Search

Magnetic resonance (MR) imaging is becoming an established tool in capturing articulatory and physiological motion of the structures and muscles throughout the vocal tract and enabling visual and quantitative assessment of real-time speech activities. Although motion capture speed has been regularly improved by the continual developments in high-speed MR technology, quantitative analysis of multi-subject group data remains challenging due to variations in speaking rate and imaging time among different subjects. In this paper, a workflow of post-processing methods that matches different MR image datasets within a study group is proposed. Each subject's recorded audio waveform during speech is used to extract temporal domain information and generate temporal alignment mappings from their matching pattern. The corresponding image data are resampled by deformable registration and interpolation of the deformation fields, achieving inter-subject temporal alignment between image sequences. A four-dimensional dynamic MR speech atlas is constructed using aligned volumes from four human subjects. Similarity tests between subject and target domains using the squared error, cross correlation, and mutual information measures all show an overall score increase after spatiotemporal alignment. The amount of image variability in atlas construction is reduced, indicating a quality increase in the multi-subject data for groupwise quantitative analysis.

Sound generation during voiced speech remains an open research topic because the underlying process within the human larynx is hardly accessible for direct measurements. In the present study, harmonic sound generation during phonation was investigated with a model that replicates the fully coupled fluid-structure-acoustic interaction (FSAI). The FSAI was captured using a multi-modal approach by measuring the flow and acoustic source fields based on particle image velocimetry, as well as the surface velocity of the vocal folds based on laser vibrometry and high-speed imaging. Strong harmonic sources were localized near the glottis, as well as further downstream, during the presence of the supraglottal jet. The strongest harmonic content of the vocal fold surface motion was verified for the area near the glottis, which directly interacts with the glottal jet flow. Also, the acoustic back-coupling of the formant frequencies onto the harmonic oscillation of the vocal folds was verified. These findings verify that harmonic sound generation is the result of a strong interrelation between the vocal fold motion, modulated flow field, and vocal tract geometry.

Classical singers use nasal consonants as “resonance exercises,” and experimental results have shown that singers may use some velopharyngeal opening (VPO), most often in [a:] and more seldom in [i:] or [u:]. In particular, male singers have been found to increase VPO as pitch rises toward register change frequencies (passaggi). Laryngoscopic findings have shown that some VPO stabilizes vocal fold vibration; the effect is related to positive reactance. This study investigates the effects of VPO on vocal tract (VT) reactance over the range of fundamental frequencies (f0) used in singing using a computerized tomography-based finite element model of the VT and nose of a female. According to the results, by raising the lowest VT resonances, the VPO increased the VT reactance in the frequency ranges 207–359 Hz for [i:], 265–411 Hz for [u:], and 500–611 Hz for [a:], depending on the VPO size (full or half VPO). These frequency ranges are close to the first and second passaggio of a female singer. The change may have an especially practical stabilizing effect for [a:], which is otherwise characterized by very large changes in VT reactance and negative reactance at the second passaggio.

This study aimed to develop an artificial intelligence (AI)-based tool for screening COVID-19 patients based on the acoustic parameters of their voices. Twenty-five acoustic parameters were extracted from voice samples of 203 COVID-19 patients and 171 healthy individuals who produced a sustained vowel, i.e., /a/, as long as they could after a deep breath. The selected acoustic parameters were from different categories including fundamental frequency and its perturbation, harmonicity, vocal tract function, airflow sufficiency, and periodicity. After the feature extraction, different machine learning methods were tested. A leave-one-subject-out validation scheme was used to tune the hyper-parameters and record the test set results. Then the models were compared based on their accuracy, precision, recall, and F1-score. Based on accuracy (89.71%), recall (91.63%), and F1-score (90.62%), the best model was the feedforward neural network (FFNN). Its precision function (89.63%) was a bit lower than the logistic regression (90.17%). Based on these results and confusion matrices, the FFNN model was employed in the software. This screening tool could be practically used at home and public places to ensure the health of each individual's respiratory system. If there are any related abnormalities in the test taker's voice, the tool recommends that they seek a medical consultant.

In recent studies, it has been assumed that vocal tract formants (Fn) and the voice source could interact. However, there are only few studies analyzing this assumption in vivo. Here, the vowel transition /i/-/a/-/u/-/i/ of 12 professional classical singers (6 females, 6 males) when phonating on the pitch D4 [fundamental frequency (ƒo) ca. 294 Hz] were analyzed using transnasal high speed videoendoscopy (20.000 fps), electroglottography (EGG), and audio recordings. Fn data were calculated using a cepstral method. Source-filter interaction candidates (SFICs) were determined by (a) algorithmic detection of major intersections of Fn/nƒo and (b) perceptual assessment of the EGG signal. Although the open quotient showed some increase for the /i-a/ and /u-i/ transitions, there were no clear effects at the expected Fn/nƒo intersections. In contrast, ƒo adjustments and changes in the phonovibrogram occurred at perceptually derived SFICs, suggesting level-two interactions. In some cases, these were constituted by intersections between higher nƒo and Fn. The presented data partially corroborates that vowel transitions may result in level-two interactions also in professional singers. However, the lack of systematically detectable effects suggests either the absence of a strong interaction or existence of confounding factors, which may potentially counterbalance the level-two-interactions.

