Audio identification and generation method, device, equipment, medium and product embedded with digital watermark

Abstract

The present application discloses a method, apparatus, device, medium, and product for audio identification and generation embedded with a digital watermark, relating to the field of audio processing technology. The method comprises performing frame processing on an acquired audio signal to be identified to obtain multiple audio frames of fixed length; performing a discrete Fourier transform on each audio frame, and determining the maximum amplitude frequency in each audio frame after the discrete Fourier transform as a first embedded reference frequency; determining a frequency having complementary characteristics to the first embedded reference frequency as a second embedded reference frequency; performing phase continuity processing on each audio frame, and performing phase compensation on each audio frame based on the phase difference between the first embedded reference frequency and the second embedded reference frequency after the phase continuity processing to obtain an original phase audio signal; and extracting watermark information from the original phase audio signal and splicing it to obtain a binary value watermark sequence. The present application can effectively improve the success rate of watermark information extraction.

Description

Technical Field

[0001] The present application relates to the field of audio processing technology, and in particular to a method, device, equipment, medium and product for audio identification and generation embedded with a digital watermark. Background Art

[0002] With the continuous iteration of speech models, the speech synthesized by audio synthesis technology is becoming closer and closer to real audio, that is, natural speech, in terms of naturalness, realism, and diversity, making it easier to tamper with the recorded content.

[0003] Currently, watermarks are added to prevent audio recordings from being tampered with. However, recordings often undergo multiple processing and transformations, such as transcoding, compression, changes in playback equipment, and environmental noise. These operations can cause the watermark information to be lost or destroyed, making it impossible to extract valid proof information during subsequent recording authentication. Therefore, there is an urgent need to provide a robust and easy-to-identify audio watermark generation and authentication method. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the related art, the purpose of this application is to provide an audio identification and generation method, device, equipment, medium and product embedded with a digital watermark, which can improve the robustness of the digital watermark in the audio signal and facilitate identification.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides an audio authentication method for embedding a digital watermark, the method comprising: performing frame processing on the acquired audio signal to be authenticated to obtain multiple audio frames of fixed length; the audio signal to be authenticated is a recording signal after embedding watermark information and bandpass filter processing, and the maximum amplitude frequency of the recording signal and a frequency with complementary characteristics to the maximum amplitude frequency are used to embed watermark information; performing discrete Fourier transform on each of the audio frames, and determining the maximum amplitude frequency in each of the audio frames after discrete Fourier transform as a first embedding reference frequency; determining a frequency with complementary characteristics to the first embedding reference frequency as a second embedding reference frequency; performing phase continuity processing on each of the audio frames, and performing phase compensation on each of the audio frames based on the phase difference between the first embedding reference frequency and the second embedding reference frequency after the phase continuity processing to obtain the original phase audio signal before bandpass filter processing; extracting the watermark information of the original phase audio signal and splicing it to obtain a binary value watermark sequence.

[0007] Optionally, the frame processing of the acquired audio signal to be identified to obtain multiple audio frames of fixed length includes: determining the total length of the audio signal to be identified; dividing the audio signal to be identified based on a preset length and a frame overlap rate to obtain multiple audio frames to be processed; performing windowing processing on each of the audio frames to be processed, and storing the timestamp of each audio frame to be processed after the windowing processing to obtain multiple audio frames of fixed length.

[0008] Optionally, performing a discrete Fourier transform on each of the audio frames and determining the maximum amplitude frequency in each of the audio frames after the discrete Fourier transform as the first embedded reference frequency includes: converting each of the audio frames from time domain audio to frequency domain audio through a discrete Fourier transform to obtain an amplitude spectrum and a phase spectrum of each of the audio frames; analyzing the amplitude spectrum of each of the audio frames, and determining the maximum amplitude frequency in each of the audio frames as the first embedded reference frequency.

[0009] Optionally, the phase continuity processing is performed on each of the audio frames, and the phase compensation is performed on each of the audio frames based on the phase difference between the first embedded reference frequency and the second embedded reference frequency after the phase continuity processing to obtain the original phase audio signal before the bandpass filter processing, including: determining the third embedded reference frequency based on the first embedded reference frequency; judging whether the phase of each of the audio frames is continuous based on the phase difference between the first embedded reference frequency and the third embedded reference frequency; if the phase of each of the audio frames is not continuous, adjusting the phase value of each of the audio frames to obtain the audio frame with phase continuity; when the phase of each of the audio frames is continuous and the bandpass filter has a linear phase characteristic within the passband, determining the phase distortion factor of the bandpass filter based on the first embedded reference frequency and the corresponding phase spectrum, the second embedded reference frequency and the corresponding phase spectrum; compensating the phase of each audio frame based on the phase distortion factor to obtain the original phase audio signal.

[0010] Optionally, the watermark information of the original phase audio signal is extracted and spliced ​​to obtain a binary value watermark sequence, including: if the corrected phase difference between the first embedded reference frequency and the second embedded reference frequency is greater than a preset threshold, the binary value is 1; if the corrected phase difference between the first embedded reference frequency and the second embedded reference frequency is less than or equal to a preset threshold, the binary value is 0; and the binary value of each audio frame is spliced ​​in timestamp order to obtain the binary value watermark sequence.

