Acoustic Impulse Response Research Articles

An improved method to determine the onset timings of reflections in an acoustic impulse response

Determining the absolute onset time of reflections in an acoustic impulse response (IR) has applications for both subjective and physical acoustics problems. Although computationally simple, a first-order energetic analysis of the IR can lead to false-positive identification of reflections. This letter reports on a method to determine reflection onset timings using a modified running local kurtosis analysis to identify regions in the IR where the distribution is non-normal. IRs from both real and virtual rooms are used to validate the method and to find optimum analysis window sizes.

Pattern recognition of eggshell crack using PCA and LDA

The eggshell crack was detected with dynamic frequency response and an egg was excited by a light mechanical impact on different locations of the eggshell. The dominant resonance frequency can be observed using flexible piezoelectric film sensors. The results showed that some frequencies were higher for the cracked eggs, whereas the dominant frequency value was lower for the intact eggs. In normalized power spectrum, the first 10 or 20 features were extracted based on interval frequency (IFM), maximum magnitudes in turn (HVM) and frequencies by magnitudes in turn (MVF), and were used as input of pattern recognition algorithms The pattern recognition was conducted by Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). Intact egg and cracked egg could be distinguished by LDA and PCA using HVM and MVF. More distinguishing effects were obtained based on LDA using 10 MVFs. The examination of cracks on eggshell was usually conducted by floodlighting before, but it gives eye fatigue, makes misjudgment and is not easy to detect hairline crack. The acoustic impulse response method was suggested in this research to measure hairline crack for eggshell using flexible piezoelectric film sensors. The result was found that the acoustic impulse response method can distinguish between intact egg and cracked egg. This research provides a technology detection of cracked egg.

Multichannel Online Blind Speech Dereverberation with Marginalization of Static Observation Parameters in a Rao-Blackwellized Particle Filter

Room reverberation leads to reduced intelligibility of audio signals and spectral coloration of audio signals. Enhancement of acoustic signals is thus crucial for high-quality audio and scene analysis applications. Multiple sensors can be used to exploit statistical evidence from multiple observations of the same event to improve enhancement. Whilst traditional beamforming techniques suffer from interfering reverberant reflections with the beam path, other approaches to dereverberation often require at least partial knowledge of the room impulse response which is not available in practice, or rely on inverse filtering of a channel estimate to obtain a clean speech estimate, resulting in difficulties with non-minimum phase acoustic impulse responses. This paper proposes a multi-sensor approach to blind dereverberation in which both the source signal and acoustic channel are directly estimated from the distorted observations using their optimal estimators. The remaining model parameters are sampled from hypothesis distributions using a particle filter, thus facilitating real-time dereverberation. This approach was previously successfully applied to single-sensor blind dereverberation. In this paper, the single-channel approach is extended to multiple sensors. Performance improvements due to the use of multiple sensors are demonstrated on synthetic and baseband speech examples.

New Insights Into the MVDR Beamformer in Room Acoustics

The minimum variance distortionless response (MVDR) beamformer, also known as Capon's beamformer, is widely studied in the area of speech enhancement. The MVDR beamformer can be used for both speech dereverberation and noise reduction. This paper provides new insights into the MVDR beamformer. Specifically, the local and global behavior of the MVDR beamformer is analyzed and novel forms of the MVDR filter are derived and discussed. In earlier works it was observed that there is a tradeoff between the amount of speech dereverberation and noise reduction when the MVDR beamformer is used. Here, the tradeoff between speech dereverberation and noise reduction is analyzed thoroughly. The local and global behavior, as well as the tradeoff, is analyzed for different noise fields such as, for example, a mixture of coherent and non-coherent noise fields, entirely non-coherent noise fields and diffuse noise fields. It is shown that maximum noise reduction is achieved when the MVDR beamformer is used for noise reduction only. The amount of noise reduction that is sacrificed when complete dereverberation is required depends on the direct-to-reverberation ratio of the acoustic impulse response between the source and the reference microphone. The performance evaluation supports the theoretical analysis and demonstrates the tradeoff between speech dereverberation and noise reduction. When desiring both speech dereverberation and noise reduction, the results also demonstrate that the amount of noise reduction that is sacrificed decreases when the number of microphones increases.