The glossectomy procedure, involving surgical resection of cancerous lingual tissue, has long been observed to affect speech production. This study aims to quantitatively index and compare complexity of vocal tract shaping due to lingual movement in individuals who have undergone glossectomy and typical speakers using real-time magnetic resonance imaging data and Principal Component Analysis. The data reveal that (i) the type of glossectomy undergone largely predicts the patterns in vocal tract shaping observed, (ii) gross forward and backward motion of the tongue body accounts for more change in vocal tract shaping than do subtler movements of the tongue (e.g., tongue tip constrictions) in patient data, and (iii) fewer vocal tract shaping components are required to account for the patients' speech data than typical speech data, suggesting that the patient data at hand exhibit less complex vocal tract shaping in the midsagittal plane than do the data from the typical speakers observed.

Semi-occluded vocal tract exercises (SOVTEs) are increasingly popular as therapeutic exercises for patients with voice disorders. This popularity is reflected in the growing research literature, investigating the scientific principles underlying SOVTEs and their practical efficacy. This study examines several acoustic, articulatory, and aerodynamic variables before, during, and after short-duration (15 s) SOVTEs with a narrow tube in air. Participants were 20 healthy young adults, and all variables were measured at threshold phonation levels. Acoustic variables were measured with a microphone and a neck accelerometer, and include fundamental frequency, glottal open quotient, and vocal efficiency. Articulatory variables were measured with ultrasound, and include measures of the tongue tip, tongue dorsum, and posterior tongue height, and horizontal tongue length. Aerodynamic variables were measured with an intraoral pressure transducer and include subglottal, intraoral, and transglottal pressures. Lowering of the posterior tongue height and tongue dorsum height were observed with gender-specific small changes in the fundamental frequency, but there were no significant effects on the transglottal pressure or vocal efficiency. These findings suggest that the voices of healthy young adults already approach optimal performance, and the continued search for scientific evidence supporting SOVTEs should focus on populations with voice disorders.

application/pdf, Toyohashi University of Technology, National Institute for Japanese Language and Linguistics

In the clinic, many diagnostic and therapeutic procedures focus on the oscillation patterns of the vocal folds (VF). Dynamic characteristics of the VFs, such as symmetry, periodicity, and full glottal closure, are considered essential features for healthy phonation. However, the relevance of these individual factors in the complex interaction between the airflow, laryngeal structures, and the resulting acoustics has not yet been quantified. Sustained phonation was induced in nine excised porcine larynges without vocal tract (supraglottal structures had been removed above the ventricular folds). The multimodal setup was designed to simultaneously control and monitor key aspects of phonation in the three essential parts of the larynx. More specifically, measurements will comprise (1) the subglottal pressure signal, (2) high-speed recordings in the glottal plane, and (3) the acoustic signal in the supraglottal region. The automated setup regulates glottal airflow, asymmetric arytenoid adduction, and the pre-phonatory glottal gap. Statistical analysis revealed a beneficial influence of VF periodicity and glottal closure on the signal quality of the subglottal pressure and the supraglottal acoustics, whereas VF symmetry only had a negligible influence. Strong correlations were found between the subglottal and supraglottal signal quality, with significant improvement of the acoustic quality for high levels of periodicity and glottal closure.

No agreed-upon method currently exists for objective measurement of perceived voice quality. This paper describes validation of a psychoacoustic model designed to fill this gap. This model includes parameters to characterize the harmonic and inharmonic voice sources, vocal tract transfer function, fundamental frequency, and amplitude of the voice, which together serve to completely quantify the integral sound of a target voice sample. In experiment 1, 200 voices with and without diagnosed vocal pathology were fit with the model using analysis-by-synthesis. The resulting synthetic voice samples were not distinguishable from the original voice tokens, suggesting that the model has all the parameters it needs to fully quantify voice quality. In experiment 2 parameters that model the harmonic voice source were removed one by one, and the voice tokens were re-synthesized with the reduced model. In every case the lower-dimensional models provided worse perceptual matches to the quality of the natural tokens than did the original set, indicating that the psychoacoustic model cannot be reduced in dimensionality without loss of fit to the data. Results confirm that this model can be validly applied to quantify voice quality in clinical and research applications.

As existing descriptions are likely too narrow to reflect a broader range of articulatory variability in Mandarin production, this study is undertaken to explore qualitative and quantitative tongue shape analysis in Mandarin sibilants. Tongue movement data are collected from 18 adult Mandarin speakers producing six sibilants in three vowel contexts. Acoustic information is also analyzed to establish the articulatory-acoustic correspondence. In addition to the common retroflex and bunched shapes, the results discovered a humped shape (e.g., a single, posterior lingual constriction) in most of Mandarin retroflex tokens. This shape is one variant of North American English /r/, but yet identified in Mandarin production. The humped shape adds to the literature and expands existing descriptions of Mandarin retroflex tongue configurations. Despite the shape differences, the general many-to-one articulatory-acoustic mappings also holds true for Mandarin retroflexes. However, while curvature analyses based on Cartesian coordinates significantly differentiated contrastive shapes in retroflex production, these analyses were not equally reliable in separating the alveolar-retroflex distinction, likely due to individual differences. The tongue contour changes in the place contrast were instead quantified by calculations with polar coordinates. The preliminary findings on Mandarin retroflexes are discussed in terms of vocal tract morphology, with possible lip protrusion.