[0011] In a second aspect, the present application provides an audio generation method embedded with a digital watermark, the method comprising: performing frame processing on the acquired recording signal to be processed to obtain multiple fixed-length frames; performing discrete Fourier transform on each of the fixed-length frames, and determining the maximum amplitude frequency in each of the fixed-length frames after discrete Fourier transform as a first reference frequency; determining a frequency having complementary characteristics to the first reference frequency as a second reference frequency; embedding watermark information in the phases of the first reference frequency and the second reference frequency to generate fixed-length frames containing watermarks; performing inverse transform and time domain processing on each of the fixed-length frames containing watermarks, and splicing the fixed-length frames containing watermarks after inverse transform and time domain processing to obtain an audio signal containing a binary value watermark sequence.

[0012] In a third aspect, the present application provides an audio authentication device embedded with a digital watermark, the device comprising:

[0013] an audio framing module configured to perform framing processing on the acquired audio signal to be authenticated, thereby obtaining a plurality of audio frames of fixed length; the audio signal to be authenticated is a recording signal that has been processed by embedding watermark information and bandpass filtering, wherein the maximum amplitude frequency of the recording signal and a frequency having a complementary characteristic to the maximum amplitude frequency are used to embed the watermark information;

[0014] a transformation determination module, configured to perform a discrete Fourier transform on each of the audio frames, and determine a maximum amplitude frequency in each of the audio frames after the discrete Fourier transform as a first embedding reference frequency;

[0015] a frequency determination module, configured to determine a frequency having a complementary characteristic to the first embedded reference frequency as a second embedded reference frequency;

[0016] a phase compensation module, configured to perform phase continuity processing on each of the audio frames, and perform phase compensation on each of the audio frames based on a phase difference between the first embedded reference frequency and the second embedded reference frequency after the phase continuity processing, to obtain an original phase audio signal before bandpass filter processing;

[0017] The extraction and splicing module is used to extract the watermark information of the original phase audio signal and splice it to obtain a binary value watermark sequence.

[0018] In a fourth aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the audio authentication method embedded with a digital watermark and the steps of the audio generation method embedded with a digital watermark as described above.

[0019] In a fifth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the audio authentication method with embedded digital watermark or the audio generation method with embedded digital watermark as described above.

[0020] In a sixth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned methods for audio authentication with embedded digital watermarks or the steps of the method for audio generation with embedded digital watermarks.

[0021] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0022] The present application provides an audio identification and generation method, apparatus, device, medium and product for embedding digital watermarks. By performing frame processing on the acquired audio signal to be identified, a plurality of audio frames of fixed length are obtained, which can realize independent processing of the plurality of audio frames, help reduce the complexity of data processing, and is suitable for batch verification of large-scale audio signals, and makes the subsequent extraction of watermark information more robust; by performing discrete Fourier transform on each audio frame, and determining the maximum amplitude frequency in each audio frame as the first embedded reference frequency; by determining the frequency with complementary characteristics to the first embedded reference frequency as the second embedded reference frequency; by performing phase continuity processing on each audio frame, and performing phase compensation on each audio frame based on the phase difference between the first embedded reference frequency and the second embedded reference frequency after the phase continuity processing, the original phase audio signal before bandpass filter processing is obtained; by extracting the watermark information of the original phase audio signal and splicing it, a binary value watermark sequence is obtained; on the one hand, due to the audio signal to be identified, the phase difference between the first embedded reference frequency and the second embedded reference frequency is obtained. The signal is a recording signal after embedding watermark information and bandpass filter processing, and the maximum amplitude frequency of the recording signal and the frequency with complementary characteristics to the maximum amplitude frequency are used to embed the watermark information, which can not only reduce the probability of the watermark information being discovered by the human ear and increase the concealment of the watermark information, so that the watermark information will not be easily discovered and tampered with; but also improve the robustness of the watermark information against bandpass filter attacks and maintain the compression resistance of the watermark information; therefore, when extracting the watermark information from the audio signal to be identified, it is necessary to perform a discrete Fourier transform on the audio signal to be identified to obtain a first embedded reference frequency and a second embedded reference frequency with complementary characteristics to the first embedded reference frequency, so as to facilitate the subsequent extraction of a complete binary value watermark sequence; on the other hand, by performing phase continuity processing and phase compensation on each audio frame, the original phase audio signal before bandpass filter processing can be obtained, which can effectively eliminate the phase distortion caused by the bandpass filter on each audio frame, and can effectively improve the success rate of watermark information extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 This is a diagram illustrating an application environment of an audio authentication method for embedding a digital watermark in one embodiment of the present application;

[0025] Figure 2 A flowchart of an audio authentication method for embedding a digital watermark provided in one embodiment of the present application;

[0026] Figure 3 A schematic diagram of the watermark information authentication process provided in one embodiment of the present application;

