channels and the amount of data that can be hidden in the original digital audio are calculated, and the embeddable channels are determined based on the frequency characteristics and volume characteristics.

[0125] Among them, the total number of channels = the number of tracks * the number of channels, and the number of bits of all hidden data = the number of information points * the number of channels * the number of tracks.

[0126] The present invention utilizes multi-channel audio processing technology to expand the information hiding capacity in the manner of total number of channels = number of tracks × number of channels, and combines frequency domain analysis and energy region selection to achieve greater capacity information hiding while ensuring audio quality and reliability of information extraction.

[0127] Let's take a specific example:

[0128] The sample is taken once at the 4th second of every minute. A music file of one and a half hours is a 4-track dual-channel file. The amount of data that can be hidden using the dual-channel audio file information hiding method is 30×4×2=240 bits.

[0129] like Figure 3 As shown, in one embodiment of the present invention, the steps of the information hiding unit are as follows:

[0130] The original digital audio is used to generate a spectrogram, and the energy value of each time-frequency unit in the spectrogram is calculated according to the noise threshold; the calculation formula is:

[0131] Energy(t,f)=Σ Δt Σ Δf |X(t,f)| 2

[0132] Wherein, Energy(t, f) represents the energy value of the time-frequency unit (t, f), t represents the time of the original digital audio, f represents the frequency of the original digital audio, Δt represents the size of the time window, Δf represents the size of the frequency window, X(t, f) represents the energy value of the spectrogram at time t and frequency f, and (t, f) represents the time-frequency unit at time t and frequency f;

[0133] Detect the intensity of environmental noise, dynamically adjust the information embedding parameters of different scenes according to the noise level, and calculate the embedding energy thresholds of different scenes based on the information embedding parameters of different scenes;

[0134] Specifically, the embedding energy thresholds of different scenes are calculated based on the information embedding parameters of different scenes, and the calculation formula is:

[0135]

[0136] Among them, α chat represents the information embedding parameter in the chat scenario, α alarm represents the information embedding parameter in the emergency alarm scenario, α monitor represents the information embedding parameter in the supervision scenario, k represents a constant, L represents the chat sensitivity level, m represents a constant, U represents the urgency level, n represents a constant, C represents the privacy protection level, and S represents the monitoring sensitivity level;

[0137] Determine the energy area to be embedded based on the energy value of each time-frequency unit and the embedding energy threshold of different scenes;

[0138] The encrypted information matrix is ​​secretly scrambled to generate a carrier matrix, and the carrier matrix is ​​information encoded to obtain an encoding matrix;

[0139] Based on the time point to be embedded, the frequency position to be embedded, the embeddable channel and the energy area to be embedded, a reversible data hiding algorithm is used to embed the coding matrix into the energy area to be embedded, and audio format packaging is performed to obtain the first digital audio.

[0140] The present invention dynamically adjusts the energy threshold and adaptively adjusts the parameters (0.2-0.7) according to different environmental noise intensities (quiet environment, general environment, high-noise environment), thereby ensuring the effectiveness and robustness of information hiding in various complex environments and improving the environmental adaptability of the system.

[0141] In the embodiment of the present invention, the dynamic information embedding parameter α is set according to different scenarios. i ,By analyzing the characteristics of audio signals, the energy threshold T of different scenes is dynamically embedded to adapt to different types of audio content and environments, thereby improving the robustness and concealment of steganography. The calculation formula is:

[0142] T=α i ×AverageEnergy;

[0143] Among them, T represents the embedding energy threshold of different scenes, α i represents the information embedding parameter under scenario i, i represents different scenarios, and AverageEnergy represents a ratio of the average energy of the spectrogram;

[0144] In the case of high ambient noise, background noise (such as traffic noise, human voice, wind noise, etc.) and other interfering sounds, if the energy of the information to be embedded is set too low, the information may be completely lost in the subsequent signal processing (such as compression or noise reduction). According to the intensity of the detected ambient noise, the information embedding parameter α is adaptively adjusted. i .

