Gender recognition from vocal source

Efficiency of automatic recognition of male and female voices based on solving the inverse problem for glottis area dynamics and for waveform of the glottal airflow volume velocity pulse is studied. The inverse problem is regularized through the use of analytical models of the voice excitation pulse and of the dynamics of the glottis area, as well as the model of one-dimensional glottal airflow. Parameters of these models and spectral parameters of the volume velocity pulse are considered. The following parameters are found to be most promising: the instant of maximum glottis area, the maximum derivative of the area, the slope of the spectrum of the glottal airflow volume velocity pulse, the amplitude ratios of harmonics of this spectrum, and the pitch. On the plane of the first two main components in the space of these parameters, an almost twofold decrease in the classification error relative to that for the pitch alone is attained. The male voice recognition probability is found to be 94.7%, and the female voice recognition probability is 95.9%.     

Laryngeal vibrations: a comparison between high-speed filming and glottographic techniques

This study was designed to compare information on laryngeal vibrations obtained by high-speed filming, photoglottography (PGG), and electroglottography (ECG). Simultaneous glottographic signals and high-speed films were obtained from two subjects producing steady phonation. Measurements of glottal width were made at three points along the glottis in the anterior--posterior dimension and aligned with the other records. Results indicate that PGG and film measurements give essentially the same information for peak glottal opening and glottal closure. The EGG signal appears to reliably indicate vocal-fold contact. Together, PGG and EGG may provide much of the information obtained from high-speed filming as well as potentially detect horizontal phase differences during opening and closing.     

On the use of the derivative of electroglottographic signals for characterization of nonpathological phonation

Electroglottography is a common method for providing noninvasive measurements of glottal activity. The derivative of the electroglottographic signal, however, has not attracted much attention, although it yields reliable indicators of glottal closing instants. The purpose of this paper is to provide a guide to the usefulness of this signal. The main features that are to be found in this signal are presented on the basis of an extensive analysis of a database of items sung by 18 trained singers. Glottal opening and closing instants are related to peaks in the signal; the latter can be used to measure glottal parameters such as fundamental frequency and open quotient. In some cases, peaks are doubled or imprecise, which points to special (but by no means uncommon) glottal configurations. A correlation-based algorithm for the automatic measurement of fundamental frequency and open quotient using the derivative of electroglottographic signals is proposed. It is compared to three other electroglottographic-based methods with regard to the measurement of open quotient in inverse-filtered derived glottal flow. It is shown that agreement with the glottal-flow measurements is much better than most threshold-based measurements in the case of sustained sounds.     

Unsteady behavior of flow in a scaled-up vocal folds model

Measurements of the fluid flow through a scaled-up model of the human glottis are presented to determine whether glottal flow may be approximated as unsteady. Time- and space-resolved velocity vector fields from digital particle image velocimetry (DPIV) measurements of the flow through the gap between two moving, rigid walls are presented in four cases, over a range of Strouhal numbers: 0.010, 0.018, 0.035, 0.040, corresponding to life-scale f(0) of 30, 58, 109, and 126 Hz, respectively, at a Reynolds number of 8000. It is observed that (1) glottal flow onset is delayed after glottal opening and (2) glottal flow shutoff occurs prior to closure. A comparison between flow through a fully open, nonmoving glottis and that through the moving vocal folds shows a marked difference in spatial structure of the glottal jet. The following features of the flow are seen to exhibit strong dependence on cycle frequency: (a) glottal exit plane velocity, (b) volume flow, (c) vortex shedding rates, and (d) vortex amplitude. Vortex shedding appears to be a factor both in controlling flow resistance and in cycle-to-cycle volume flow variations. All these observations strongly suggest that glottal flow is inherently unsteady.     

