Objective analysis of the singing voice as a training aid.

A new tool for robust tracking of fundamental frequency is proposed, along with an objective measure of main singing voice parameters, such as vibrato rate, vibrato extent, and vocal intonation. High-resolution Power Spectral Density estimation is implemented, based on AutoRegressive models of suitable order, allowing reliable formant tracking also in vocalizations characterized by highly varying values. The proposed techniques are applied to about 1000 vocalizations, coming from both professional and non-professional singers, and show better performance as compared to classical Fourier-based approaches. If properly implemented, and with a user-friendly interface, the new tool would allow real-time analysis of singing voice. Hence, it could be of help in giving non-professional singers and singing teachers reliable measures of possible improvements during and after training.

Developmental aspects of infant's cry melody and formants.

This paper deals with the analysis of cry melodies (time variations of the fundamental frequency) as well as vocal tract resonance frequencies (formants) from infant cry signals. The increase of complexity of cry melodies is a good indicator for neuro-muscular maturation as well as for the evaluation of pre-speech development. The variation of formant frequencies allows an estimation of articulatory activity during pre-speech vocalization. Subjects are three pairs of healthy identical twins (monocygozity determined by DNA-fingerprint). Spontaneous cries of these six children were recorded at different ages: 8th-9th week, 15th-17th week and 23rd-24th week. Analysis of 136 cry melodies and intensity contours was made using KAY-CSL 4300/MDVP. For formant estimation a spectral parametric technique was applied, which was based on autoregressive models (Digital spectral analysis with applications, 1987) whose order is adaptively estimated on subsequent signal frames by means of a new method (Med. Eng. Phys. 20 (1998) 432; Utras. Med. Biol. 21 (1995) 793). Cry melodies exhibited an increasing complexity during the observation period. Beginning with the second observation period (15th-17th week) an increasing coupling and tuning between melody and resonance frequencies was observed, which was interpreted as "intentional" articulatory activity. Possible applications are in cry diagnosis as well as in the evaluation of pre-speech development.

A simple subspace approach for speech denoising.

For pathological voices, hoarseness is mainly due to airflow turbulence in the vocal tract and is often referred to as noise. This paper focuses on the enhancement of speech signals that are supposedly degraded by additive white noise. Speech enhancement is performed in the time-domain, by means of a fast and reliable subspace approach. A low-order singular value decomposition (SVD) allows separating the signal and the noise contribution in subsequent data frames of the analysed speech signal. The noise component is thus removed from the signal and the filtered signal is reconstructed along the directions spanned by the eigenvectors associated with the signal subspace eigenvalues only, thus giving enhanced voice quality. This approach was tested on synthetic data, showing higher performance in terms of increased SNR when compared with linear prediction (LP) filtering. It was also successfully applied to real data, from hoarse voices of patients that had undergone partial cordectomisation. The simple structure of the proposed technique allows a real-time implementation, suitable for portable device realisation, as an aid to dysphonic speakers. It could be useful for reducing the effort in speaking, which is closely related to social problems due to awkwardness of voice.

A comparative analysis of fundamental frequency estimation methods with application to pathological voices.

One of the basic parameters characterising voiced phonation is the fundamental frequency, named pitch, which is the rate of vibration of the folds. In pathological voices. pitch variations within an utterance are indicative of the patient status. As such voices are corrupted by 'noise', robust pitch estimation methods are required in order to track its variations. This paper aims to compare some pitch estimation methods, pointing out their main advantages and drawbacks for present application. For each method, modifications are proposed in order to enhance performance. The methods are tested on simulated signals and then applied to real signals, coming both from healthy and pathological voices. The latter we obtained from patients who have undergone surgery for vocal folds via laser or traditional lancet techniques.

Parametric and non-parametric estimation of speech formants: application to infant cry.