[0145] In a preferred embodiment of the present invention, generally, α i Take 0.5; in a quiet environment, α i Appropriately reduce it to 0.2.

[0146] In the embodiment of the present invention, the information embedding parameter α i The maximum value of is 0.7.

[0147] In one embodiment of the present invention, the steps of the information encryption unit are as follows:

[0148] Convert the data to be embedded into ASCII code, and convert the ASCII code into 8-bit binary code to obtain binary data of the data to be embedded;

[0149] Setting an initial matrix, serializing the binary data in sequence, and filling the serially serialized binary data into the initial matrix to obtain an encrypted information matrix;

[0150] or,

[0151] Extract the RGB value of each pixel of the data to be embedded, and convert the RGB value of each pixel into 8-bit binary to obtain binary data of the data to be embedded;

[0152] An initial matrix is ​​set, the binary data is concatenated in sequence, and the binary data is filled into the initial matrix to obtain an encrypted information matrix.

[0153] Specifically, a specific embodiment is used to illustrate:

[0154] If the secret data is already in binary form, use the text "miyou" as an example to show how to convert the secret data into binary form and organize it into a matrix for embedding into other media. Specifically:

[0155] Convert the text "miyou" to ASCII code: 'm'->109, 'i'->105, 'y'->121, 'o'->111, 'u'->117;

[0156] Convert each ASCII code to 8-bit binary form:

[0157] 109->01101101, 105->01101001, 121->01111001, 111->01101111, 117->01110101;

[0158] Concatenate binary data in sequence and organize them into a matrix form:

[0159] Binary_Text=01101101 01101001 01111001 01101111 01110101

[0160] Choose a 3x5 matrix Q to organize these data (because 3x5=15, and we have 5 characters, each character is 8 bits, a total of 40 bits, which can be divided by 3x5), where the matrix Q is an M×N matrix, M×N≥L, M represents the number of rows of the matrix Q, N represents the number of columns of the matrix Q, and L represents the length of the binary data, that is:

[0161] Q=[01101101 01101001 01111001 01101111 0111010];

[0162] Divide the binary data S into M×N data blocks, each of which contains b bits (where b is the number of data bits in each row or column). If L cannot be divided by M×N, fill S with padding bits (such as 0) to make its length M×N. Organize the divided data blocks into a matrix Q in rows or columns. The formula is:

[0163] Q ij =S index(i，j) ;

[0164] Among them, Q ij represents the element in the i-th row and j-th column of the matrix Q. index(i, j) represents the index position calculated for row i and column j. i represents the number of rows in the matrix Q, and j represents the number of columns in the matrix Q. S index(i，j) Represents the corresponding in the binary data S.

[0165] If the secret data is not binary data but an image, then convert each color value of the RGB image into 8-bit binary form, where each pixel P(a, b) consists of three color values ​​(R, G, B):

[0166]

[0167] Wherein, P(a, b) represents the pixel point at the ath row and bth column of the image, a represents the row coordinate of the pixel in the image, a represents the column coordinate of the pixel in the image, ConvertToB inary(R) represents converting the decimal value (0-255) of the red component R into 8-bit binary, ConvertToB inary(G) represents converting the decimal value (0-255) of the green component G into 8-bit binary, ConvertToB inary(B) represents converting the decimal value (0-255) of the blue component B into 8-bit binary, R represents the red component value of the image, G represents the green component value of the image, and B represents the blue component value of the image. Represents the exclusive-or operation.

[0168] A specific embodiment is used to illustrate:

[0169] A 3x3 RGB image, the RGB value of each pixel is as follows:

[0170]

[0171] Convert each color value to 8-bit binary form and concatenate them row by row:

[0172]

[0173] Among them, Binary Image represents the converted binary sequence,

[0174] The encoded binary data is organized into a matrix form for easy embedding.