[0027] Figure 4 A schematic diagram of the watermark information extraction process provided in one embodiment of the present application;

[0028] Figure 5 A schematic diagram of phase unwrapping provided in one embodiment of the present application;

[0029] Figure 6 A schematic diagram of the watermark embedding process provided in one embodiment of the present application;

[0030] Figure 7 A schematic diagram of a process for framing watermark information embedding provided in an embodiment of the present application;

[0031] Figure 8 A schematic diagram of a weight function for a splicing process according to an embodiment of the present application;

[0032] Figure 9 A schematic diagram of the functional modules of an audio authentication device for embedding a digital watermark provided in one embodiment of the present application;

[0033] Figure 10 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0035] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0036] The audio authentication method for embedding digital watermarks provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, the terminal 102 communicates with the server 104 via a network. The data storage system can store data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. Terminal 102 may send an audio signal to be authenticated to server 104. The audio signal to be authenticated is a recording signal that has been processed by embedding watermark information and bandpass filtering. The maximum amplitude frequency of the recording signal and a frequency complementary to the maximum amplitude frequency are used to embed the watermark information. After receiving the audio signal to be authenticated, server 104 may frame the audio signal to obtain multiple audio frames of fixed length. A discrete Fourier transform may be performed on each audio frame, and the maximum amplitude frequency in each audio frame after the discrete Fourier transform may be determined as a first embedding reference frequency. A frequency complementary to the first embedding reference frequency may be determined as a second embedding reference frequency. Phase continuity processing may be performed on each audio frame, and phase compensation may be performed on each audio frame based on the phase difference between the first and second embedded reference frequencies after the phase continuity processing to obtain the original phase audio signal before the bandpass filtering. The watermark information of the original phase audio signals may be extracted and concatenated to obtain a binary watermark sequence. Server 104 may then provide the obtained video tag for the video to terminal 102. In addition, in some embodiments, the audio authentication method of embedding a digital watermark can also be implemented independently by the server 104 or the terminal 102. For example, the terminal 102 can directly perform digital watermark authentication on the audio signal to be authenticated, or the server 104 can obtain the audio signal to be authenticated from the data storage system and perform digital watermark authentication on the audio signal to be authenticated.

[0037] Terminal 102 may include, but is not limited to, various desktop computers, laptops, smartphones, tablet computers, IoT devices, and portable wearable devices. IoT devices may include smart speakers, smart TVs, smart air conditioners, and smart car devices. Portable wearable devices may include smart watches, smart bracelets, and head-mounted devices. Server 104 may be implemented as a standalone server or a server cluster consisting of multiple servers, or may be a cloud server.

[0038] In an exemplary embodiment, Figure 2As shown, a method for audio identification with embedded digital watermark is provided. The method is executed by a computer device, specifically a computer device such as a terminal or a server, or a terminal and a server. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate the process, including the following steps S210 to S250.

[0039] Step S210 : performing frame processing on the acquired audio signal to be identified to obtain a plurality of audio frames of fixed length.

[0040] In an exemplary embodiment, the audio signal to be identified involved in the embodiment of the present application is a recording signal after embedding watermark information and bandpass filter processing, and the maximum amplitude frequency of the recording signal and the frequency with complementary characteristics to the maximum amplitude frequency are used to embed watermark information.

[0041] It should be noted that, based on the characteristic that a sound and a sound one octave apart are perceived as similar by the human ear, in the embodiment of the present application, the maximum amplitude frequency of the recording signal and the frequency one octave apart from the maximum amplitude are used to embed watermark information.

[0042] Optionally, the above-mentioned step S210 may include: determining the total length of the audio signal to be identified; dividing the audio signal to be identified based on a preset length and a frame overlap rate to obtain a plurality of audio frames to be processed; performing windowing processing on each audio frame to be processed, and storing the timestamp of each audio frame to be processed after the windowing processing to obtain a plurality of audio frames of fixed length.

[0043] By processing the audio signal to be identified through framing, distortion can be reduced and robustness can be enhanced.

[0044] Step S220 , performing discrete Fourier transform on each audio frame, and determining the maximum amplitude frequency in each audio frame after the discrete Fourier transform as the first embedding reference frequency.

[0045] In the example embodiment, since the audio signal to be identified is a recording signal that has been processed by embedding watermark information and a band-pass filter, and the maximum amplitude frequency of the recording signal and a frequency with complementary characteristics to the maximum amplitude frequency are used to embed the watermark information, it can not only reduce the probability of the watermark information being discovered by the human ear and increase the concealment of the watermark information, so that the watermark information will not be easily discovered and tampered with; it can also improve the robustness of the watermark information against band-pass filter attacks and maintain the compression resistance of the watermark information; therefore, when extracting the watermark information from the audio signal to be identified, it is necessary to perform a discrete Fourier transform on the audio signal to be identified to obtain the maximum amplitude frequency of each audio frame.

[0046] Optionally, each audio frame is converted from time domain audio to frequency domain audio through discrete Fourier transform to obtain the amplitude spectrum and phase spectrum of each audio frame; the amplitude spectrum of each audio frame is analyzed, and the maximum amplitude frequency in each audio frame is determined as the first embedded reference frequency.