Directivity of speech and singing is determined primarily by the morphology of a person, i.e., head size, torso dimensions, posture, and vocal tract. Previous works have suggested from measurements that voice directivity in singing is controlled unintentionally by spectral emphasis in the range of 2-4 kHz. The attempt is made to try to identify to what extent voice directivity is affected by the mouth configuration and the torso. Therefore, simulations, together with measurements that investigate voice directivity in more detail, are presented. Simulations are presented for a piston in an infinite baffle, a radiating spherical cap, and an extended spherical cap model, taking into account transverse propagation modes. Measurements of a classical singer, an amateur singer, and a head and torso simulator are undertaken simultaneously in the horizontal and vertical planes. In order to assess differences of voice directivity common metrics, e.g., horizontal and vertical directivity indexes, are discussed and compared to improved alternatives. The measurements and simulations reveal that voice directivity in singing is affected if the mouth opening is changed significantly. The measurements show that the torso generates side lobes due to diffraction and reflections at frequencies related to the torso's dimensions.

In the Northern Horpa (NH) language of Sichuan, vowels are divided between plain and pharyngealized sets, with the latter pronounced with auxiliary articulatory gestures involving more constriction in the vocal tract. The current study examines how the NH vocalic contrast is manifested in line with the process of pharyngealization both acoustically and articulatorily, based on freshly gathered data from two varieties of the language (i.e., Rtsangkhog and Yunasche). Along with formant analyses, ultrasound imaging was employed to capture the tongue postures and positions during vowel production. The results show that in contrast with plain vowels, pharyngealized vowels generally feature lower F2 values and higher F1 and F3 values. Mixed results for F2 and F3 suggest that the quality contrasts are vowel-dependent. Ultrasound images, on the other hand, reveal that the vocalic distinction is affected by different types of tongue movements, including retraction, backing, and double bunching, depending on the inherent tongue positions for each vowel. The two NH varieties investigated are found to display differential formant changes and different types of tongue displacements. The formant profiles along with ultrasound images support the view that the production of the NH phonologically marked vowels is characteristic of pharyngealization.

This letter introduces a parametrization of the vocal tract area function based on the position of a few points along the vocal tract. A QR decomposition algorithm is applied to area function data in various vowel configurations in order to identify those points with the most independent position patterns across vowels. Each point defines the shape of an associated kinematic region, and the overall area function is determined by the combination of the kinematic regions' shapes. The results show that only four data points, located at the tongue body, lips, and two at the tongue back, are enough to obtain accurate reconstructions of the vowels' area functions.

The purpose of this study was to determine the threshold of velopharyngeal coupling area at which listeners switch from identifying a consonant as a stop to a nasal in North American English, based on V1CV2 stimuli generated with a speech production model that encodes phonetic segments as relative acoustic targets. Each V1CV2 was synthesized with a set of velopharyngeal coupling functions whose area ranged from 0 to 0.1 cm2. Results show that consonants were identified by listeners as a stop when the coupling area was less than 0.035–0.057 cm2, depending on place of articulation and final vowel. The smallest coupling area (0.035 cm2) at which the stop-to-nasal switch occurred was found for an alveolar consonant in the /ɑCi/ context, whereas the largest (0.057 cm2) was for a bilabial in /ɑCɑ/. For each stimulus, the balance of oral versus nasal acoustic energy was characterized by the peak nasalance during the consonant. Stimuli with peak nasalance below 40% were mostly identified by listeners as stops, whereas those above 40% were identified as nasals. This study was intended to be a precursor to further investigations using the same model but scaled to represent the developing speech production system of male and female talkers.

This computational study aims to identify vocal tract adjustments that minimize the peak vocal fold contact pressure during phonation and thus should be targeted in voice therapy treating phonotraumatic vocal hyperfunction. The results showed that for a given subglottal pressure, the effect of vocal tract adjustments on the peak vocal fold contact pressure was generally small except when such adjustments caused noticeable changes in the glottal flow amplitude. In this study, this occurred mainly when the lip opening was reduced and at conditions of large initial glottal angles or high subglottal pressures, which decreased the peak contact pressure but also significantly reduced the output sound pressure level (SPL). On the other hand, increasing lip opening significantly increased sound radiation efficiency from the mouth and reduced the subglottal pressure required to produce a target SPL. Because of the large effect of the subglottal pressure on the peak contact pressure, increasing lip opening thus was able to significantly reduce the peak contact pressure in voice tasks targeting a specific SPL. In contrast, the effect of pharyngeal expansion alone had only a small effect on the peak contact pressure, whether controlling for the subglottal pressure or targeting a specific SPL.

In this work, vocal tract characteristic changes under the out-of-breath condition are explored. Speaking under the influence of physical exercise is called out-of-breath speech. The change in breathing pattern results in perceptual changes in the produced sound. For vocal tract, the first four formants show a lowering in their average frequency. The bandwidths BF1 and BF2 widen, whereas the other two get narrowed. The change in bandwidth is small for the last three. For a speaker, the change in frequency and bandwidth may not be uniform across formants. Subband analysis is carried out around formants for comparing the variation of the vocal tract with the source. A vocal tract adaptive empirical wavelet transform is used for extracting formant specific subbands from speech and source. The support vector machine performs the subband-based binary classification between the normal and out-of-breath speech. For all speakers, it shows an F1-score improvement of 4% over speech subbands. Similarly, a performance improvement of 5% can be seen for both male and female speakers. Furthermore, the misclassification amount is less for source compared to speech. These results suggest that physical exercise influences the source more than the vocal tract.