The present paper addresses the issue of correctly estimating the peaks in the speech envelope (formants) occurring in newborn infant cry. Clinical studies have shown that the analysis of such spectral characteristics is a helpful noninvasive diagnostic tool. In fact it can be applied to explore brain function at very early stage of child development, for a timely diagnosis of neonatal disease and malformation. The paper focuses on the performance comparison between some classical parametric and non-parametric estimation techniques particularly well suited for the present application, specifically the LP, ARX and cepstrum approaches. It is shown that, if the model order is correctly chosen, parametric methods are in general more reliable and robust against noise, but exhibit a less uniform behaviour than cepstrum. The methods are compared also in terms of tracking capability, since the signals under study are nonstationary. Both simulated and real signals are used in order to outline the relevant features of the proposed approaches.

Laboratory simulations of lidar returns from clouds.

The results of lidar measurements on laboratory-scaled cloud models are presented. The lidar system was based on a picosecond laser source and a streak camera. The cloud was simulated by a homogeneous aqueous suspension of calibrated microspheres. Measurements were repeated for different concentrations of diffusers and for different values of the receiver angular field of view. The geometric situation was similar to one of an actual lidar sounding a 300-m-thick cloud at a distance of 1200 or 7800 m. The results show how the effect of multiple scattering depends on the extinction coefficient of the sounded medium and on the geometric parameters. The depolarization introduced by multiple scattering was also investigated. Measurements were carried out in well-controlled conditions. The results can thus be useful to validate the accuracy of numerical or analytical procedures that have been developed to study multiple-scattering contribution in lidar returns.

Multiple scattering from Chebyshev particles: Monte Carlo simulations for backscattering in lidar geometry.

Lidar measurements are often interpreted on the basis of two fundamental assumptions: absence of multiple scattering and sphericity of the particles that make up the diffusing medium. There are situations in which neither holds true. We focus our interest on multiply-scattered returns from homogeneous layers of monodisperse, randomly oriented, axisymmetric nonspherical particles. T(2) Chebyshev particles have been chosen and their single-scattering properties have been reviewed. A Monte Carlo procedure has been employed to calculate the backscattered signal for several fields of view. Comparisons with the case of scattering from equivalent (equal-volume) spheres have been carried out (narrow polydispersions have been used to smooth the phase functions' oscillations). Our numerical effort highlights a considerable variability in the intensity of the multiply-scattered signal, which is a consequence of the strong dependence of the backscattering cross section on deformation of the particles. Even more striking effects have been noted for depolarization; peculiar behavior was observed at moderate optical depths when particles characterized by a large backscattering depolarization ratio were employed in our simulations. The sensitivity of depolarization to even small departures from sphericity, in spite of random orientation of the particles, has been confirmed. The results obtained with the Monte Carlo codes have been successfully checked with an analytical formula for double scattering.

Computerized cry analysis in infants affected by severe protein energy malnutrition.

A new method of computerized cry analysis has been utilized to evaluate the cries of infants affected by severe protein energy malnutrition. We studied 17 Kenian babies affected by severe malnutrition for more than four months (9 cases of marasmus and 8 of kwashiorkor) and a control group of 17 well-nourished babies. The cries of the malnourished children showed lower inter-utterance variability, formants' frequencies and cry score, assigned by the Infant Cry Modulation Assessment Scale. The melodic pattern was more often flat, rising or falling-rising, when compared to the cries of the well-nourished babies. We hypothesize that these differences reflect the state of brain damage associated with protein energy malnutrition. No differences were found between the cries of infants affected by marasmus and those affected by kwashiorkor, between the cries recorded before and after nutritional therapy and between the first cries of malnourished children who subsequently died during hospitalization and those of infants who survived.

Laboratory simulations of lidar returns from clouds: experimental and numerical results.

The experimental results of laboratory simulations of lidar returns from clouds are presented. Measurements were carried out on laboratory-scaled cloud models by using a picosecond laser and a streak-camera system. The turbid structures simulating clouds were suspensions of polystyrene spheres in water. The geometrical situation was similar to that of an actual lidar sounding a cloud 1000 m distant and with a thickness of 300 m. Measurements were repeated for different concentrations and different sizes of spheres. The results show how the effect of multiple scattering depends on the scattering coefficient and on the phase function of the diffusers. The depolarization introduced by multiple scattering was also investigated. The results were also compared with numerical results obtained by Monte Carlo simulations. Substantially good agreement between numerical and experimental results was found. The measurements showed the adequacy of modern electro-optical systems to study the features of multiple-scattering effects on lidar echoes from atmosphere or ocean by means of experiments on well-controlled laboratory-scaled models. This adequacy provides the possibility of studying the influence of different effects in the laboratory in well-controlled situations.