[0047] Step S230 : determining a frequency having complementary characteristics to the first embedded reference frequency as a second embedded reference frequency.

[0048] In conjunction with the above embodiment, a sound that differs by one octave from another will be perceived as similar by the human ear, and the frequency component that is one octave away can "support" the main frequency component, making the human ear's perception of the two sounds complementary. Therefore, the watermark information in the audio signal to be authenticated is embedded in two frequency phases that differ by one octave. This embedding method can ensure that the sound quality of the recorded signal is not affected, especially within the sensitive frequency range of human hearing, ensuring that the sound quality of the recorded signal is minimal. This makes it easy for the watermark information to go unnoticed and prevent malicious tampering or deletion. Correspondingly, when authenticating the watermark information, it is necessary to determine the maximum amplitude frequency in the audio signal to be authenticated and the frequency that differs by one octave from the maximum amplitude frequency.

[0049] Step S240 , performing phase continuity processing on each audio frame, and performing phase compensation on each audio frame based on the phase difference between the first embedded reference frequency and the second embedded reference frequency after the phase continuity processing, to obtain the original phase audio signal before bandpass filter processing.

[0050] In the example embodiment, the recorded signal often undergoes multiple processing and conversion, such as transcoding, compression, changes in playback equipment, environmental noise, etc. to become an audio signal, for example, the audio signal to be identified in the embodiment of the present application. After the above multiple processing and conversion, the audio signal to be identified will be phase distorted, thereby destroying the integrity of the watermark information. Therefore, it is necessary to perform phase compensation on each audio frame. When compensating the phase of each audio frame, since the phase is usually obtained by Fourier transform or other frequency domain analysis methods, the phase is based on is the period. When the phase change exceeds When the phase value changes from Jump back to 0. This phenomenon causes phase discontinuity, affecting the subsequent extraction and identification of watermark information. Therefore, it is necessary to unwrap the phase, that is, to restore the "wrapped" phase value (the discontinuous phase caused by periodic jumps) to a continuous phase value. Then, based on the phase difference between the first embedded reference frequency and the second embedded reference frequency of each audio frame, the phase of each audio frame is compensated to obtain the original phase audio signal before bandpass filter processing.

[0051] Step S250: extract the watermark information of the original phase audio signal and concatenate them to obtain a binary value watermark sequence.

[0052] In an exemplary embodiment, the watermark information is binary watermark information, including 0 and 1. A binary-valued watermark sequence is a complete digital sequence consisting of 0 and 1.

[0053] Optionally, the above-mentioned step S250 may include: if the corrected phase difference between the first embedded reference frequency and the second embedded reference frequency is greater than a preset threshold, the binary value is 1; if the corrected phase difference between the first embedded reference frequency and the second embedded reference frequency is less than or equal to the preset threshold, the binary value is 0; and the binary values ​​of each audio frame are spliced ​​in the order of timestamps to obtain a binary value watermark sequence.

[0054] By implementing the above-mentioned steps S210 to S250, the acquired audio signal to be identified is subjected to frame processing to obtain multiple audio frames of fixed length, which can realize independent processing of multiple audio frames, help reduce the complexity of data processing, and is suitable for batch verification of large-scale audio signals, and makes the subsequent watermark information extraction more robust; by performing discrete Fourier transform on each audio frame, and determining the maximum amplitude frequency in each audio frame as the first embedded reference frequency; by determining the frequency with complementary characteristics to the first embedded reference frequency as the second embedded reference frequency; by performing phase continuity processing on each audio frame, and performing phase compensation on each audio frame based on the phase difference between the first embedded reference frequency and the second embedded reference frequency after the phase continuity processing, the original phase audio signal before bandpass filter processing is obtained; by extracting the watermark information of the original phase audio signal and splicing it, a binary value watermark sequence is obtained; on the one hand, since the audio signal to be identified is obtained by embedding watermark signal The watermark information is embedded in the audio signal after the audio signal is processed by the band-pass filter, and the maximum amplitude frequency of the recording signal and the frequency with complementary characteristics to the maximum amplitude frequency are used to embed the watermark information, which can not only reduce the probability of the watermark information being discovered by the human ear and increase the concealment of the watermark information, so that the watermark information will not be easily discovered and tampered with; but also improve the robustness of the watermark information against the band-pass filter attack and maintain the compression resistance of the watermark information; therefore, when extracting the watermark information from the audio signal to be identified, it is necessary to perform a discrete Fourier transform on the audio signal to be identified to obtain a first embedding reference frequency and a second embedding reference frequency with complementary characteristics to the first embedding reference frequency, so as to facilitate the subsequent extraction of a complete binary value watermark sequence; on the other hand, by performing phase continuity processing and phase compensation on each audio frame, the original phase audio signal before band-pass filter processing can be obtained, which can effectively eliminate the phase distortion caused by the band-pass filter on each audio frame, and can effectively improve the success rate of watermark information extraction.