The presence of hypernasality in repaired cleft palate (CP) speech is a consequence of velopharyngeal insufficiency. The coupling of the nasal tract with the oral tract adds nasal formant and antiformant pairs in the hypernasal speech spectrum. This addition deviates the spectral and linear prediction (LP) residual characteristics of hypernasal speech compared to normal speech. In this work, the vocal tract constriction feature, peak to side-lobe ratio feature, and spectral moment features augmented by low-order cepstral coefficients are used to capture the spectral and residual deviations for hypernasality detection. The first feature captures the lower-frequencies prominence in speech due to the presence of nasal formants, the second feature captures the undesirable signal components in the residual signal due to the nasal antiformants, and the third feature captures the information about formants and antiformants in the spectrum along with the spectral envelope. The combination of three features gives normal versus hypernasal speech detection accuracies of 87.76%, 91.13%, and 93.70% for /a/, /i/, and /u/ vowels, respectively, and hypernasality severity detection accuracies of 80.13% and 81.25% for /i/ and /u/ vowels, respectively. The speech data are collected from 30 control normal and 30 repaired CP children between the ages of 7 and 12.

A model is described in which the effects of articulatory movements to produce speech are generated by specifying relative acoustic events along a time axis. These events consist of directional changes of the vocal tract resonance frequencies that, when associated with a temporal event function, are transformed via acoustic sensitivity functions, into time-varying modulations of the vocal tract shape. Because the time course of the events may be considerably overlapped in time, coarticulatory effects are automatically generated. Production of sentence-level speech with the model is demonstrated with audio samples and vocal tract animations.

The goal of this study is to identify vocal fold geometric and mechanical conditions that are likely to produce large contact pressure and thus high risk of vocal fold injury. Using a three-dimensional computational model of phonation, parametric simulations are performed with co-variations in vocal fold geometry and stiffness, with and without a vocal tract. For each simulation, the peak contact pressure is calculated. The results show that the subglottal pressure and the transverse stiffness of the vocal folds in the coronal plane have the largest and most consistent effect on the peak contact pressure, indicating the importance of maintaining a balance between the subglottal pressure and transverse stiffness to avoiding vocal fold injury. The presence of a vocal tract generally increases the peak contact pressure, particularly for an open-mouth vocal tract configuration. While a low degree of vocal fold approximation significantly reduces vocal fold contact pressure, for conditions of moderate and tight vocal fold approximation changes in vocal fold approximation may increase or decrease the peak contact pressure. The effects of the medial surface thickness and vocal fold stiffness along the anterior-posterior direction are similarly inconsistent and vary depending on other control parameters and the vocal tract configuration.

The estimation of formant frequencies from acoustic speech signals is mostly based on Linear Predictive Coding (LPC) algorithms. Since LPC is based on the source-filter model of speech production, the formant frequencies obtained are often implicitly regarded as those for an infinite glottal impedance, i.e., a closed glottis. However, previous studies have indicated that LPC-based formant estimates of vowels generated with a realistically varying glottal area may substantially differ from the resonances of the vocal tract with a closed glottis. In the present study, the deviation between closed-glottis resonances and LPC-estimated formants during phonation with different peak glottal areas has been systematically examined both using physical vocal tract models excited with a self-oscillating rubber model of the vocal folds, and by computer simulations of interacting source and filter models. Ten vocal tract resonators representing different vowels have been analyzed. The results showed that F1 increased with the peak area of the time-varying glottis, while F2 and F3 were not systematically affected. The effect of the peak glottal area on F1 was strongest for close-mid to close vowels, and more moderate for mid to open vowels.

Unlike aspiration in stops, occurrence of aspiration in non-stop consonants is quite rare. Most of the languages that have aspirated non-stop consonants are low-resource languages. Hence, data driven, quantitative, and statistical analysis of their aspiration phenomena is fairly limited. Rabha and Angami are considered in this study, as previous studies have confirmed the existence of aspiration contrast in fricatives and nasals. This study reports the acoustic characteristics of aspiration in stops, fricatives, and nasals. Among them, distinguishing the aspirated fricatives and aspirated nasals from their unaspirated counterparts is a challenging task. A set of acoustic features is proposed to automatically detect the presence of aspiration in fricatives and nasals. Acoustic features, such as vocal tract constriction (VTC), normalized autocorrelation peak strength (NAPS), strength of excitation (SoE), and variance of successive epoch intervals (VSEI) are used to detect aspiration in fricatives and nasals. These features are extracted from zero-frequency filtered signal of the speech sounds, as it preserves the aspiration information. Results show that VTC, NAPS, and SoE can detect aspiration in nasals, whereas SoE and VSEI can detect aspiration in fricatives. The proposed method improves the performance of an automatic phoneme recognizer by reducing the confusion between aspirated and unaspirated counterparts.

Lateral approximant speech sounds are notoriously difficult to measure and describe due to their complex articulation and acoustics. This has prevented researchers from reaching a unifying description of the articulatory and acoustic characteristics of laterals. This paper examines articulatory and acoustic properties of Brazilian Portuguese alveolar and palatal lateral approximants (/l/ and /ʎ/) produced by six native speakers. The methodology for obtaining vocal tract area functions was based on three-dimensional/four-dimensional (3D/4D) ultrasound recordings and 3D digitized palatal impressions with simultaneously recorded audio signals. Area functions were used to calculate transfer function spectra, and predicted formant and anti-resonance frequencies were compared with the acoustic recordings. Mean absolute error in formant frequency prediction was 4% with a Pearson correlation of r = 0.987. Findings suggest anti-resonances from the interdental channels are less important than a prominent anti-resonance from the supralingual cavity but can become important in asymmetrical articulations. The use of 3D/4D ultrasound to study articulatory-acoustic relations is promising, but significant limitations remain and future work is needed to make better use of 3D/4D ultrasound data, e.g., by combining it with magnetic resonance imaging.