Influence of glottal closure configuration on vocal efficacy in young normal-speaking women

Posterior closure insufficiency of the glottis is often mentioned in connection with permanent voice disorders. Recently published studies have revealed that an incomplete closure of the glottis can be found also in normal-speaking voices, especially in women. However, the effect of glottal closure configuration on vocal efficacy is not sufficiently clarified. The purpose of this study was to determine the effect of glottal closure configuration on singing and speaking voice characteristics. Overall, 520 young female normal-speaking subjects were examined by videostroboscopy for different phonation conditions in the combination of soft, loud, low, and/or high phonation and by voice range profile measurements. According to the videostroboscopic analysis, the subjects were subdivided into four groups: complete closure of the vocal folds already in soft phonation (group 1), closure of the vocal fold with increasing intensity (group 2), persistent closure insufficiencies despite increasing intensity (group 3), and hourglass-shaped closure in subjects with vocal nodules (group 4). Subjects in which the glottal closure could not be evaluated sufficiently were subclassified into group 5 (missing values).

Selected criteria of the singing and speaking voice were evaluated and statistically processed according to the mentioned subclassification. Group 1 reached significantly the highest sound pressure levels (SPL_max) for the singing voice as well as for the shouting voice. Group 3 showed a limited capacity to increase the intensity of the singing and speaking voice. The results gathered in this study objectify the relationship of insufficient glottal closure and reduced vocal capabilities. As long as no conclusive data on long-term consequences of insufficient glottal closure are available, a prophylactic improvement of the laryngeal situation especially in female professional voice users by voice therapy should be recommended.     

腭裂语音中齿龈塞音的声门代偿现象声学分析与判定

对齿龈塞音在腭裂语音中的声门塞音代偿现象进行了声学分析,计算频谱分布的多阶统计量—谱矩,并将代偿塞音和正常塞音进行对比。结果显示声门塞音爆破段的第一阶谱矩即频谱质心的频率位置比正常塞音低,因为声门塞音的阻塞部位在声门,导致声道腔体偏长从而共振频率偏低。还观察到声门塞音的第二阶谱矩即标准偏差偏高,说明其谱能量分布比正常塞音更加分散。声门塞音的第三阶谱矩即偏度大多为正值,反映了声门塞音功率谱的非对称性且大头朝向低频区而长尾朝向高频区。采用逻辑回归模型进行样本分类,通过交叉验证选出最优的四阶谱矩作为模型自变量,分类正确率为89.7%。结合塞音爆破时刻自动检测,实现了音节/di/的声门塞音客观判定。      

The effect of glottal angle on intraglottal pressure

Intraglottal pressure distributions depend upon glottal shape, size, and diameter. This study reports the effects of varying glottal angle on intraglottal and transglottal pressures using a three-dimensional Plexiglas model with a glottis having nine symmetric glottal angles and a constant minimal glottal diameter of 0.06 cm. The empirical data were supported by computational results using FLUENT. The results suggested that (1) the greater the convergent glottal angle, the greater outward driving forces (higher intraglottal pressures) on the vocal folds; (2) flow resistance was greatest for the uniform glottis, and least for the 10 degrees divergent glottis; (3) the greatest negative pressure in the glottis and therefore the greatest pressure recovery for diverging glottal shapes occurred for an angle of 10 degrees; (4) the smaller the convergent angle, the greater the flow resistance; (5) FLUENT was highly accurate in predicting the empirical pressures of this model; (6) flow separation locations (given by FLUENT) for the divergent glottis moved upstream for larger flows and larger glottal angles. The results suggest that phonatory efficiency related to aerodynamics may be enhanced with vocal fold oscillations that include large convergent angles during glottal opening and small (5 degrees - 10 degrees) divergent angles during glottal closing.     

Analysis of voice source characteristics using a constrained polynomial representation of voice source signals

To analyze the characteristics of voice source signals from speech, a model is presented in the form of polynomial function by expanding the definition of the Rosenberg model. In combination with the all-pole assumption of the vocal-tract filter, methods are described for the pitch-synchronous speech analysis and temporal search of the glottal opening and closing instants. Because the source and filter models are both linear, the parameter estimation problem can be conveniently solved. In addition, the temporal search method can refine the locations of the glottal events and improve the accuracy of the parameter estimation. Analyses of non-nasalized voiced speech are conducted using an electroglottographic device from which the initial estimate of the temporal information is given.     

