Audio Authentication Research Articles

Audio authenticity: Duplicated audio segment detection in waveform audio file

Waveform audio (WAV) file is a widely used file format of uncompressed audio. With the rapid development of digital media technology, one can easily insert duplicated segments with powerful audio editing software, e.g. inserting a segment of audio with negative meaning into the existing audio file. The duplicated segments can change the meaning of the audio file totally. So for a WAV file to be used as evidence in legal proceedings and historical documents, it is very importance to identify if there are any duplicated segments in it. This paper proposes a method to detect duplicated segments in a WAV file. Our method is based on the similarity calculation between two different segments. Duplicated segments are prone to having similar audio waveform, i.e., a high similarity. We use fast convolution algorithm to calculate the similarity, which makes our method quit efficient. We calculate the similarity between any two different segments in a digital audio file and use the similarity to judge which segments are duplicated. Experimental results show the feasibility and efficiency of our method on detecting duplicated audio segments.

법음성학에서의 오디오 신호의 위변조 구간 자동 검출 방법 연구

We propose a novel scheme for digital audio authentication of given audio files which are edited by inserting small audio segments from different environmental sources. The purpose of this research is to detect inserted sections from given audio files. We expect that the proposed method will assist human investigators by notifying suspected audio section which considered to be recorded or transmitted on different environments. GMM-UBM and GSV-SVM are applied for modeling the dominant environment of a given audio file. Four kinds of likelihood ratio based scores and SVM score are used to measure the likelihood for a dominant environment model. We also use an ensemble score which is a combination of the aforementioned five kinds of scores. In the experimental results, the proposed method shows the lowest average equal error rate when we use the ensemble score. Even when dominant environments were unknown, the proposed method gives a similar accuracy.

A game-theoretic analysis of preventing spam over Internet Telephony via audio CAPTCHA-based authentication

Spam over Internet Telephony SPIT is a potential source of disruption in Voice over IP VoIP systems. The use of anti-SPIT mechanisms, such as filters and audio CAPTCHA Completely Automated Public T...

A game-theoretic analysis of preventing spam over Internet Telephony via audio CAPTCHA-based authentication

Spam over Internet Telephony SPIT is a potential source of disruption in Voice over IP VoIP systems. The use of anti-SPIT mechanisms, such as filters and audio CAPTCHA Completely Automated Public Turing Test to Tell Computer and Humans Apart can prevent unsolicited calls and lead to less unwanted traffic. In this paper, we present a game-theoretic model, in which the game is played between SPIT senders and internet telephony users. The game includes call filters and audio CAPTCHA, so as to classify incoming calls as legitimate or malicious. We show how the resulting model can be used to decide upon the trade-offs present in this problem and help us predict the SPIT sender's behavior. We also highlight the advantages in terms of SPIT call reduction of merely introducing CAPTCHA, and provide experimental verification of our results.

Authenticity examination of compressed audio recordings using detection of multiple compression and encoders’ identification

Since the appearance of digital audio recordings, audio authentication has been becoming increasingly difficult. The currently available technologies and free editing software allow a forger to cut or paste any single word without audible artifacts. Nowadays, the only method referring to digital audio files commonly approved by forensic experts is the ENF criterion. It consists in fluctuation analysis of the mains frequency induced in electronic circuits of recording devices. Therefore, its effectiveness is strictly dependent on the presence of mains signal in the recording, which is a rare occurrence. Recently, much attention has been paid to authenticity analysis of compressed multimedia files and several solutions were proposed for detection of double compression in both digital video and digital audio. This paper addresses the problem of tampering detection in compressed audio files and discusses new methods that can be used for authenticity analysis of digital recordings. Presented approaches consist in evaluation of statistical features extracted from the MDCT coefficients as well as other parameters that may be obtained from compressed audio files. Calculated feature vectors are used for training selected machine learning algorithms. The detection of multiple compression covers up tampering activities as well as identification of traces of montage in digital audio recordings. To enhance the methods’ robustness an encoder identification algorithm was developed and applied based on analysis of inherent parameters of compression. The effectiveness of tampering detection algorithms is tested on a predefined large music database consisting of nearly one million of compressed audio files. The influence of compression algorithms’ parameters on the classification performance is discussed, based on the results of the current study.