Dysfunction of the velopharyngeal valve in the human airway causes speech disorders because there is no separation between the oral and nasal cavities during normal oral speech. The speech literature hypothesizes that undesired sound is formed by turbulent flow in the nasal cavity in cases of small velopharyngeal openings. The aim is to determine the flow behavior and the sound-generating mechanism in the vocal tract using computational fluid dynamics in two patient-specific models with small and large velopharyngeal openings and contrast it with cases of complete velopharyngeal closure. The geometry for the models was reconstructed from computed tomography scans that were taken while the patients were sustaining a sibilant sound. The results for the turbulence are correlated with the broadband acoustic models of Proudman and Curle. The models show that turbulence in the vocal tract increases downstream of a constriction and that sound may be generated from it. Furthermore, most of the sound due to turbulence in the nasal cavity is governed by a dipole source where turbulence interacts with the nasal cavity walls. The generated sound power by turbulence itself in the nasal cavity (the quadrupole source) is two orders of magnitude less than the dipole source.

This paper describes the inverse estimation of the vocal tract shape for vowels by using a vocal tract mapping interface. In prior research, an interface capable of generating a vocal tract shape by clicking on its window was developed. The vocal tract shapes for five vowels are located at the vertices of a pentagonal chart and a different shape that corresponds to an arbitrary mouse-pointer position on the interface window is calculated by interpolation. In this study, an attempt was made to apply the interface to the inverse estimation of vocal tract shapes based on formant frequencies. A target formant frequency data set was searched based on the geometry of the interface window by using a coarse to fine algorithm. It was revealed that the estimated vocal tract shapes obtained from the mapping interface were close to those from magnetic resonance imaging data in another study and to lip area data captured using video recordings. The results of experiments to evaluate the estimated vocal tract shapes showed that each subject demonstrated unique trajectories on the interface window corresponding to the estimated vocal tract shapes. These results suggest the usefulness of inverse estimation using the interface.

In speech production, the motor system organizes articulators such as the jaw, tongue, and lips into synergies whose function is to produce speech sounds by forming constrictions at the phonetic places of articulation. The present study tests whether synergies for different constriction tasks differ in terms of inter-articulator coordination. The test is conducted on utterances [ɑpɑ], [ɑtɑ], [ɑiɑ], and [ɑkɑ] with a real-time magnetic resonance imaging biomarker that is computed using a statistical model of the forward kinematics of the vocal tract. The present study is the first to estimate the forward kinematics of the vocal tract from speech production data. Using the imaging biomarker, the study finds that the jaw contributes least to the velar stop for [k], more to pharyngeal approximation for [ɑ], still more to palatal approximation for [i], and most to the coronal stop for [t]. Additionally, the jaw contributes more to the coronal stop for [t] than to the bilabial stop for [p]. Finally, the study investigates how this pattern of results varies by participant. The study identifies differences in inter-articulator coordination by constriction task, which support the claim that inter-articulator coordination differs depending on the active articulator synergy.

Nasal cavities are known to introduce antiresonances (dips) in the sound spectrum reducing the acoustic power of the voice. In this study, a three-dimensional (3D) finite element (FE) model of the vocal tract (VT) of one female subject was created for vowels [a:] and [i:] without and with a detailed model of nasal cavities based on CT (Computer Tomography) images. The 3D FE models were then used for analyzing the resonances, antiresonances and the acoustic pressure response spectra of the VT. The computed results were compared with the measurements of a VT model for the vowel [a:], obtained from the FE model by 3D printing. The nasality affects mainly the lowest formant frequency and decreases its peak level. The results confirm the main effect of nasalization, i.e., that sound pressure level decreases in the frequency region of the formants F1-F2 and emphasizes the frequency region of the formants F3-F5 around the singer's formant cluster. Additionally, many internal local resonances in the nasal and paranasal cavities were found in the 3D FE model. Their effect on the acoustic output was found to be minimal, but accelerometer measurements on the walls of the 3D-printed model suggested they could contribute to structure vibrations.

Differences in voice pitch (F0) and vocal tract length (VTL) improve intelligibility of speech masked by a background talker (speech-on-speech; SoS) for normal-hearing (NH) listeners. Cochlear implant (CI) users, who are less sensitive to these two voice cues compared to NH listeners, experience difficulties in SoS perception. Three research questions were addressed: (1) whether increasing the F0 and VTL difference (Delta F0; Delta VTL) between two competing talkers benefits CI users in SoS intelligibility and comprehension, (2) whether this benefit is related to their F0 and VTL sensitivity, and (3) whether their overall SoS intelligibility and comprehension are related to their F0 and VTL sensitivity. Results showed: (1) CI users did not benefit in SoS perception from increasing Delta F0 and Delta VTL; increasing Delta VTL had a slightly detrimental effect on SoS intelligibility and comprehension. Results also showed: (2) the effect from increasing Delta F0 on SoS intelligibility was correlated with F0 sensitivity, while the effect from increasing Delta VTL on SoS comprehension was correlated with VTL sensitivity. Finally, (3) the sensitivity to both F0 and Delta VTL, and not only one of them, was found to be correlated with overall SoS performance, elucidating important aspects of voice perception that should be optimized through future coding strategies. (C) 2019 Author(s).