[0055] In another embodiment, in order to effectively eliminate the phase distortion caused by the bandpass filter on each audio frame, the robustness of the watermark information against the bandpass filter attack can be effectively improved, thereby improving the success rate of watermark information extraction after the bandpass filter attack; the above step S240 may include the following steps S2401 to S2405, specifically:

[0056] Step S2401, determining a third embedded reference frequency based on the first embedded reference frequency;

[0057] Step S2402: determining whether the phase of each audio frame is continuous based on the phase difference between the first embedded reference frequency and the third embedded reference frequency;

[0058] Step S2403: If the phase of each audio frame is not continuous, adjust the phase value of each audio frame to obtain an audio frame with continuous phase;

[0059] Step S2404: determining a phase distortion factor of the bandpass filter based on the first embedded reference frequency and the corresponding phase spectrum, the second embedded reference frequency and the corresponding phase spectrum, when the phase of each audio frame is continuous and the bandpass filter has a linear phase characteristic within the passband;

[0060] Step S2405 : Compensate the phase of each audio frame based on the phase distortion factor to obtain an original phase audio signal.

[0061] It can be understood from the above embodiments that Figure 3 As shown, step S1 is executed to process the audio signal Perform frame processing, that is, the audio signal Divide into multiple small frames of length N Step S2: extract each small frame Watermark information in , among which, Figure 4 As shown, step S2 includes steps S21 to S22. Step S21 is to process each small frame. Perform discrete Fourier transform to obtain each small frame , calculate each small frame Amplitude spectrum of and phase spectrum ; Then execute step S22 to find each small frame The maximum frequency of the amplitude , ,in and is the passband of the bandpass filter and determines the frequency In order to improve the robustness of the audio information after watermark information is embedded to the bandpass filter attack, the frequency , where the frequency ,frequency and frequency The relationship is shown in the following formula (1) and formula (2), the frequency and frequency The phase distortion difference at is given by the following equation (2): ;

[0062] (1);

[0063] (2);

[0064] in, , 、 and Respectively represent the frequency ,frequency and frequency Phase characteristics at ; 、 and Respectively represent the frequency ,frequency and frequency Phase distortion.

[0065] In order to compensate for the phase distortion, the extraction process assumes that the bandpass filter has a linear phase characteristic within the passband; then the phase distortion A linear function can be used Indicates that and is a constant. That is, in the above formula (2) It can be expressed as In the above formula (2), It can be expressed as .

[0066] However, in audio signal processing, phase is usually obtained through Fourier transform or other frequency domain analysis methods. For periodic, when the phase change exceeds When the phase value changes from Jump back to 0. This phenomenon will cause phase information discontinuity, affecting the subsequent extraction and analysis of watermark information. Therefore, phase unwrapping is required. and , the phase wraps (i.e., phase jumps), and the phase value needs to be adjusted according to the following formula (3) to restore its continuity.

[0067] (3)

[0068] Pass frequency and frequency The phase difference at Figure 5 The phase distortion factor can be estimated, that is, the phase distortion factor is expressed by the following formula (4):

[0069] (4)

[0070] According to the phase distortion factor, the following formula (5) is obtained:

[0071] (5)

[0072] Substituting formula (4) into formula (2), we get , that is, the influence of phase distortion caused by the bandpass filter can be eliminated, thereby restoring the original phase modulation information.

[0073] Among them, if , then it means , that is, the embedded information =0; otherwise, embed information =1.

[0074] By more accurately estimating the phase distortion and then compensating for it, it helps to maintain the detectability of the watermark information in the face of bandpass filter attacks, while ensuring the robustness of the watermark information so that the watermark information can still be successfully extracted after being processed by the bandpass filter.

[0075] In an exemplary embodiment, a method for generating an audio signal embedded with a digital watermark is provided. The method is executed by a computer device, specifically a computer device such as a terminal or a server, or a terminal and a server. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in FIG. 1 is used as an example to illustrate the process, including steps S310 to S350, wherein:

[0076] Step S310 , performing frame processing on the acquired recording signal to be processed to obtain a plurality of fixed-length frames.

[0077] Step S320 , performing discrete Fourier transform on each fixed-length frame, and determining the maximum amplitude frequency in each fixed-length frame as the first reference frequency.

[0078] It should be noted that because the maximum amplitude frequency in the video frame represents the most significant frequency component in the current frame and carries the main energy of the recorded signal, and these frequency components are more likely to be retained in subsequent signal processing (such as compression, filtering, and noise interference), embedding the watermark information in the phase of the main frequency component can avoid the loss of watermark information due to filtering out or compression of secondary frequencies, which is conducive to ensuring the integrity of the watermark information and facilitating subsequent extraction and identification of the watermark information.

[0079] Step S330: Determine a frequency that differs from the first reference frequency by an octave as a second reference frequency.

[0080] It should be noted that the first reference frequency and the second reference frequency differ by one octave, which will not produce obvious auditory distortion. After the watermark information is embedded, it is not easily detected by the human ear and can also enhance the robustness of the watermark information.