Eggshell crack detection based on acoustic impulse response and supervised pattern recognition

A system based on acoustic resonance was developed for eggshell crack detection. It was achieved by the analysis of the measured frequency response of eggshell excited with a light mechanism. The response signal was processed by recursive least squares adaptive filter, which resulted in the signal-to-noise ratio of the acoustic impulse response reing remarkably enhanced. Five features variables were exacted from the response frequency signals. To develop a robust discrimination model, three pattern recognition algorithms (i.e. K-nearest neighbours, artificial neural network, and support vector machine) were examined comparatively in this work. Some parameters of the model were optimised by cross-validation in the building model. The experimental results showed that the performance of the support vector machine model is the best in comparison to k-nearest neighbours and artificial neural network models. The optimal support vector machine model was obtained with the identification rates of 95.1% in the calibration set, and 97.1% in the prediction set, respectively. Based on the results, it was concluded that the acoustic resonance system combined with the supervised pattern recognition has a significant potential for the cracked eggs detection.

Acoustic impulse response in apple tissues treated by pulsed electric field

This work discusses application of acoustically-derived index of disintegration for characterization of pulsed electric field (PEF)-treated whole apples in aqueous media. It was demonstrated that PEF efficiency was dependent on the conductivity of the aqueous media and orientation of the apple with respect to the external field. Electrical treatment also stimulated transient changes after PEF application in acoustic disintegration index (Za). Relation between the conductivity disintegration index (Zc) and Za was established. A non-linear relation between Zc and Za was observed and it was demonstrated that the Za value seems to be a better measure of the real degree of damage than the widely used value of Zc.

A Class of Sparseness-Controlled Algorithms for Echo Cancellation

In the context of acoustic echo cancellation (AEC), it is shown that the level of sparseness in acoustic impulse responses can vary greatly in a mobile environment. When the response is strongly sparse, convergence of conventional approaches is poor. Drawing on techniques originally developed for network echo cancellation (NEC), we propose a class of AEC algorithms that can not only work well in both sparse and dispersive circumstances, but also adapt dynamically to the level of sparseness using a new sparseness-controlled approach. Simulation results, using white Gaussian noise (WGN) and speech input signals, show improved performance over existing methods. The proposed algorithms achieve these improvement with only a modest increase in computational complexity.

Blind deconvolution of quadratic time-frequency representations of musical signals for reverberation feature extraction.

An acoustic space is uniquely characterized by its reverberant behavior. Due to the complexity of the multiple reflections and diffraction of sound in an enclosure, currently available system identification algorithms cannot effectively estimate the impulse response of a concert hall simply from musical recordings in that space. This paper reports the use of blind image deconvolution methods to construct echo patterns from quadratic time-frequency representations of reverberant recordings of music. First, a quadratic time-frequency analysis is performed to decompose the musical signal into its constituent harmonic components. Quadratic time-frequency analysis methods are known to give enhanced resolution in the time-frequency plane in comparison to conventional Fourier analysis. Reverberant features then appear as blur in the time-frequency plane, which can be estimated by employing the methods of blind deconvolution of this “image” of the sound. The proposed algorithm retrieves both the blur pattern, which corresponds to the reverberation echogram, and the direct acoustic signal. By choosing the time-frequency frame scale and smoothing window, a multi-resolution analysis of the underlying acoustic impulse response is obtained. Various quadratic time-frequency analysis methods are evaluated and their relative performance is reported. The proposed methods are also compared to existing dereverberation algorithms.

Eggshell crack detection based on acoustic response and support vector data description algorithm

A system based on acoustic resonance and combined with pattern recognition was attempted to discriminate cracks in eggshell. Support vector data description (SVDD) was employed to solve the classification problem due to the imbalanced number of training samples. The frequency band was between 1,000 and 8,000 Hz. Recursive least squares adaptive filter was used to process the response signal. Signal-to-noise ratio of acoustic impulse response was remarkably enhanced. Five characteristics descriptors were extracted from response frequency signals, and some parameters were optimized in building model. Experiment results showed that in the same condition SVDD got better performance than conventional classification methods. The performance of SVDD model was achieved with crack detection level of 90% and a false rejection level of 10% in the prediction set. Based on the results, it can be concluded that the acoustic resonance system combined with SVDD has significant potential in the detection of cracked eggs.