Audio Watermarking Scheme Using IMFs and HHT for Forensic Applications

Audio signals and applications are numerous and ubiquitous. Most of these applications especially those on the Internet require authentication and proof(s) of ownership. There are several efficient methods in the literature address these crucial and critical concerns. In this paper, the authors present a new non-blind audio watermarking scheme for forensic audio authentication and proof of ownership. The proposed scheme is based on empirical mode decomposition and Hilbert Haung Transformation (HHT). In this method, the audio signal is decomposed into frames of 1024 sample each. These frames are further decomposed into its several mono-component signals called Intrinsic Mode Functions (IMF). These Intrinsic Mode Functions will serve as the addressee for the watermark. In this research, the chosen watermark is a pseudo random number generated by Matlab-7, which is added to the highest and lowest IMFs of each frame of the decomposed signal. This is done to accommodate for time scale modification attacks as well as MP3 compression respectively. Experimental results show that the watermarked audio signals maintained high fidelity of more than 20 dBs which meets the International Federation of Phonographic Industry requirements. The results also show that the proposed scheme is robust against signal processing attacks such as MP3, time scale modification, and resizing attacks.

Research on Digital Audio Authenticity Analysis

Removed due to poor citations and verbatim copying of texts from the following original papers: “Cooper, A. J. (2006) Detection of Copies of Digital Audio Recordings for Forensic Purposes, PhD Thesis, The Open University, Faculty of Technology, Department of Information and Communication Technology, Milton Keynes, UK”; “Daniel Lawn Rappaport. (2012) ESTABLISHING A STANDARD FOR DIGITAL AUDIO AUTHENTICITY: A CRITICAL ANALYSIS OF TOOLS, METHODOLOGIES, AND CHALLENGES, PhD [sic] Thesis”; “Bruce E. Koenig and Douglas S. Lacey – Forensic Authentication of Digital Audio Recordings, J. Audio Eng. Soc., Vol. 57, No. 9, 2009.”

An Improved Discrete Fourier Transform-Based Algorithm for Electric Network Frequency Extraction

This paper introduces a Discrete Fourier Transform (DFT)-based algorithm to extract the Electric Network Frequency (ENF) information from an audio recording for use in audio authentication. The basic idea of the proposed algorithm is to calculate the specific spectral lines by DFT in the frequency domain at the desired frequency point instead of throughout the entire frequency band. Then a binary search technique is employed to search the next desired frequency bin to repeat the spectral line calculation until the hidden ENF information is extracted. The purpose is to improve the accuracy and precision of conventional ENF extraction methods and also to enhance the calculation efficiency. Both simulated audio signals with different signal-to-noise ratios (SNRs) and actual audio recordings are studied to verify the performance of the proposed algorithm. Two error-evaluation criteria, frequency offset and frequency bias, are defined to evaluate the algorithm performance on accuracy and precision. The test results and the error evaluation prove the validation and demonstrate the improvement of the proposed algorithm.

Music content authentication based on beat segmentation and fuzzy classification

Digital audio has been ubiquitous over the past decade. Since it can be easily modified by editing tools, there has been a strong need to protect its content for secure multimedia applications. Previous audio authentication algorithms are mainly focused on either human speech or general audio with music as part of the test data, while special research on music authentication has been somewhat neglected. In this article, we propose a novel algorithm to protect the integrity and authenticity of music signals. Its main contributions include the following: (1) Music is segmented into beat-based frames, which not only endows the authentication units with more semantic meaning but also perfectly resolves the challenging synchronization problem. (2) Robust hashes are generated from chroma-based mid-level audio feature which can appropriately characterize the music content and integrated with an encryption procedure to ensure the security against malicious block-wise vector quantization attack. (3) Fuzzy logic is adopted to make the authentication decision in the light of three measures defined on bit errors, coinciding with the inherent blurred nature of authentication. The experiments exhibit good discriminative ability between admissible and malicious operations.