[0081] Step S340: embed watermark information into the phases of the first reference frequency and the second reference frequency to generate a fixed-length frame containing the watermark.

[0082] It should be noted that the watermark information is binary digital information, including 0 and 1.

[0083] Step S350 , inverse transform and time domain processing are performed on each fixed-length frame containing a watermark, and the fixed-length frames containing a watermark after inverse transform and time domain processing are spliced ​​to obtain an audio signal containing a binary value watermark sequence.

[0084] Understandable, such as Figure 6 As shown, step S10 is executed to record the signal Perform frame processing, that is, record the signal Split into multiple small frames of length N ; Then execute step S20, in each small frame Embed watermark information , get each small frame Then, step S30 is executed to smooth the boundary, that is, to avoid each small frame A mutation occurs at the boundary of Figure 8 The weight function shown Time domain frame With the original frame Mixing, the mixed continuous signal is obtained by the following formula (6): ; Then execute step S40, all continuous signals Splice in time sequence to restore the continuity of the audio and form a complete audio signal with watermark information .

[0085] (6)

[0086] Further, if Figure 7 As shown, the above step S20 includes steps S201 to S203, and step S201 is executed to perform Perform discrete Fourier transform and transform each small frame Convert from time domain to frequency domain and calculate each small frame Amplitude spectrum and phase spectrum ; Then execute step S202 to calculate each small frame The frequency component with the largest amplitude , ,in and is the passband of the bandpass filter, which As embedded watermark information b i The base frequency of the sound. Based on the fact that a sound is perceived as similar to another sound that is one octave away from it, the maximum amplitude frequency Another frequency is calculated by the following formula (7): ; Embed watermark information into frequency and frequency In the phase of , the watermark information is binary watermark information (0 or 1), if the initial phase is Then the phase shift after embedding the binary watermark information is Execute step S203, the modified frequency domain signal (amplitude spectrum unchanged, the phase spectrum is ) performs inverse discrete Fourier transform to obtain the time domain frame .

[0087] (7)

[0088] It should be noted that, compared to traditional watermark embedding methods that modify the amplitude, the embodiments of this application adjust the phase of the recorded signal. That is, by selecting the phase corresponding to the maximum amplitude frequency for adjustment, the watermark information is embedded in the phase of the maximum amplitude frequency without significantly affecting the sound quality. Because the maximum amplitude frequency dominates the auditory perception, its phase changes are more easily masked by the energy of the recorded signal itself, which can reduce the perceptual distortion of the recorded signal caused by the watermark information embedding. In addition, by embedding the watermark information in the phase of the maximum amplitude frequency of the recorded signal, the compression resistance of the watermark information in the generated audio signal can be guaranteed to a certain extent, making it easier to extract the watermark information later.

[0089] Based on the same inventive concept, embodiments of the present application also provide a digital watermarked audio authentication device for implementing the aforementioned digital watermarked audio authentication method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the digital watermarked audio authentication device provided below can be found in the limitations of the digital watermarked audio authentication method described above and will not be repeated here.

[0090] In an exemplary embodiment, Figure 9 As shown, an audio authentication device for embedding a digital watermark is provided, and the audio authentication device for embedding a digital watermark 900 includes:

[0091] The audio framing module 910 is configured to perform framing processing on the acquired audio signal to be authenticated, thereby obtaining a plurality of audio frames of fixed length. The audio signal to be authenticated is a recording signal that has been processed by embedding watermark information and bandpass filtering. The maximum amplitude frequency of the recording signal and a frequency complementary to the maximum amplitude frequency are used to embed the watermark information.

[0092] a transformation determination module 920 configured to perform a discrete Fourier transform on each audio frame and determine a maximum amplitude frequency in each audio frame after the discrete Fourier transform as a first embedding reference frequency;

[0093] A frequency determination module 930 is configured to determine a frequency having a complementary characteristic to the first embedded reference frequency as a second embedded reference frequency;

[0094] a phase compensation module 940 for performing phase continuity processing on each audio frame and performing phase compensation on each audio frame based on a phase difference between the first embedded reference frequency and the second embedded reference frequency after the phase continuity processing, to obtain an original phase audio signal before bandpass filter processing;

[0095] The extraction and splicing module 950 is used to extract the watermark information of the original phase audio signal and splice it to obtain a binary value watermark sequence.

[0096] As an optional implementation, the above-mentioned audio framing module 910 is specifically used to determine the total length of the audio signal to be identified; divide the audio signal to be identified based on a preset length and frame overlap rate to obtain multiple audio frames to be processed; perform windowing processing on each audio frame to be processed, and store the timestamp of each audio frame to be processed after the windowing processing to obtain multiple audio frames of fixed length.

[0097] As an optional implementation, the above-mentioned transformation determination module 920 is specifically used to convert each audio frame from time domain audio to frequency domain audio through discrete Fourier transform to obtain the amplitude spectrum and phase spectrum of each audio frame; analyze the amplitude spectrum of each audio frame, and determine the maximum amplitude frequency in each audio frame as the first embedded reference frequency.