EVALUATION OF TOMATO QUALITY DURING STORAGE BY ACOUSTIC IMPULSE RESPONSE

ABSTRACT The aim of this research was to evaluate the ability of acoustic impulse response to monitor “Heatwave” tomato quality during storage. The tomato was excited and the response signal is captured using the film sensors. The results showed that acoustic impulse response parameters (dominant frequency, firmness index and elasticity coefficient) of “Heatwave” tomato decreased with storage time. The dominant frequency, firmness index and elasticity coefficient had no correlation with soluble solids contents or pH, but had a good correlation with compression force and puncture force. There was a better correlation between elasticity coefficient and compression force (or puncture force); hence, elasticity coefficient could be a better index of “Heatwave” tomato firmness. The acoustic impulse response is a potential factor in order to determine the firmness of “Heatwave” tomato. PRACTICAL APPLICATIONS Tomato firmness is an important characteristic because firmness of tomato flesh influences the mouth sensory feel. This results showed that acoustic impulse response parameters (dominant frequency, firmness index and elasticity coefficient) of “Heatwave” tomato decreased with storage time. The dominant frequency, firmness index and elasticity coefficient had not correlation with soluble solids contents or pH, but had a good correlation with compression force and puncture force. There were better correlation between elasticity coefficient and compression force (or puncture force), and hence elasticity coefficient could be a better index of “Heatwave” tomato firmness. The acoustic impulse response is potential to determine the firmness of “Heatwave” tomato.

Multichannel Eigenspace Beamforming in a Reverberant Noisy Environment With Multiple Interfering Speech Signals

In many practical environments we wish to extract several desired speech signals, which are contaminated by nonstationary and stationary interfering signals. The desired signals may also be subject to distortion imposed by the acoustic room impulse responses (RIRs). In this paper, a linearly constrained minimum variance (LCMV) beamformer is designed for extracting the desired signals from multimicrophone measurements. The beamformer satisfies two sets of linear constraints. One set is dedicated to maintaining the desired signals, while the other set is chosen to mitigate both the stationary and nonstationary interferences. Unlike classical beamformers, which approximate the RIRs as delay-only filters, we take into account the entire RIR [or its respective acoustic transfer function (ATF)]. The LCMV beamformer is then reformulated in a generalized sidelobe canceler (GSC) structure, consisting of a fixed beamformer (FBF), blocking matrix (BM), and adaptive noise canceler (ANC). It is shown that for spatially white noise field, the beamformer reduces to a FBF, satisfying the constraint sets, without power minimization. It is shown that the application of the adaptive ANC contributes to interference reduction, but only when the constraint sets are not completely satisfied. We show that relative transfer functions (RTFs), which relate the desired speech sources and the microphones, and a basis for the interference subspace suffice for constructing the beamformer. The RTFs are estimated by applying the generalized eigenvalue decomposition (GEVD) procedure to the power spectral density (PSD) matrices of the received signals and the stationary noise. A basis for the interference subspace is estimated by collecting eigenvectors, calculated in segments where nonstationary interfering sources are active and the desired sources are inactive. The rank of the basis is then reduced by the application of the orthogonal triangular decomposition (QRD). This procedure relaxes the common requirement for nonoverlapping activity periods of the interference sources. A comprehensive experimental study in both simulated and real environments demonstrates the performance of the proposed beamformer.

Equalization of Multichannel Acoustic Systems in Oversampled Subbands

Equalization of room transfer functions (RTFs) is an important topic with several applications in acoustic signal processing. RTFs are often modeled as finite-impulse response filters, characterized by orders of thousands of taps and non-minimum phase. In practice, only approximate estimates of the actual RTFs are available due to measurement noise, limited estimation accuracy, and temporal variation of source-receiver position. These issues make equalization a difficult problem. In this paper, we discuss multichannel equalization with focus on inexact RTF estimates. We present a multichannel method for the equalization filter design utilizing decimated and oversampled subbands, where the full-band acoustic impulse response is decomposed into equivalent subband filters prior to equalization. This technique is not only more computationally efficient but also more robust to impulse response inaccuracies compared with the full-band counterpart.