Glottal closure, transglottal airflow, and voice quality in healthy middle-aged women

Seventeen healthy women, 45 to 61 years old, were examined using videofiberstroboscopy during phonation at three loudness levels. Two phoniatricians evaluated glottal closure using category and ratio scales. Transglottal airflow was studied by inverse filtering of the oral airflow signal recorded in a flow mask (Glottal Enterprises System) during the spoken phrase /ba:pa:pa:pa:p/ at three loudness levels. Subglottal pressure was estimated from the intraoral pressure during p occlusion. Running speech and the repeated /pa:/ syllables were perceptually evaluated by three speech pathologists regarding breathiness, hypo-, and hyperfunction, using continuous scales. Incomplete glottal closure was found in 35 of 46 phonations (76%). The degree of glottal closure increased significantly with raised loudness. Half of the women closed the glottis completely during loud phonation. Posterior glottal chink (PGC) was the most common gap configuration and was found in 28 of 46 phonations (61%). One third of the PGCs were in the cartilaginous glottis (PGCc) only. Two thirds extended into the membranous portion (PGCm); most of these occurred during soft phonation. Peak flow, peak-to-peak (AC) flow, and the maximum rate of change for the flow in the closing phase increased significantly with raised loudness. Minimum flow decreased significantly from normal to loud voice. Breathiness decreased with increased loudness. The results suggest that the incomplete closure patterns PGCc and PGCm during soft phonation ought primarily to be regarded as normal for Swedish women in this age group.     

On the role of glottis-interior sources in the production of voiced sound

The voice source is dominated by aeroacoustic sources downstream of the glottis. In this paper an investigation is made of the contribution to voiced speech of secondary sources within the glottis. The acoustic waveform is ultimately determined by the volume velocity of air at the glottis, which is controlled by vocal fold vibration, pressure forcing from the lungs, and unsteady backreactions from the sound and from the supraglottal air jet. The theory of aerodynamic sound is applied to study the influence on the fine details of the acoustic waveform of "potential flow" added-mass-type glottal sources, glottis friction, and vorticity either in the glottis-wall boundary layer or in the portion of the free jet shear layer within the glottis. These sources govern predominantly the high frequency content of the sound when the glottis is near closure. A detailed analysis performed for a canonical, cylindrical glottis of rectangular cross section indicates that glottis-interior boundary/shear layer vortex sources and the surface frictional source are of comparable importance; the influence of the potential flow source is about an order of magnitude smaller.     

Glottal configuration associated withfundamental frequency and vocal register

Simultaneous measurements of mean airflow rate, vocal intensityand fundamental frequency were made during flexible video endoscopic recording of the vowel /i/ sustained in two vocal registers, modal and falsetto. The glottal closure patterns of four males and four females were evaluated by visually inspecting the video images. Acoustic signals were recorded and analyzed to verify the frequency and intensity criteria. Aerodynamic analysis of mean airflow rate was done via Rothenberg mask and commercial software. Incomplete glottic closure was common in both males and females. The degree of closure was significantly higher for modal samples than for falsetto samples with frequency and intensity held constant. The shape of the glottal closure did not vary with changes in the mode of phonation. As expected, the mean airflow rate increased with decreased glottal closure. The results suggest that incomplete glottic closure should be considered as a normal glottal configuration in high frequency modal and falsetto phonation. Moreover, hourglass and spindle glottal configurations may also be found in both the modal and falsetto registers of normal subjects. These results also confirm the positive relationships between degree of glottal gap and mean airflow rate. Thus, mean airflow rate may be regarded as a criterion for judging degree of glottal closure.     

Investigation of a glottal related harmonics-to-noise ratio and spectral tilt as indicators of glottal noise in synthesized and human voice signals

The harmonics-to-noise ratio (HNR) of the voiced speech signal has implicitly been used to infer information regarding the turbulent noise level at the glottis. However, two problems exist for inferring glottal noise attributes from the HNR of the speech wave form: (i) the measure is fundamental frequency (f0) dependent for equal levels of glottal noise, and (ii) any deviation from signal periodicity affects the ratio, not just turbulent noise. An alternative harmonics-to-noise ratio formulation [glottal related HNR (GHNR')] is proposed to overcome the former problem. In GHNR' a mean over the spectral range of interest of the HNRs at specific harmonic/between-harmonic frequencies (expressed in linear scale) is calculated. For the latter issue [(ii)] two spectral tilt measures are shown, using synthesis data, to be sensitive to glottal noise while at the same time being comparatively insensitive to other glottal aperiodicities. The theoretical development predicts that the spectral tilt measures reduce as noise levels increase. A conventional HNR estimator, GHNR' and two spectral tilt measures are applied to a data set of 13 pathological and 12 normal voice samples. One of the tilt measures and GHNR' are shown to provide statistically significant differentiating power over a conventional HNR estimator.     