[0098] As an optional implementation, the above-mentioned phase compensation module 940 is specifically used to determine the third embedded reference frequency based on the first embedded reference frequency; based on the phase difference between the first embedded reference frequency and the third embedded reference frequency, determine whether the phase of each audio frame is continuous; if the phase of each audio frame is not continuous, adjust the phase value of each audio frame to obtain an audio frame with continuous phase; when the phase of each audio frame is continuous and the bandpass filter has a linear phase characteristic within the passband, determine the phase distortion factor of the bandpass filter based on the first embedded reference frequency and the corresponding phase spectrum, the second embedded reference frequency and the corresponding phase spectrum; compensate the phase of each audio frame based on the phase distortion factor to obtain the original phase audio signal.

[0099] As an optional implementation, the above-mentioned extraction and splicing module 950 is specifically used to: if the corrected phase difference between the first embedded reference frequency and the second embedded reference frequency is greater than a preset threshold, the binary value is 1; if the corrected phase difference between the first embedded reference frequency and the second embedded reference frequency is less than or equal to the preset threshold, the binary value is 0; and the binary values ​​of each audio frame are spliced ​​in the order of timestamps to obtain a binary value watermark sequence.

[0100] Among them, the implementation of this embodiment, on the one hand, by performing frame processing on the acquired audio signal to be identified, a plurality of audio frames of fixed length are obtained, which can realize independent processing of the plurality of audio frames, help to reduce the complexity of data processing, and is suitable for batch verification of large-scale audio signals, and makes the subsequent extraction of watermark information more robust; on the other hand, because the audio signal to be identified is a recording signal processed by embedding watermark information and bandpass filter, and the maximum amplitude frequency of the recording signal and the frequency with complementary characteristics to the maximum amplitude frequency are used to embed watermark information, it can not only reduce the probability of the watermark information being discovered by the human ear, but also increase the concealment of the watermark information, so that the watermark information will not be easily discovered and tampering; and it can improve the robustness of the watermark information against bandpass filter attacks and maintain the compression resistance of the watermark information; therefore, when extracting the watermark information from the audio signal to be identified, it is necessary to perform a discrete Fourier transform on the audio signal to be identified to obtain a first embedded reference frequency and a second embedded reference frequency having complementary characteristics to the first embedded reference frequency, so as to facilitate the subsequent extraction of a complete binary value watermark sequence; on the other hand, by performing phase continuity processing and phase compensation on each audio frame, the original phase audio signal before bandpass filter processing can be obtained, which can effectively eliminate the phase distortion caused by the bandpass filter on each audio frame, and can effectively improve the success rate of watermark information extraction.

[0101] In an exemplary embodiment, the embodiment of the present application also provides an audio generation device embedded with a digital watermark, which can be used to: perform frame processing on the acquired recording signal to be processed to obtain multiple fixed-length frames; perform discrete Fourier transform on each fixed-length frame, and determine the maximum amplitude frequency in each fixed-length frame after discrete Fourier transform as the first reference frequency; determine the frequency with complementary characteristics to the first reference frequency as the second reference frequency; embed watermark information in the phase of the first reference frequency and the second reference frequency to generate a fixed-length frame containing a watermark; perform inverse transform and time domain processing on each fixed-length frame containing a watermark, and splice the fixed-length frames containing watermarks after inverse transform and time domain processing to obtain an audio signal containing a binary value watermark sequence.

[0102] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 10As shown. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, memory and input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store audio signals to be identified or data generated during the identification process, as well as audio recording signals to be processed or data generated during the generation process. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements a method for audio identification embedded with a digital watermark and a method for audio generation embedded with a digital watermark.

[0103] Those skilled in the art will understand that Figure 10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0104] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0105] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0106] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0107] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0108] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-mentioned embodiments. In particular, any reference to the memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0109] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, and the like.

[0110] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0111] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for audio authentication by embedding a digital watermark, characterized in that: The audio authentication method for embedding a digital watermark comprises: The obtained audio signal to be identified is subjected to frame processing to obtain multiple audio frames of fixed length; the audio signal to be identified is a recording signal that has been processed by embedding watermark information and bandpass filtering, and the maximum amplitude frequency of the recording signal and the frequency that differs from the maximum amplitude frequency by an octave are used to embed the watermark information; Performing a discrete Fourier transform on each of the audio frames, and determining a maximum amplitude frequency in each of the audio frames after the discrete Fourier transform as a first embedding reference frequency; determining a frequency that differs from the first embedded reference frequency by an octave as a second embedded reference frequency; performing phase continuity processing on each of the audio frames, and performing phase compensation on each of the audio frames based on a phase difference between the first embedded reference frequency and the second embedded reference frequency after the phase continuity processing, to obtain an original phase audio signal before bandpass filter processing; The watermark information of the original phase audio signal is extracted and concatenated to obtain a binary value watermark sequence.