[0175] In one embodiment of the present invention, the binary form of the secret data is T, and the length is W bits. We organize it into a matrix P of R×Y, where R×Y≥W; R represents the number of rows of the matrix P, Y represents the number of columns of the matrix P, and W represents the length of the binary data T;

[0176] Divide the binary data T into R×Y data blocks, each containing τ bits (where τ is the number of data bits per row or column). If W is not divisible by R×Y, fill T with padding bits (such as 0) to make its length R×Y. Organize the divided data blocks into a matrix P in rows or columns, and the calculation formula is as follows:

[0177]

[0178] Among them, P i′j′ represents the element in the i'th row and j'th column of the matrix P. If P i′j′ =T index(i′，j′) , indicating that the element in the i'th row and j'th column of the matrix P is the element in the i'th row and j'th column of the binary data T. If P i′j′ =0, it means that the element in the i'th row and j'th column in the matrix P exceeds the length of the binary data T and is filled with padding bits 0.

[0179] If R = 4, Y = 5, and W = 17, then the matrix P can be organized as follows:

[0180]

[0181] Wherein, T1 represents the first bit of binary data T, T2 represents the second bit of binary data T, T3 represents the third bit of binary data T, T4 represents the fourth bit of binary data T, T5 represents the fifth bit of binary data T, T6 represents the sixth bit of binary data T, T7 represents the seventh bit of binary data T, T8 represents the eighth bit of binary data T, T9 represents the ninth bit of binary data T, T 10 Represents the tenth bit of binary data T, T 11 Represents the eleventh bit of binary data T, T 12 Represents the twelfth bit of binary data T, T 13 Indicates the thirteenth digit of binary data T, T 14 Indicates the fourteenth bit of binary data T, T 15 Represents the fifteenth bit of binary data T, T 16 Represents the sixteenth bit of binary data T, T 17 represents the seventeenth bit of binary data T;

[0182] By using the matrix drawing method, the audio binary data is organized into a matrix form suitable for embedding.

[0183] As can be understood by those skilled in the art, data embedding and data extraction are inverse processes of encrypting and decrypting information, and therefore the principles and techniques are the same.

[0184] In one embodiment of the present invention, the steps of the information decryption unit are as follows:

[0185] A first digital audio is received, a plurality of same-type matrices are selected as transmission carriers of data to be embedded, a binary sequence of the data to be embedded is received, and secret information is recovered based on linear code theory.

[0186] It can be understood that after the digital audio is digitized, sampled, quantized and encoded, a set of binary data will be generated, which is the secret information that needs to be hidden.

[0187] A specific embodiment is used to illustrate:

[0188] Divide the long string of binary information into equal length and smaller sub-information segments, and then cut the sub-information segments into equal parts to generate a matrix V with equal determinant by row. n*n , taking "miyou" as an example,

[0189]

[0190] Scramble the matrix to improve its encryption security. Introduce a quasi-affine transformation on the integer domain and multiply each element of the matrix V by e. Since Q is a binary matrix, this operation is actually a bitwise multiplication, but here we treat it as a modulo 256 multiplication to handle possible overflow. The parameters e and f are selected here to ensure the security and randomness of the transformation. The parameters e and f should meet the following conditions:

[0191] e should be an integer that is relatively prime to the domain size of the transformation (usually 256, because a byte has 256 possible values). This ensures that the multiplication operation is reversible modulo 256. f should be a random integer. The role of f is to add an offset to each element to increase the complexity and security of the transformation. The formula for the quasi-affine transformation is as follows:

[0192] V′=(V·e+f)mod256

[0193] Wherein, V′ represents a quasi-affine transformation matrix, V represents a matrix with equal determinants, e represents a coprime parameter, f represents an offset parameter, · represents a bitwise multiplication, and mod256 represents modulo 256;

[0194] After the quasi-affine transformation, the matrix V is realized by integer lifting transformation, and e and f are rounded to integers to complete the bijection V on the integer domain. n*n →V′ n*n , the calculation formula is as follows:

[0195]