Physiologic and acoustic differences between male and female voices

Comparison is drawn between male and female larynges on the basis of overall size, vocal fold membranous length, elastic properties of tissue, and prephonatory glottal shape. Two scale factors are proposed that are useful for explaining differences in fundamental frequency, sound power, mean airflow, and glottal efficiency. Fundamental frequency is scaled primarily according to the membranous length of the vocal folds (scale factor of 1.6), whereas mean airflow, sound power, glottal efficiency, and amplitude of vibration include another scale factor (1.2) that relates to overall larynx size. Some explanations are given for observed sex differences in glottographic waveforms. In particular, the simulated (computer-modeled) vocal fold contact area is used to infer male-female differences in the shape of the glottis. The female glottis appears to converge more linearly (from bottom to top) than the male glottis, primarily because of medial surface bulging of the male vocal folds.     

Flow visualization and pressure distributions in a model of the glottis with a symmetric and oblique divergent angle of 10 degrees

Modeling the human larynx can provide insights into the nature of the flow and pressures within the glottis. In this study, the intraglottal pressures and glottal jet flow were studied for a divergent glottis that was symmetric for one case and oblique for another. A Plexiglas model of the larynx (7.5 times life size) with interchangeable vocal folds was used. Each vocal fold had at least 11 pressure taps. The minimal glottal diameter was held constant at 0.04 cm. The glottis had an included divergent angle of 10 degrees. In one case the glottis was symmetric. In the other case, the glottis had an obliquity of 15 degrees. For each geometry, transglottal pressure drops of 3, 5, 10, and 15 cm H2O were used. Pressure distribution results, suggesting significantly different cross-channel pressures at glottal entry for the oblique case, replicate the data in another study by Scherer et al. [J. Acoust. Soc. Am. 109, 1616-1630 (2001b)]. Flow visualization using a LASER sheet and seeded airflow indicated separated flow inside the glottis. Separation points did not appear to change with flow for the symmetric glottis, but for the oblique glottis moved upstream on the divergent glottal wall as flow rate increased. The outgoing glottal jet was skewed off-axis for both the symmetric and oblique cases. The laser sheet showed asymmetric circulating regions in the downstream region. The length of the laminar core of the glottal jet was less than approximately 0.6 cm, and decreased in length as flow increased. The results suggest that the glottal obliquity studied here creates significantly different driving forces on the two sides of the glottis (especially at the entrance to the glottis), and that the skewed glottal jet characteristics need to be taken into consideration for modeling and aeroacoustic purposes.     

Qualitative and quantitative analysis of voice onset by means of a multidimensional voice analysis system (MVAS) using high-speed imaging

High-speed filming is one of the most informative methods for assessing voice physiology data. Tracing high-speed images of the glottis provides quantitative parameters such as the glottal area and the glottal width function. By way of example, a number of studies are discussed which extract quantitative data from high-speed images showing voice onsets. Furthermore, a new computer system (MVAS; multi-dimensional voice analysis system) is presented that synchronously displays a laryngoscopic high-speed film, the electroglottographical signal, and several acoustic analyses of the recorded voice sample. The automatic measurement of glottal width and glottal area from the laryngoscopic images is also provided. Looking at former studies and our analyses of voice onsets reveals a tremendous intersubject and even intrasubject variability (different prephonatory closure, different time span until full amplitude is reached, different open quotient).     