Monte Carlo calculations of the modulation transfer function of an optical system operating in a turbid medium.

Using a Monte Carlo method, we investigate the effect of a turbid medium on image transmission by means of the modulation transfer function approach. We present results that refer to a medium that consists of a random distribution of water spherical particles in air. We analyze the effect of geometric conditions (medium width and position) and source characteristics (Lambertian, beam emission). We present results for small spheres (Rayleigh scattering) and spheres (1.0-microm diameter) that are not small in comparison with the wavelength lambda = 0.6328 microm. Numerical data show a large modulation transfer function dependence on the source emission aperture and a substantial independence of the medium width for a fixed value of the optical depth. In accordance with reciprocity principles, we test an inverse scheme of Monte Carlo calculation, the advantage of this scheme being a substantial reduction in calculation time.

Transmission of a pulsed polarized light beam through thick turbid media: numerical results.

We present numerical results on the change in polarization state of light pulses transmitted through thick turbid media. These results were obtained with a modified version of a previous Monte Carlo code that takes into account depolarization introduced by multiple scattering. The results have shown that for scattered received power pulse shape, polarization and total received power mainly depend on the transport cross section, σ(d), of the medium. The effect of the angular field of view of the receiver or of the distance between the diffusing medium and the receiver is shown, whereas the effect of the lateral displacement of the receiver elements proves to be of minor importance. An example of measurements showed a good agreement with numerical results, indicating the adequacy of our numerical code.

Transmission of a pulsed thin light beam through thick turbid media: experimental results.

Experimental results of light pulse transmission through thick turbid media are presented. Measurements have been carried out on polystyrene latex spheres by using a picosecond thin laser beam and a streak camera system. The results show that the shape of the received pulse depends mostly on the transport mean free path and on the absorption coefficient of the medium, indicating that both the absorption coefficient and the asymmetry factor of the scattering function can be obtained from the pulse shape. The results also show that a detectable amount of received photons follows trajectories near the source receiver line even for large values of optical depth, indicating the potential of a time-gated scanning imaging system to detect absorbing structures inside thick turbid media.

Simple inexpensive method of measuring the temporal spreading of a light pulse propagating in a turbid medium.

A simple and inexpensive method of measuring statistical parameters related to the time lengthening arising in the propagation of a light pulse in a turbid medium is presented. The method is based on the repetition of attenuation measurements of a light beam passing through the turbid medium when the absorption coefficient of the medium surrounding the diffusing particles is varied. The measurements are carried out using a cw source and a simple optical receiver with a common photodiode as a detector. The results of two measurements are reported together with the results of numerical simulations carried out using the scattering properties and geometric parameters corresponding to the experimental situation. Numerical results were obtained using a Monte Carlo based method. Good agreement between experimental and numerical results was found.

Method of measuring the phase function of a turbid medium in the small scattering angle range.

A method is described for measuring the phase scattering function of a turbid medium in the near forward direction. The method is based on the use of a transmissometer with a variable field of view and is suitable for measurements on natural media, such as water or turbid atmospheres. Some precautions to be taken to avoid the effects of multiple scattering are indicated. Results of laboratory tests on suspensions of latex spheres in water are presented. An example of the use of the instrument in natural fog is also shown.

Separation and analysis of forward scattered power in laboratory measurements of light beam transmittance through a turbid medium.

In laboratory measurements of the transmittance of a light beam through a diffusing medium (water plus latex spheres), a distinction between the attenuated beam power and the received forward scattered power was made possible by the use of a transmissometer whose receiver has a variable field of view. The dependence of the received scattered power on the FOV angle and on the medium optical depth was analyzed. The deduced separated contributions of first- and second-order scattering, as well as the total received scattered power, were compared to the results of calculations.