[0196] in, represents the result matrix after bijective transformation, x′ represents the x-coordinate value of the result matrix, and y′ represents the y-coordinate value of the result matrix. represents the quasi-affine transformation matrix, u represents the element of the first row and first column of the quasi-affine transformation matrix, v represents the element of the first row and second column of the quasi-affine transformation matrix, c represents the element of the second row and first column of the quasi-affine transformation matrix, and d represents the element of the second row and second column of the quasi-affine transformation matrix. represents the original matrix V, x represents the x-coordinate value of the original matrix V, y represents the y-coordinate value of the original matrix V, represents the translation vector;

[0197] If e=3, f=5, then the first element of V (0*3+5) mod 256=5;

[0198] The second element of V (1*3+5) mod 256 = 8, and so on:

[0199]

[0200] Generate n N×M matrices Ki, which will be used as carriers of information to be transmitted. According to linear coding theory, a generator matrix G of N×M size is generated, where G h，g ∈Z, where h represents the number of rows of matrix G and g represents the number of columns of matrix G. For each matrix V′, the operation is performed to obtain the encoded matrix G i :

[0201] G i =V′*Gmod2

[0202] Each row of V′ can be regarded as a linearly coded codeword. The results are XORed to obtain the coded matrix G i .

[0203] Generate n matrices Ci on Z2 of the same size as Ki. And define the sequence: S1, S2, ..., Sm, C1, C2, ..., Cη, D0, D1, ..., Dη + λ, where Di is generated by summing the previous n + λ matrices and taking the modulo 2:

[0204] D λ+η =(D0+D1+…+D η+λ-1 )mod2

[0205] Among them, D λ+η represents the last D value in the sequence, D0 represents the first D value in the sequence, D1 represents the second D value in the sequence, and D η+λ-1 Indicates the second to last D value in the sequence;

[0206] Di is the cumulative sum of all previous D values, S values, and values, which is equivalent to adding all matrices bit by bit. If the result is an odd number, Di = 1; if it is an even number, Di = 0;

[0207] Randomly select a number r∈Z and a two-variable one-way function f(r,β). For each Ki, we calculate its same-sized matrix Ki'=f(r,β)mod 2 and calculate Ti=Di-Ki'mod2. Finally, the generated matrix G, random number r, one-way function f(r,β), and sequence T1,T2,…,Tη,Dη+1,Dη+2,…,Dη+λ are output;

[0208] After receiving ω carriers Ki (i = 1, 2, 3 ... η), the receiver calculates Ki' (i = 1, 2, 3 ... η), Di (i = 1, 2, 3 ... η) according to the random number r and f (r, k), Ti (i = 1, 2, 3 ... η), Di (i = η + 1, η + 2 ... η + λ). Combined with the coefficient matrix, we can now solve S1, ..., Sλ, C1, ..., Cη. Thus, the encrypted message "miyou" is obtained.

[0209]

[0210] Among them, Q represents the encryption matrix, a 11 Represents the element in the first row and first column of the encryption matrix, a 21 Represents the element in the second row and first column of the encryption matrix, a (η+λ)1 represents the element in the first column of the η+λth row of the encryption matrix, a 12 Represents the element in the first row and second column of the encryption matrix, a 22 Represents the element in the second row and second column of the encryption matrix, a (η+λ)2 represents the element in the η+λth row and second column of the encryption matrix, a 1(η+λ) represents the element in the first row and the n+λth column of the encryption matrix, a 2(η+λ) represents the η+λth element in the second row of the encryption matrix, a (η+λ)(η+λ) represents the element in the n+λth row and the n+λth column of the encryption matrix;

[0211] In the embodiment of the present application, the python code is reasonably merged and decompiled into a wav audio with secret information. Decoding separates different coded information, obtains secret information and retains binary encoding. The audio signal with embedded data is encapsulated into an appropriate file format for easy transmission. That is, the technology is encapsulated and implemented in code, leaving only the interface part. The user runs the encode file code, selects the secret file, selects the audio carrier file for encryption, and obtains the encrypted audio file. Run the decode file code, upload the encrypted audio file, and obtain the secret information.