2. The audio authentication method of embedding digital watermark according to claim 1, characterized in that: The step of performing frame processing on the acquired audio signal to be identified to obtain a plurality of audio frames of fixed length includes: Determining the total length of the audio signal to be identified; Dividing the audio signal to be identified based on a preset length and a frame overlap rate to obtain a plurality of audio frames to be processed; Windowing processing is performed on each of the audio frames to be processed, and a timestamp of each of the audio frames to be processed after the windowing processing is stored to obtain a plurality of audio frames of fixed length.

3. The audio authentication method of embedding digital watermark according to claim 1, characterized in that: The performing discrete Fourier transform on each of the audio frames and determining the maximum amplitude frequency in each of the audio frames after the discrete Fourier transform as the first embedded reference frequency includes: Convert each of the audio frames from time-domain audio to frequency-domain audio by discrete Fourier transform to obtain an amplitude spectrum and a phase spectrum of each of the audio frames; The amplitude spectrum of each of the audio frames is analyzed, and the maximum amplitude frequency in each of the audio frames is determined as the first embedding reference frequency.

4. The audio authentication method of embedding digital watermark according to claim 3, characterized in that: The performing phase continuity processing on each of the audio frames, and performing phase compensation on each of the audio frames based on the phase difference between the first embedded reference frequency and the second embedded reference frequency after the phase continuity processing to obtain the original phase audio signal before bandpass filter processing, includes: determining a third embedded reference frequency based on the first embedded reference frequency; determining whether the phase of each of the audio frames is continuous based on a phase difference between the first embedded reference frequency and the third embedded reference frequency; If the phase of each of the audio frames is not continuous, adjusting the phase value of each of the audio frames to obtain the audio frames with continuous phase; determining a phase distortion factor of the bandpass filter based on the first embedded reference frequency and the corresponding phase spectrum, and the second embedded reference frequency and the corresponding phase spectrum, when the phase of each audio frame is continuous and the bandpass filter has a linear phase characteristic within the passband; The phase of each audio frame is compensated based on the phase distortion factor to obtain an original phase audio signal.

5. The audio authentication method of embedding digital watermark according to claim 4, characterized in that: The step of extracting the watermark information of the original phase audio signal and splicing the original phase audio signal to obtain a binary value watermark sequence includes: If the corrected phase difference between the first embedded reference frequency and the second embedded reference frequency is greater than a predetermined threshold, the binary value is 1; If the corrected phase difference between the first embedded reference frequency and the second embedded reference frequency is less than or equal to a preset threshold, the binary value is 0; The binary value of each audio frame is concatenated in a timestamp order to obtain the binary value watermark sequence.

6. A method for generating audio embedded with a digital watermark, characterized in that: The method for generating audio embedded with a digital watermark comprises: Performing frame processing on the acquired recording signal to be processed to obtain multiple fixed-length frames; Performing a discrete Fourier transform on each of the fixed-length frames, and determining a maximum amplitude frequency in each of the fixed-length frames after the discrete Fourier transform as a first reference frequency; determining a frequency that differs from the first reference frequency by an octave as a second reference frequency; embedding watermark information in the phases of the first reference frequency and the second reference frequency to generate a fixed-length frame containing the watermark; Each of the fixed-length watermarked frames is inversely transformed and time-domain processed, and the fixed-length watermarked frames after inverse transformation and time-domain processing are spliced ​​to obtain an audio signal containing a binary-valued watermark sequence.

7. An audio authentication device embedded with a digital watermark, characterized in that: The audio identification device embedded with a digital watermark comprises: an audio framing module configured to perform framing processing on the acquired audio signal to be authenticated, thereby obtaining a plurality of audio frames of fixed length; the audio signal to be authenticated is a recording signal that has been processed by embedding watermark information and bandpass filtering, wherein the maximum amplitude frequency of the recording signal and a frequency that differs from the maximum amplitude frequency by an octave are used to embed the watermark information; a transformation determination module, configured to perform a discrete Fourier transform on each of the audio frames, and determine a maximum amplitude frequency in each of the audio frames after the discrete Fourier transform as a first embedding reference frequency; a frequency determination module, configured to determine a frequency that differs from the first embedded reference frequency by an octave as a second embedded reference frequency; a phase compensation module, configured to perform phase continuity processing on each of the audio frames, and perform phase compensation on each of the audio frames based on a phase difference between the first embedded reference frequency and the second embedded reference frequency after the phase continuity processing, to obtain an original phase audio signal before bandpass filter processing; The extraction and splicing module is used to extract the watermark information of the original phase audio signal and splice it to obtain a binary value watermark sequence.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the method for audio authentication with embedded digital watermarks as described in any one of claims 1 to 5, and to implement the steps of the method for audio generation with embedded digital watermarks as described in claim 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the computer program implements the steps of the method for authenticating an audio program embedded with a digital watermark according to any one of claims 1 to 5, or implements the steps of the method for generating an audio program embedded with a digital watermark according to claim 6.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the steps of the method for authenticating an audio program embedded with a digital watermark according to any one of claims 1 to 5, or implements the steps of the method for generating an audio program embedded with a digital watermark according to claim 6.

Images