Three-dimensional (3D/4D) ultrasound (US) imaging of the tongue has emerged as a useful instrument for articulatory studies. However, extracting quantitative measurements of the shape of the tongue surface remains challenging and time-consuming. In response to these challenges, this paper documents and evaluates the first automated method for extracting tongue surfaces from 3D/4D US data. The method draws on established methods in computer vision, and combines image phase symmetry measurements, eigen-analysis of the image Hessian matrix, and a fast marching method for surface evolution towards the automatic detection of the sheet-like surface of the tongue amidst noisy US data. The method was tested on US recordings from eight speakers and the resulting automatically extracted tongue surfaces were generally found to lie within 1 to 2 mm from their corresponding manually delineated surfaces in terms of mean-sum-of-distances error. Further experiments demonstrate that the accuracy of 2D midsagittal tongue contour extraction is also improved using 3D data and methods. This is likely because the additional information afforded by 3D US compared to 2D US images strongly constrains the possible location of the midsagittal contour. Thus, the proposed method seems appropriate for immediate practical use in the analysis of 3D/4D US recordings of the tongue.

Previous research showed that aspiration noise difference limens in moderately breathy /a/ vowels decreased as the spectral slope of the glottal source spectrum became increasingly steep [Kreiman and Gerratt, J. Acoust. Soc. Am. 131(1), 492–500 (2012)]. The current study investigated whether discrimination of aspiration noise levels was affected by differences in spectral shape due to vowel quality (/ae/ and /i/) and speaker identity (three male speakers) when the slope of the glottal source spectrum was fixed. The results showed that discrimination performance was worse overall for /i/ than /ae/, but the result may have resulted from relatively poor performance for the /i/ vowel of one speaker. Acoustic analyses of the stimuli were performed to estimate the association between acoustic properties and the perceptual outcomes. The results showed that both the smoothed cepstral peak prominence and the harmonic energy level between 2 and 5 kHz may account for the observed differences in aspiration noise discrimination among speakers within each vowel, but not for differences between vowel categories. It is possible that the relationship between the aspiration noise discrimination and aforementioned acoustic properties may be modulated by the spectral distribution of energy across frequency.Previous research showed that aspiration noise difference limens in moderately breathy /a/ vowels decreased as the spectral slope of the glottal source spectrum became increasingly steep [Kreiman and Gerratt, J. Acoust. Soc. Am. 131(1), 492–500 (2012)]. The current study investigated whether discrimination of aspiration noise levels was affected by differences in spectral shape due to vowel quality (/ae/ and /i/) and speaker identity (three male speakers) when the slope of the glottal source spectrum was fixed. The results showed that discrimination performance was worse overall for /i/ than /ae/, but the result may have resulted from relatively poor performance for the /i/ vowel of one speaker. Acoustic analyses of the stimuli were performed to estimate the association between acoustic properties and the perceptual outcomes. The results showed that both the smoothed cepstral peak prominence and the harmonic energy level between 2 and 5 kHz may account for the observed differences in aspiration noise discri...

Real-time magnetic resonance imaging (MRI) speech production data have expanded the understanding of vocal tract actions. This letter presents an Automatic Centroid Tracking tool, ACT, which obtains both spatial and temporal information characterizing multi-directional articulatory movement. ACT auto-segments an articulatory object composed of connected pixels in a real-time MRI video, by finding its intensity centroids over time and returns kinematic profiles including direction and magnitude information of the object. This letter discusses the utility of ACT, which outperforms other similar object tracking techniques, by demonstrating its successful online tracking of vertical larynx movement. ACT can be deployed generally for dynamic image processing and analysis.

The goal of this study is to identify vocal fold conditions that produce irregular vocal fold vibration and the underlying physical mechanisms. Using a three-dimensional computational model of phonation, parametric simulations are performed with co-variations in vocal fold geometry, stiffness, and vocal tract shape. For each simulation, the cycle-to-cycle variations in the amplitude and period of the glottal area function are calculated, based on which the voice is classified into three types corresponding to regular, quasi-steady or subharmonic, and chaotic phonation. The results show that vocal folds with a large medial surface vertical thickness and low transverse stiffness are more likely to exhibit irregular vocal fold vibration when tightly approximated and subject to high subglottal pressure. Transition from regular vocal fold vibration to vocal instabilities is often accompanied by energy redistribution among the first few vocal fold eigenmodes, presumably due to nonlinear interaction between eigenmodes during vocal fold contact. The presence of a vocal tract may suppress such contact-related vocal instabilities, but also induce new instabilities, particularly for less constricted vocal fold conditions, almost doubling the number of vocal fold conditions producing irregular vibration.

In this paper, acoustic analysis of misarticulated trills in cleft lip and palate speakers is carried out using excitation source based features: strength of excitation and fundamental frequency, derived from zero-frequency filtered signal, and vocal tract system features: first formant frequency (F1) and trill frequency, derived from the linear prediction analysis and autocorrelation approach, respectively. These features are found to be statistically significant while discriminating normal from misarticulated trills. Using acoustic features, dynamic time warping based trill misarticulation detection system is demonstrated. The performance of the proposed system in terms of the F1-score is 73.44%, whereas that for conventional Mel-frequency cepstral coefficients is 66.11%.