基于最小二乘支持向量机的国公酒中橙皮苷含量测定

应用近红外光谱技术结合最小二乘支持向量机建立了国公酒中橙皮苷含量的模型。利用Kernard-Stone法对训练集样本进行划分,对光谱数据预处理方法进行了选择,比较了平滑、范围标度化、自标度化、一阶微分、二阶微分以及这几种预处理相互结合的六种方法,确定了以平滑、一阶微分,范围标度化作为国公酒近红外光谱的数据预处理方法,采用组合的间隔偏最小二乘法筛选出有效波段8 211～8 312 cm-1及9 712～9 808 cm-1。应用最小二乘支持向量机建立模型,所建模型的交叉验证误差均方根为0.000 1,预测误差均方根为0.004,预测集的相对偏差小于5%。与组合的间隔偏最小二乘法、径向基-人工神经网络和支持向量机进行了比较。该方法快速、无损且可靠,可作为国公酒中橙皮苷含量快速测定的手段。     

低浓度水质化学需氧量紫外吸光检测中浊度与pH影响分析

采用紫外吸光法采集浓度范围1～25 mg·L-1的邻苯二甲酸氢钾配置的标准液图谱,对光谱进行不同的预处理后运用偏最小二乘法建立紫外吸光光谱下的COD定量分析模型,并对所建模型进行了浊度、pH值的影响分析。结果显示采用平滑一阶导数预处理方法,内部交叉验证均方根RMSECV值为0.122 27,主成份数为4,预测模型的平方相关系数r为0.999 8,预测相对误差在0.03％～1.7％的范围内;对0～100 NTU不同浊度的溶液,应用所建模型补偿后相对标准偏差RSD达到2.3%;pH在3～10变化时,影响可以忽略。     

The effect of entrance radii on intraglottal pressure distributions in the divergent glottis

Modeling laryngeal aerodynamics requires specification of the glottal geometry. Changing the glottal exit radius alters the intraglottal pressure distributions in the converging glottis [Scherer et al., J. Acoust. Soc. Am. 110, 2267-2269 (2001)]. This study examined the effects of the glottal entrance radius on the intraglottal pressure distributions for divergent angles of 5°, 10°, 20°, 30°, and 40°. Glottal airflow and minimal glottal diameter were held constant at 73.2 cm(3)/s and 0.02 cm, respectively. The computational code FLUENT was used to obtain the pressure distributions. Results suggest that a smaller glottal entrance radius tends to (1) lower the transglottal pressure (reduce glottal flow resistance), although this is angle dependent, (2) make the pressure dip near the glottal entrance more negative in value, (3) increase the slope of the pressure distribution just upstream of the glottal entrance, and (4) make the initial pressure recovery (rise) in the glottis steeper. A general empirical equation for transglottal pressure as a function of radius, angle, and separation point location is offered. These results suggest that glottal entrance curvature for the divergent glottis significantly affects the driving pressures on the vocal folds, and needs to be well specified when building computational and physical models.     

近红外光谱法快速测定工作场所莠去津浓度的研究

莠去津是一种广泛使用的除草剂,我国是其原药的主要生产国家。为加强工作场所莠去津暴露浓度的检测力度,保障职业接触工人身体健康,研究开发工作场所莠去津浓度的现场快速检测方法具有重要现实意义。利用自行组装的便携式近红外光谱仪,采集了实验室配置的浓度为10～1 000 mg·L-1的莠去津溶液样本光谱,并比较了多元散射校正、变量标准化、一阶导数方法、二阶导数方法及其组合等光谱预处理方法,竞争自适应重加权采样变量选择法和遗传算法等变量选择方法,偏最小二乘算法和支持向量机等回归方法对近红外光谱模型分析精度的影响。研究发现一阶导数是最佳光谱预处理方法;遗传算法优选的光谱变量表现优于竞争自适应重加权采样变量选择法;支持向量机模型表现优于偏最小二乘模型。基于遗传算法选择的16个光谱变量建立的支持向量机模型分析精度最高,其定标决定系数、验证决定系数、定标均方差、预测均方差和相对分析误差(成分浓度的标准偏差与预测均方差的比值)分别为1,0.99,17.54 mg·L-1,25.42 mg·L-1和11.43,有望应用于工作场所莠去津浓度的实际检测中。该研究探讨了近红外光谱法检测工作场所莠去津浓度的可行性,相关结果对于未来类似工作的开展具有重要参考价值。