[0212] The decode file has two functions. The first function reads the hidden data bytes from the wav file, starting from the lowest two bits of each text. The second function extracts the hidden data from the wav file and saves it as an image file or text file. First read all the bytes of the wav, parse the wav structure, get the wav data part, and then read the hidden data size. Finally, save it as a file in the corresponding format and output the data memory size.

[0213] encode is an encrypted file, which contains three functions. The first function converts signed integers into unsigned integers. The second function hides one byte of data into the PCM data of the WAV file and loops until all are hidden. The third function calculates the space available for hiding in the WAV file by reading all bytes of the WAV, image file, and text file. If the image can be converted into four-byte binary data, the file size and data are put into the WAV file, and then the written WAV file is written into a new file.

[0214] These two files are the core files for technical implementation. Calling these two file codes in the online dating system can achieve encryption and decryption.

[0215] like Figure 4 As shown, the network friendship system of the present invention is described by a specific embodiment (the network friendship system of the present invention is applied to the "Miyou" friendship software):

[0216] Users can browse the website information to find "About Miyou" and "About Us" to learn more about "Miyou Friend";

[0217] Log in and register through "Personal User Registration and Login". If you forget your password, you can reset it.

[0218] After successfully logging in, you can manage your friends through "Personal Information and Management", such as recommending friends, adding friends, deleting friends, etc. You can also manage your friends through "Search and Match";

[0219] When chatting with friends, you can use text chat or voice chat. In voice chat, the sound information will be hidden. When the hidden sound information contains hidden distress information or sensitive words are detected, the alarm will automatically send a distress signal or give a warning or prompt;

[0220] After the user logs in to "Miyou", security and privacy will be protected. When sensitive words appear during use, the app will automatically alarm and send a distress signal or give warnings and prompts.

[0221] like Figure 5 As shown, the present invention also provides an audio information hiding method, which is applied to the online dating system as described above, and includes the following steps:

[0222] S1. Perform timestamp analysis on the original audio signal, select characteristic words for timestamp marking according to different security requirements, and obtain keyword timestamps;

[0223] S2, performing multi-channel analysis on the original audio signal, and determining the embeddable channels according to the total number of channels in the original digital audio and the amount of data that can be hidden;

[0224] S3, performing time domain analysis on the original audio data, calculating the energy of the original audio data according to the energy adjustment factor, and determining the energy area to be embedded;

[0225] S4, performing frequency domain analysis on the original audio data, and determining the frequency position to be embedded according to the information embedding parameter;

[0226] S5, converting the data to be embedded into binary and forming an encrypted information matrix;

[0227] S6. Based on the keyword timestamp, the embeddable channel, the energy area to be embedded, and the frequency position to be embedded, a reversible data hiding algorithm is used to add the encrypted information matrix to the energy area to be embedded to obtain the first audio data.

[0228] like Figure 6 As shown, specifically, at the sending end, the secret data H is first embedded into the original carrier audio I using the data embedding method, and then the audio W embedded with the secret data is sent to the receiver. At the receiving end, the embedded secret data H is extracted using the data extraction and audio recovery methods, and the original carrier audio I is restored.

[0229] The network friendship device provided by the present invention is described below. The network friendship device described below and the audio information hiding method described above can be referred to each other.

[0230] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 7As shown, the electronic device may include: a processor (processor) 710, a communication interface (Communications Interface) 720, a memory (memory) 730 and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other through the communication bus 740. The processor 710 can call the logic instructions in the memory 730 to execute the audio information hiding method, which includes: performing timestamp marking analysis on the original audio signal, selecting feature words for timestamp marking according to different security requirements, and obtaining a keyword timestamp; performing multi-channel analysis on the original audio signal, and determining the embeddable channels according to the total number of channels in the original digital audio and the amount of data that can be hidden; performing time domain analysis on the original audio data, calculating the energy of the original audio data according to the energy adjustment factor, and determining the energy area to be embedded; performing frequency domain analysis on the original audio data, and determining the frequency position to be embedded according to the information embedding parameters; converting the data to be embedded into binary and forming an encrypted information matrix; based on the keyword timestamp, the embeddable channels, the energy area to be embedded, and the frequency position to be embedded, using a reversible data hiding algorithm to add the encrypted information matrix to the energy area to be embedded to obtain the first audio data.