International audience; Acoustic variation in expressive speech at the syllable level is studied. As emotions or attitudes can be conveyed by short spoken words, analysis of paradigmatic variations in vowels is an important issue to characterize the expressive content of such speech segments. The corpus contains 160 sentences produced under seven expressive conditions (Neutral, Anger, Fear, Surprise, Sensuality, Joy, Sadness) acted by a French female speaker (a total of 1120 sentences, 13 140 vowels). Eleven base acoustic parameters are selected for voice source and vocal tract related feature analysis. An acoustic description of the expressions is drawn, using the dimensions of melodic range, intensity, noise, spectral tilt, vocalic space, and dynamic features. The first three functions of a discriminant analysis explain 95% of the variance in the data. These statistical dimensions are consistently associated with acoustic dimensions. Covariation of intensity and F0 explains over 80% of the variance, followed by noise features (8%), covariation of spectral tilt, and F0 (7%). On the basis of isolated vowels alone, expressions are classified with a mean accuracy of 78%.

Mental health illnesses like Major Depressive Disorder and Schizophrenia affect the coordination between articulatory gestures in speech production. Coordination features derived from Vocal tract variables (TVs) predicted by a speech inversion system can quantify the changes in articulatory gestures and have proven to be effective in the classification of mental health disorders. In this study we use data from the IEMOCAP (acted emotions) and MSP Podcast (natural emotions) datasets to understand how coordination features extracted from TVs can be used to capture changes between different emotions for the first time. We compared the eigenspectra extracted from channel delay correlation matrices for Angry, Sad and Happy emotions with respect to the “Neutral” emotion. Across both the datasets, it was observed that the “Sad” emotion follows a pattern suggesting simpler articulatory coordination while the “Angry” emotion follows the opposite showing signs of complex articulatory coordination. For the majority of subjects, the ‘Happy’ emotion follows a complex articulatory coordination pattern, but has significant confusion with “Neutral” emotion. We trained a Convolutional Neural Network with the coordination features as inputs to perform emotion classification. A detailed interpretation of the differences in eigenspectra and the results of the classification experiments will be discussed.

Dispersion Theory [DT; Liljencrants and Lindblom (1972). Language 12(1), 839-862] claims that acoustically dispersed vowel inventories should be typologically common. Dispersion is often quantified using triangle area between three mean vowel formant points. This approach is problematic; it ignores distributions, which affect speech perception [Clayards, Tanenhaus, Aslin, and Jacobs (2008). Cognition 108, 804-809]. This letter proposes a revised metric for calculating dispersion which incorporates covariance. As a test case, modeled vocal tract articulatory-acoustic data of stop consonants [Schwartz, Boe, Badin, and Sawallis (2012). J. Phonetics 40, 20-36] are examined. Although the revised metric does not recover DT predictions for stop inventories, it changes results, showing that dispersion results depend on metric choice, which is often overlooked. The metric can be used in any acoustic space to include information about within-category variation when calculating dispersion.

The effects of tongue position on sound properties were investigated by using simplified models of sibilant fricatives /s/ and /ʃ/. These were constructed from medical images of a native Japanese-speaking subject who pronounced both fricative sounds. The sounds generated by the models were experimentally measured and compared with the subjects /s/ and /ʃ/. The position of tongue models altered the main peak frequency and spectral mean of the generated sound, which resulted in the reproduction of frequency characteristics of /s/ or /ʃ/ with the subjects tongue position.

Speech produced by children is characterized by a high fundamental frequency which complicates measurement of vocal tract resonances, and hence coarticulation. Here two whole-spectrum measures of coarticulation are validated, one temporal and one spectral, that are less sensitive to these challenges. Using these measures, consonant-vowel coarticulation is calculated in the speech of a large sample of 4-year-old children. The measurements replicate known lingual coarticulatory findings from the literature, demonstrating the utility of these acoustic measures of coarticulation in speakers of all ages.

Aperiodicity in the voice source is caused by changes in the vocal fold vibrations, other than the normal quasi-periodicity and the turbulence at the glottis. The aperiodicity appears to be one of the main properties that is responsible for conveying the emotion in artistic voices. In this paper, the feasibility of representing the excitation source characteristics in artistic (Noh) singing voice by an impulse-like sequence in the time domain is examined. The impulses at the glottal closure instants contribute to the major excitation of the vocal tract system. The sequence of such impulses produces harmonics of the fundamental frequency in the spectrum. The amplitude variation or amplitude modulation (AM) of these impulses in the sequence contributes to the aperiodicity in the excitation, and can result in appearance of subharmonics in the spectrum. The variation in the impulse intervals or frequency modulation (FM) can also contribute to the aperiodicity in the excitation. The aperiodic component of the excitation in the Noh voice is examined in the impulse-like sequence derived from the signal using the single frequency filtering (SFF) analysis. The effects of aperiodicity are explained for synthetic AM and FM sequence of impulses using spectrograms and saliency plots.

This paper proposes a modular architecture for articulatory synthesis from a gestural specification comprising relatively simple models for the vocal tract, the glottis, aero-acoustics, and articulatory control. The vocal tract module combines a midsagittal statistical analysis articulatory model, derived by factor analysis of air-tissue boundaries in real-time magnetic resonance imaging data, with an [Formula: see text] model for converting midsagittal section to area function specifications. The aero-acoustics and glottis models were based on a software implementation of classic work by Maeda. The articulatory control module uses dynamical systems, which implement articulatory gestures, to animate the statistical articulatory model, inspired by the task dynamics model. Results on synthesizing vowel-consonant-vowel sequences with plosive consonants, using models that were built on data from, and simulate the behavior of, two different speakers are presented.

