Luiz W. P. Biscainho

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 efficient model-based multirate method for reconstruction of audio signals across long gaps

This paper presents an efficient model-based interpolator for reconstructing long portions of missing samples in audio signals. A simple modification to an existing interpolation method is proposed, optimizing the balance between computational requirements and qualitative performance. Moreover, a post-processing multirate interpolation scheme is introduced to further enhance the quality of the restored signals. Subjective quality assessments of the reconstructed signals via listening tests substantiate the improved performance of the proposed algorithm

Optimal variable step size for the LMS/Newton algorithm with application to subband adaptive filtering

An efficient approach for the computation of the optimum convergence factor for the LMS (least mean square)/Newton algorithm applied to a transversal FIR structure is proposed. The approach leads to a variable step size algorithm that results in a dramatic reduction in convergence time. The algorithm is evaluated in system identification applications where two alternative implementations of the adaptive filter are considered: the conventional transversal FIR realization and adaptive filtering in subbands. >

On the quality assessment of sound signals

This paper constitutes an introduction to the field of quality evaluation of sound (speech and audio) signals. The need for such an assessment is inherent to modern communications: VoIP, mobile phone, or teleconference systems require meaningful measures of performance, which may ultimately assure good service or profitable business. A brief survey on subjective and objective evaluation methods is provided. Recent developments as well as new topics to be investigated are also addressed. Experiments are conducted to illustrate how to validate quality assessment methods.

High-selectivity filter banks for spectral analysis of music signals

This paper approaches, under a unified framework, several algorithms for the spectral analysis of musical signals. Such algorithms include the fast Fourier transform (FFT), the fast filter bank (FFB), the constant- transform (CT), and the bounded- transform (BT), previously known from the associated literature. Two new methods are then introduced, namely, the constant- fast filter bank (CFFB) and the bounded- fast filter bank (BFFB), combining the positive characteristics of the previously mentioned algorithms. The provided analyses indicate that the proposed BFFB achieves an excellent compromise between the reduced computational effort of the FFT, the high selectivity of each output channel of the FFB, and the efficient distribution of frequency channels associated to the CT and BT methods. Examples are included to illustrate the performances of these methods in the spectral analysis of music signals.

Edit Detection in Speech Recordings via Instantaneous Electric Network Frequency Variations

In this paper, an edit detection method for forensic audio analysis is proposed. It develops and improves a previous method through changes in the signal processing chain and a novel detection criterion. As with the original method, electrical network frequency (ENF) analysis is central to the novel edit detector, for it allows monitoring anomalous variations of the ENF related to audio edit events. Working in unsupervised manner, the edit detector compares the extent of ENF variations, centered at its nominal frequency, with a variable threshold that defines the upper limit for normal variations observed in unedited signals. The ENF variations caused by edits in the signal are likely to exceed the threshold providing a mechanism for their detection. The proposed method is evaluated in both qualitative and quantitative terms via two distinct annotated databases. Results are reported for originally noisy database signals as well as versions of them further degraded under controlled conditions. A comparative performance evaluation, in terms of equal error rate (EER) detection, reveals that, for one of the tested databases, an improvement from 7% to 4% EER is achieved, respectively, from the original to the new edit detection method. When the signals are amplitude clipped or corrupted by broadband background noise, the performance figures of the novel method follow the same profile of those of the original method.

AR model estimation from quantized signals

This letter investigates the deviation of least squares AR-model estimates induced by linear quantization of the observable signal under modeling. Model correction is attained by describing the undesirable bias as a function of the number of bits used to represent the signal. Sensitivity considerations are employed to explain the behavior of the deviations on the generator filter poles. Numerical examples are included.

Robust Acoustic Self-Localization of Mobile Devices

Self-localization of smart portable devices serves as foundation for several novel applications. This work proposes a set of algorithms that enable a mobile device to passively determine its position relative to a known reference with centimeter precision, based exclusively on the capture of acoustic signals emitted by controlled sources around it. The proposed techniques tackle typical practical issues such as reverberation, unknown speed of sound, line-of-sight obstruction, clock skew, and the need for asynchronous operation. After their theoretical developments and off-line simulations, the methods are assessed as real-time applications embedded into off-the-shelf mobile devices operating in real scenarios. When line of sight is available, position estimation errors are at most 4 cm using recorded signals.

Bayesian Restoration of Audio Signals Degraded by Impulsive Noise Modeled as Individual Pulses

Impulsive noise, also known as clicks, is a very common type of distortion in old gramophone recordings. Existing methods (both heuristic and statistical) for removal of this type of defect usually do not exploit its underlying physical generation. This work proposes a model in which each click is individually modeled, which is more physically meaningful. A Bayesian method based on the reversible-jump Metropolis-Hastings algorithm for joint detection and removal of impulsive noise in audio signals is devised. Simulations with artificial and real audio signals as well as comparisons with competing approaches are presented to illustrate and validate the proposed method.

HRTF Interpolation Through Direct Angular Parameterization

Under anechoic conditions, a system that generates 3D binaural sound should simulate the head-related transfer functions (HRTFs) that represent sound changes from the source to the listeners ears. Such a system could directly use HRTFs measured over a sphere around the listener. However, since it is not feasible to perform measurements for every position, practical systems interpolate the necessary functions between measured HRTFs. This paper proposes a generalization of the continuous variable digital delay (CVDD) described in the literature, yielding a two spatial-variable dependent transfer function. It performs a polynomial interpolation whose coefficients are optimized on a region over the sphere, so that the angular coordinates of any point inside the region map directly into the corresponding HRTFs. Performance assessment is done through graphical means. The issue of continuity in moving sound generation is also approached, through the combination of adjacent-region structures