[0231] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0232] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the audio information hiding method provided by the above-mentioned methods, which includes: performing timestamp marking analysis on the original audio signal, selecting feature words for timestamp marking according to different security requirements, and obtaining a keyword timestamp; performing multi-channel analysis on the original audio signal, and determining the embeddable channel according to the total number of channels in the original digital audio and the amount of hidden data; performing time domain analysis on the original audio data, calculating the energy of the original audio data according to the energy adjustment factor, and determining the energy area to be embedded; performing frequency domain analysis on the original audio data, and determining the frequency position to be embedded according to the information embedding parameters; converting the data to be embedded into binary and forming an encrypted information matrix; based on the keyword timestamp, the embeddable channel, the energy area to be embedded, and the frequency position to be embedded, using a reversible data hiding algorithm to add the encrypted information matrix to the energy area to be embedded to obtain the first audio data.

[0233] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the audio information hiding method provided by the above-mentioned methods, the method comprising: performing timestamp marking analysis on the original audio signal, selecting feature words for timestamp marking according to different security requirements, and obtaining a keyword timestamp; performing multi-channel analysis on the original audio signal, and determining the embeddable channels according to the total number of channels in the original digital audio and the amount of data that can be hidden; performing time domain analysis on the original audio data, calculating the energy of the original audio data according to the energy adjustment factor, and determining the energy area to be embedded; performing frequency domain analysis on the original audio data, and determining the frequency position to be embedded according to the information embedding parameters; converting the data to be embedded into binary and forming an encrypted information matrix; based on the keyword timestamp, the embeddable channels, the energy area to be embedded, and the frequency position to be embedded, using a reversible data hiding algorithm to add the encrypted information matrix to the energy area to be embedded, and obtaining the first audio data.

[0234] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0235] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0236] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A network friendship system based on audio information hiding, characterized in that: include: A secure chat module configured to encrypt the data to be embedded into the original digital audio using a reversible data hiding algorithm to form first audio data, and use the first audio data for communication; A hidden help-seeking module is configured to set hidden alarm information in the data to be embedded, and the online dating system monitors the embedded information in real time and issues an alarm after detecting the hidden alarm information; The sensitive word detection and automatic alarm module is configured so that the online dating system monitors the data to be embedded in real time and automatically triggers the alarm mechanism after detecting that the data to be embedded contains sensitive words; A multi-channel audio processing module is configured to provide additional information hiding capacity for an online dating system.

2. According to claim 1, a network friendship system based on audio information hiding is characterized in that: The secure chat module specifically includes: An audio signal processing unit configured to perform audio signal analysis on the original digital audio to determine a time point to be embedded, a frequency position to be embedded, and a channel to be embedded in the original digital audio; wherein the audio signal analysis includes timestamp analysis, multi-channel analysis, time domain analysis, and frequency domain analysis; An information encryption unit configured to convert the data to be embedded into binary and form an encrypted information matrix; The information hiding unit is configured to generate a spectrogram from the original digital audio, and use the spectrogram as a reference to embed the encrypted information matrix into the original digital audio based on the time point to be embedded, the frequency position to be embedded, and the channel to be embedded of the original digital audio, so as to obtain a first digital audio; the first audio includes the original digital audio data and the encrypted data; The information decryption unit is configured to receive the first digital audio at the receiving end, extract the encrypted data based on the linear code theory, and post-process the encrypted audio data to obtain the data to be embedded.