Addressing potential sources of variation in several non-destructive techniques for measuring firmness in apples

Measurements of firmness have traditionally been carried out according to the Magness Taylor (MT) procedure; using a texture analyser or penetrometer in reference texture tests. Non-destructive tests like the acoustic impulse response of acoustic firmness sensors (AFSs), a low-mass impact firmness sensor Sinclair International (SIQ-FT) and impact test (Lateral Impact – UPM) have also been used to measure texture and firmness. The objectives of this study were to evaluate the influence of different sources of variation in these three non-destructive tests and to evaluate their respective capabilities of discriminating between fruit maturity at two different harvest dates, turgidity before and after dehydration treatment and ripening after different storage periods. According to our results, fruit studied an unexpected AFS trend with turgidity. Contact measurements (Lateral Impact – UPM and SIQ-FT) appeared highly sensitive to changes in turgidity, but were less able to follow changes in ripening caused by storage period. Contact measurements were suitable for detecting differences between fruits from different harvest dates and showed higher correlation coefficients with reference texture tests than acoustic measurements. The Lateral Impact – UPM test proved better at separating fruits according to turgidity than the SIQ-FT instrument.

Couchet Harpsichord soundboard vibroacoustics behaviour: An application of the Impact Nearfield Acoustical Holography (IPNAH)

The Music Museum in Paris recently acquired a harpsichord made by Ioannes Couchet in 1652 in Anvers. As a masterpiece this instrument is considered as a "National Treasure" and therefore protected. A challenging problematic has risen when its restoration was decided since the aim was to play this instrument again in concert. In the objectives of increasing our understanding of the harpsichord ageing, improve a numerical model currently in process and develop a diagnostic method for conservation, an experimental modal analysis of the soundboard was performed by processing its sound field. A non intrusive method, the Impact Planar Nearfield Acoustic Holography, was used. This technique, developed by the authors, implements the well known inverse method NAH on the basis of the acoustic impulse response field and is well adapted to modal analysis. NAH is performed here in unusual conditions compared to literature, as they are far from the ideal: unbaffled source, low sound pressure level, unusually large measurement distance, preponderance of evanescent waves. An additional challenge was to muffle strings, as they should not be removed nor slackened. However, a very satisfying modal decomposition for [30-1500Hz] bandwidth is obtained. Results are confronted with literature and an energetic analysis is proposed.

Measuring and Modeling Violin Sound Radiation for Sound Equalization

During a performance a violin is put into vibration and this vibration is radiated to the air. Depending on the distance and direction of the listener, he perceives a slightly different sound. Sound radiation at each point in the space is determined by the impulse response or transfer function between vibration of the violin and sound pressure at that point. There is variety of methodologies to obtain acoustic impulse responses of violins. They differ mainly in the way of exciting the violin, the point of excitation, the position where the acoustic response is measured and how excitation and response signals are deconvolved. In this work we measure violin vibration with a bridge pickup and we 1) propose a method to calculate the transfer function between signals from the bridge pickup and a microphone, 2) use a 3D motion tracker to get the position of the microphone respect to the violin so that we can calculate the response for different directions and distances and 3) we model sound radiation as a "reference" impulse response and a parametric filter whose response depends on distance and orientation with respect to the "reference" position.

Multivariate data analysis for classification of pineapple maturity

A range of physical and chemical parameters were investigated for their efficacy in classifying pineapple according to maturity or the presence of marbling defects. The tests were conducted on a sample of 115 “Pattavia” cultivar pineapples which were later sorted into three sets according to their differing maturity levels based on translucent yellow area of the longitudinal cut-open surface, and a further set comprising marbled fruit. The destructive parameters that were measured included soluble solids content, titratable acidity and flesh firmness. Nondestructive measurement techniques including acoustic impulse response tests and specific gravity determination by means of water replacement were carried out on each sample. Discriminant analysis using a canonical discriminant function with leave-one-out cross validation was performed to segregate the fruits into each class. The classification into maturity and marbling classes was achieved using multiple parameters with overall performance of 75.7%. The specific gravity, stiffness coefficient ( f 3 2 m 2/3 where f 3 is the third resonant frequency at the peaks of the acoustic spectrum, and m is fruit weight) and the soluble solids content appeared to be the most important parameters for distinguishing pineapples into three main groups: (1) class A (more mature fruit), (2) class B (mature fruit), and (3) class C (less mature fruit) and class M (marbled fruit). In addition, class M could be separated from class C by the titratable acidity, the flesh firmness and the stiffness coefficients based on f 1. For classification of mere maturity class, the performance was improved to 77.0%. The maturity was showed to be characterized by the specific gravity, the acidity and the soluble solids content. In differentiation of maturity using only nondestructive parameters, the specific gravity again gave the highest correlation for discriminant score of the discriminant function 1. The total classification performance was 82.8%. The stiffness coefficients complemented the classification through the discriminant function 2.