Application of Power System Frequency for Digital Audio Authentication

Frequency is an important parameter for the operation and control of power systems. One novel application of frequency measurements involves using this information to authenticate a digital audio/video recording presented as forensic evidence in legal proceedings. To apply this technique, called the electrical network frequency (ENF) criterion, both a reference frequency database and an accurate frequency estimation method are required. This paper briefly introduces a wide-area frequency monitoring network (FNET) as the reference frequency database and analyzes statistical features of frequency of the four North American interconnections in terms of different time scales. Combined with digital filtering and a three-sample interpolation, the short-time Fourier transform (STFT) is adjusted to estimate the ENF embedded in digital audio recordings. A procedure of using the ENF criterion, ranging from signal preprocessing to ENF estimation and reference frequency database matching, is then proposed. Further, oscillator error of the digital recording device is considered and an iterative error correction method is given to assist with the frequency database matching. Factors which influence the accuracy of frequency estimation, such as parameter selections of the STFT and signal-to-noise ratio, are also discussed. Test results show that the procedure is capable of performing digital audio authentication.

Pseudo-zernike moments-based audio content authentication algorithm robust against feature-analysed substitution attack

Content-based audio content authentication algorithms provide a method to solve the veracity and integrity of audio content. On the basic of pseudo-Zernike moments, an audio content authentication algorithm robust against feature-analysed substitution attack is proposed, which is aimed at some insecure issues in the existing content-based audio content authentication schemes. Firstly, the audio signal is cut into non-overlapping frames and each frame is divided into two segments, and each segment is scrambled. Then, synchronization codes generated by pseudo random sequence and watermark bits generated by pseudo-Zernike moments are embedded in the first and second segment, respectively, which are completed by quantizing the modulus of pseudo-Zernike moments. The scrambled segments used to generate and extract watermark are unknown to attackers. So, it is difficult for attackers to get the watermark generated and extracted to perform feature-analysed substitution attack. The synchronization code and watermark embedding method proposed is inaudible and has excellent ability to tolerance against common signal processing operations. Compared with the existing audio watermark algorithms based on pseudo-Zernike moments, the algorithm increases the embedding capacity and improves the security of the watermarking system.

A fingerprint-based audio authentication scheme using frequency domain statistical characteristic

Recently, a major problem faced by digital data providers and owners is protecting data from malicious falsification and distribution. As a solution to this problem, digital fingerprinting technique is now attracting attention as new method for integrity protection. According to frequency domain statistical characteristic, an efficient fingerprint-based audio authentication scheme is proposed in this paper. Initially, the host audio signal is equally divided into non-overlapping frames, moreover, each audio frame is then equally split into non-overlapping segments, and the DFT (discrete Fourier transform) domain relation of each audio segment is computed. Then DFT domain relations of each audio frame are adopted to generate the corresponding binary pattern using non-uniform quantization. Finally, the fingerprint is obtained by performing XOR operation between the binary pattern and binary pseudo-random sequence. Simulation results demonstrate the effectiveness of our scheme in terms of inaudibility, detection reliability, robustness against some non-malicious signal processing operations and vulnerability against malicious falsification.

Current Developments and Future Trends in Audio Authentication

Recent developments in the audio- authentication field include basic, preliminary audio analysis and advanced audio- authentication techniques that exploit audio recording conditions and compressed audio features. Multimedia is involved in every aspect of our lives, and many of us rely on various websites for information on events taking place all over the world. We form opinions based on the content of camera footage, phone conversations, and other recordings. Although some sources give us authentic information, others contain forged content. Concerns regarding how to authenticate multimedia data have led to research in forgery detection, but studies in audio are generally still limited compared to those in image and video.

A multipurpose audio watermarking algorithm with synchronization and encryption

We propose a new multipurpose audio watermarking scheme in which two complementary watermarks are used. For audio copyright protection, the watermark data with copyright information or signature are first encrypted by Arnold transformation. Then the watermark data are inserted in the low frequency largest significant discrete cosine transform (DCT) coefficients to obtain robustness performance. For audio authentication, a chaotic signal is inserted in the high frequency insignificant DCT coefficients to detect tampered regions. Furthermore, the synchronization code is embedded in the audio statistical characteristics to resist desynchronization attacks. Experimental results show that our proposed method can not only obtain satisfactory detection and tampered location, but also achieve imperceptibility and robustness to common signal processing attacks, such as cropping, shifting, and time scale modification (TSM). Comparison results show that our method outperforms some existing methods.