3. According to claim 2, a network friendship system based on audio information hiding is characterized in that: The steps of the information hiding unit are as follows: Generate a spectrogram from the original digital audio, and calculate the energy value of each time-frequency unit in the spectrogram according to a noise threshold; Detect the intensity of environmental noise, dynamically adjust the information embedding parameters of different scenes according to the noise level, and calculate the embedding energy thresholds of different scenes based on the information embedding parameters of different scenes; Determine the energy area to be embedded based on the energy value of each time-frequency unit and the embedding energy threshold of different scenes; The encrypted information matrix is ​​secretly scrambled to generate a carrier matrix, and the carrier matrix is ​​information encoded to obtain an encoding matrix; Based on the time point to be embedded, the frequency position to be embedded, the embeddable channel and the energy area to be embedded, a reversible data hiding algorithm is used to embed the coding matrix into the energy area to be embedded, and audio format packaging is performed to obtain the first digital audio.

4. The online friendship system based on audio information hiding according to claim 3 is characterized in that: The information embedding parameters of different scenes are used to calculate the embedding energy thresholds of different scenes, and the calculation formula is: Among them, α chay represents the information embedding parameter in the chat scenario, α alarm represents the information embedding parameter in the emergency alarm scenario, α monitor represents the information embedding parameter in the supervision scenario, k represents a constant, L represents the chat sensitivity level, m represents a constant, U represents the urgency level, n represents a constant, C represents the privacy protection level, and S represents the monitoring sensitivity level.

5. The online friendship system based on audio information hiding according to claim 2 is characterized in that: The time domain analysis specifically includes: Calculate the average energy of the audio signal in the original digital audio; Divide the audio signal of the original data audio into multiple time windows of length t, and calculate the energy of each time window; Identify the current scene, set energy adjustment factors according to different scenes, and calculate energy thresholds based on the energy adjustment factors; According to the energy and energy threshold of each time window, the position and range of the low-energy segment and the position and range of the silent segment are determined, and the low-energy segment is recorded as the time point to be embedded.

6. The online friendship system based on audio information hiding according to claim 5 is characterized in that: The calculating of the energy threshold based on the energy adjustment factor specifically includes: i chat =0.3×E avg ×(1-β); i alarm =0.3×E avg ×(1+γ); i monitor =0.3×E avg ×(1+ω×S); Among them, θ chat represents the energy threshold in the chat scenario, θ alarm represents the energy threshold in the alarm scenario, θ monitor Indicates the energy threshold of the automatically built scene, E avg represents the average energy of the audio signal, β represents the adjustment factor, γ represents the enhancement factor, ω represents the real-time adjustment factor, and S represents the monitoring sensitivity level.

7. The online friendship system based on audio information hiding according to claim 2 is characterized in that: The steps of the information encryption unit are as follows: Convert the data to be embedded into ASCII code, and convert the ASCII code into 8-bit binary code to obtain binary data of the data to be embedded; Setting an initial matrix, serializing the binary data in sequence, and filling the serially serialized binary data into the initial matrix to obtain an encrypted information matrix; or, Extract the RGB value of each pixel of the data to be embedded, and convert the RGB value of each pixel into 8-bit binary to obtain binary data of the data to be embedded; An initial matrix is ​​set, the binary data is concatenated in sequence, and the binary data is filled into the initial matrix to obtain an encrypted information matrix.

8. An audio information hiding method, characterized in that: The method is applied to the online friendship system as claimed in any one of claims 1 to 7, comprising the following steps: S1. Perform timestamp analysis on the original audio signal, select characteristic words for timestamp marking according to different security requirements, and obtain keyword timestamps; S2, performing multi-channel analysis on the original audio signal, and determining the embeddable channels according to the total number of channels in the original digital audio and the amount of data that can be hidden; S3, performing time domain analysis on the original audio data, calculating the energy of the original audio data according to the energy adjustment factor, and determining the energy area to be embedded; S4, performing frequency domain analysis on the original audio data, and determining the frequency position to be embedded according to the information embedding parameter; S5, converting the data to be embedded into binary and forming an encrypted information matrix; S6. Based on the keyword timestamp, the embeddable channel, the energy area to be embedded, and the frequency position to be embedded, a reversible data hiding algorithm is used to add the encrypted information matrix to the energy area to be embedded to obtain the first audio data.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the audio information hiding method as claimed in claim 8 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the audio information hiding method as claimed in claim 8 is implemented.

Images