Application of the ISO 13472-1 in situ technique for measuring the acoustic absorption coefficient of grass and artificial turf surfaces

This paper presents the applicability of an in situ technique based on ISO 13472-1 standard for measuring the acoustic absorption coefficient of grass and artificial turf surfaces for normal incidence from a sound source. The in situ method is based on acoustic impulse response measurement of the material surface. A maximum length sequence (MLS) signal is played through a loudspeaker and the acoustic response from the surface is recorded using a single microphone. The fast Hadamard transform and fast Fourier transform based digital signal post-processing algorithm provides the acoustic absorption coefficient of the surface under test. The normal incidence acoustic absorption coefficient of a commercial artificial quash surface of Dow Co. obtained from this method was compared with the results from the ASTM E1050 impedance tube method for the same surface. The acoustic absorption coefficients of a test-site grass surfaces were measured for 30mm and 100mm length of grass blades in wet and dry soil conditions. Substantial difference in the acoustic absorption coefficient was observed for a similar grass-like artificial surface used for estimating sound power of commercial garden equipments and lawnmowers. The advantage of the in situ method lies in its ability to measure the normal incident acoustic absorption coefficient of any planar surface as installed or in situ. Additionally a quick testing time of less than a minute with the use of a laptop sound card based inexpensive data acquisition system is the main feature of this robust method.

Detection of Internal Cracks in Manchego Cheese Using the Acoustic Impulse-Response Technique and Ultrasounds

Nowadays, due to the more global nature of markets, the commercialization of cheese relies on the high quality of the product. Internal defects such as cracks or flaws may affect quality. Two different nondestructive inspection techniques (ultrasonic and acoustic experiments) were used to detect cracks in Manchego cheese. The existence of small eyes in this type of cheese limited the use of ultrasonic pulse-echo experiments due to high scattering, and only cracks close to the surface of the cheese could be detected. The acoustic impulse-response technique, however, allowed us to study wheel pieces with cracks located elsewhere in the cheese. Two different impact probes (A and B) were assayed. The energy content of the acoustic spectrum was higher for cracked wheel pieces (7,116 and 17,520V Hz1/2 for probes A and B, respectively) than for normal ones (6,841 and 16,821V Hz1/2). The differences were mainly found for frequencies higher than 150Hz, which made the centroid for cracked pieces higher (162 and 170Hz for probes A and B, respectively) than that for normal cheeses (132 and 148Hz for probes A and B, respectively). Discriminant functions were developed to classify wheel pieces, and the input variables used were the acoustic parameters from the spectrum and the principal components extracted from the whole spectrum. The best classification procedure used the principal components from the principal components analysis of the spectrum for probe B. In this case, the 50 wheel pieces used in this study were correctly classified. These results showed that a simple and low-cost acoustic impulse-response technique could be used to detect cheese cracks, formed at different moments of Manchego cheese maturation.

수박 내부결함판정을 위한 휴대형 압전형 장갑 센서시스템

Dynamic excitation and response analysis is an acceptable method to determine some of physical properties of agricultural product for quality evaluation. There is a difference in the internal viscoelasticity between sound and defective fruits due to the difference of geometric structures, thereby showing different vibration characteristics. This study was carried out to develop a portable piezoelectric film-based glove sensor system that can separate internally damaged watermelons from sound ones using an acoustic impulse response technique. Two piezoelectric sensors based on polyvinylidene fluoride (PVDF) films to measure an impact force and vibration response were separately mounted on each glove. Various signal parameters including number of peaks, energy ratio, standard deviation of peak to peak distance, zero-crossing rate, and integral value of peaks were examined to develop a regression-estimated model. When using SMLR (Stepwise Multiple Linear Regression) analysis in SAS, three parameters, i.e., zeros value, number of peaks, and standard deviation of peaks were selected as usable factors with a coefficient of determination () of 0.92 and a standard error of calibration (SEC) of 0.15. In the validation tests using twenty watermelon samples (sound 9, defective 11), the developed model provided good capability showing a classification accuracy of 95%.

Measurement and Identification of Acoustic Impulse Response

A description is given of an implementation of a system of measurement for acoustic impulse responses. The acoustic impulse response of a room is measurable with a dynamic range of 48 dB by cross correlation of the test signals. This proved to be better than by the identification by the algorithm of the stochastic gradient with decreasing step especially in noisy environments.