An Inconclusive Digital Audio Authenticity Examination: A Unique Case

This case report sets forth an authenticity examination of 35 encrypted, proprietary-format digital audio files containing recorded telephone conversations between two codefendants in a criminal matter. The codefendant who recorded the conversations did so on a recording system he developed; additionally, he was both a forensic audio authenticity examiner, who had published and presented in the field, and was the head of a professional audio society's writing group for authenticity standards. The authors conducted the examination of the recordings following nine laboratory steps of the peer-reviewed and published 11-step digital audio authenticity protocol. Based considerably on the codefendant's direct involvement with the development of the encrypted audio format, his experience in the field of forensic audio authenticity analysis, and the ease with which the audio files could be accessed, converted, edited in the gap areas, and reconstructed in such a way that the processes were undetected, the authors concluded that the recordings could not be scientifically authenticated through accepted forensic practices.

Audio Authenticity: Detecting ENF Discontinuity With High Precision Phase Analysis

This paper addresses a forensic tool used to assess audio authenticity. The proposed method is based on detecting phase discontinuity of the power grid signal; this signal, referred to as electric network frequency (ENF), is sometimes embedded in audio signals when the recording is carried out with the equipment connected to an electrical outlet or when certain microphones are in an ENF magnetic field. After down-sampling and band-filtering the audio around the nominal value of the ENF, the result can be considered a single tone such that a high-precision Fourier analysis can be used to estimate its phase. The estimated phase provides a visual aid to locating editing points (signalled by abrupt phase changes) and inferring the type of audio editing (insertion or removal of audio segments). From the estimated values, a feature is used to quantify the discontinuity of the ENF phase, allowing an automatic decision concerning the authenticity of the audio evidence. The theoretical background is presented along with practical implementation issues related to the proposed technique, whose performance is evaluated on digitally edited audio signals.

An automated approach to the Electric Network Frequency (ENF) criterion - Theory and practice

A relatively recent forensic technique developed to help establish the authenticity of audio and video recordings is the Electric Network Frequency (ENF) criterion. This paper confirms the applicability of the ENF criterion for use in mainland UK and describes an automated approach to matching ENF estimates taken from a questioned recording to a database of ENF values. It will be shown that the automated approach has the added advantage of producing statistical data capable of indicating the strength of the match. Consideration has been given to the robust extraction of ENF signals from evidential recordings and the resulting system and signal processing procedures outlined have been used successfully by the Metropolitan Police Forensic Audio Laboratory in London to extract and match ENF data from evidential recordings to an ENF database.

Adaptive Signal-Dependent Audio Watermarking Based on Human Auditory System and Neural Networks

Based the characteristics of the human auditory system (HAS) and the techniques of neural networks, this work proposes an Adaptive Signal-Dependent Audio Watermarking (ASDAW) technique for protecting audio copyrights. First, a signal-dependent watermark for the ASDAW technique is generated by using the characteristics of the HAS (the temporal and frequency maskings). Next, the signal-dependent watermark is hidden in an original audio on the temporal domain. The ASDAW technique can make the signal-dependent audio watermark imperceptive (inaudible) because of the characteristics of the HAS. Moreover, an artificial neural network (ANN) is trained in the ASDAW technique so that the ASDAW technique can memorize the relationships between an original audio and the corresponding watermarked audio. Using the trained ANN (TANN), the ASDAW technique can extract the signal-dependent watermarks without the original audio. The extracted watermarks are then exploited in verifying legal duplications made of an audio during audio authentication. Consequently, the copyright forgery for audio can be suppressed greatly. Furthermore, experimental results illustrate that the ASDAW technique significantly possesses memorized, adaptive, and robust capabilities, making it immune against common audio manipulations and pirate attacks for counterfeiting audio copyrights.