The cause of individual acoustic characteristics of sibilant fricatives /s/ and /ʃ/ was analyzed by extracting vocal tract geometries and conducting aeroacoustic experiments and simulations on each geometry. The vocal tract geometries of five Japanese subjects while sustaining /s/ and /ʃ/ were collected by magnetic resonance imaging. Flow and sound generation in the vocal tracts was predicted by large eddy simulations of compressible flow. The characteristic dimensions of the vocal tracts were extracted and simplified vocal tract models were constructed to clarify the relationship between the geometries and the acoustic characteristics. The acoustic characteristics of sounds generated by the simplified models agreed well with the sounds predicted by the simulation, indicating that the proposed model is able to express the individual characteristics in the production of sibilant fricatives. A comparison of the models showed that the volume and length of a space downstream from the constriction are key factors controlling the acoustic characteristics of each subject.

Under certain conditions, e.g., singing voice, the fundamental frequency of the vocal folds can go up and interfere with the formant frequencies. Acoustic feedback from the vocal tract filter to the vocal fold source then becomes strong and non-negligible. An experimental study was presented on such source–filter interaction using three types of synthetic vocal fold models. Asymmetry was also created between the left and right vocal folds. The experiment reproduced various nonlinear phenomena, such as frequency jump and quenching, as reported in humans. Increase in phonation threshold pressure was also observed when resonant frequency of the vocal tract and fundamental frequency of the vocal folds crossed each other. As a combined effect, the phonation threshold pressure was further increased by the left-right asymmetry. Simulation of the asymmetric two-mass model reproduced the experiments to some extent. One of the intriguing findings of this study is the variable strength of the source–filter interaction over different model types. Among the three models, two models were strongly influenced by the vocal tract, while no clear effect of the vocal tract was observed in the other model. This implies that the level of source–filter interaction may vary considerably from one subject to another in humans.

The goal of the study is to quantify the salient vocal tract acoustic, subglottal acoustic, and vocal tract physiological characteristics during phonation into a narrow flow-resistant tube with 2.53 mm inner diameter and 124 mm length in typically developing vocally healthy children using simultaneous microphone, accelerometer, and 3D/4D ultrasound recordings. Acoustic measurements included fundamental frequency (fo), first formant frequency (F1), second formant frequency (F2), first subglottal resonance (FSg1), and peak-to-peak amplitude ratio (Pvt:Psg). Physiological measurements included posterior tongue height (D1), tongue dorsum height (D2), tongue tip height (D3), tongue length (D4), oral cavity width (D5), hyoid elevation (D6), pharynx width (D7). All measurements were made on eight boys and ten girls (6–9 years) during sustained /o:/ production at typical pitch and loudness, with and without flow-resistant tube. Phonation with the flow-resistant tube resulted in a significant decrease in F1, F2, and Pvt:Psg and a significant increase in D2, D3, and FSg1. A statistically significant gender effect was observed for D1, with D1 higher in boys. These findings agree well with reported findings from adults, suggesting common acoustic and articulatory mechanisms for narrow flow-resistant tube phonation. Theoretical implications of the findings are discussed.

Speech inversion is a well-known ill-posed problem and addition of speaker differences typically makes it even harder. Normalizing the speaker differences is essential to effectively using multi-speaker articulatory data for training a speaker independent speech inversion system. This paper explores a vocal tract length normalization (VTLN) technique to transform the acoustic features of different speakers to a target speaker acoustic space such that speaker specific details are minimized. The speaker normalized features are then used to train a deep feed-forward neural network based speech inversion system. The acoustic features are parameterized as time-contextualized mel-frequency cepstral coefficients. The articulatory features are represented by six tract-variable (TV) trajectories, which are relatively speaker invariant compared to flesh point data. Experiments are performed with ten speakers from the University of Wisconsin X-ray microbeam database. Results show that the proposed speaker normalization approach provides an 8.15% relative improvement in correlation between actual and estimated TVs as compared to the system where speaker normalization was not performed. To determine the efficacy of the method across datasets, cross speaker evaluations were performed across speakers from the Multichannel Articulatory-TIMIT and EMA-IEEE datasets. Results prove that the VTLN approach provides improvement in performance even across datasets.

Studies with three-dimensional (3D) vocal tract visualization using magnetic resonance imaging (MRI) have suggested that hypopharyngeal cavities, i.e., laryngeal cavity and bilateral piriform fossa, may be the acoustic loci to express speaker characteristics in male speech sounds. Previous studies mainly investigated the hypopharynx in males but few for females. This study explored the hypopharynx morphological characteristics for the subjects in both genders by MRI. 3D numerical vocal tracts at vowels were reconstructed from the MRI datasets of three male and four female Chinese subjects. Geometrical measurements were conducted for the hypopharyngeal cavities. Morphological observations and statistical analyses revealed both commonalities and differences between the male and female subjects. The laryngeal cavity shapes in females were found similar to males resembling a Helmholtz resonator rather than a simple straight closed tube, and the bilateral piriform fossa cavities showed an asymmetry: the right is longer and wider than the left in both genders. As for the cavity size across vowels, for both the male and female subjects the laryngeal cavity and piriform fossa in /i/ were observed larger than those in /a/. To summarize gender characteristics, the female subjects were characterized by the smaller laryngeal cavity and piriform fossa compared with males.

I will demonstrate how I use browser-based software authored by Mark Huckvale to promote active learning of the acoustic theory of speech production. Students create the complex glottal tone using ESynth, filter the glottal tone with various vocal tract filters with ESystem, explore the impact of vocal tract shapes on formants using VTDemo, and examine their own vowel formants with ESection.