Lower Formant Frequencies Research Articles

Perception of age, gender, and talker height from children’s vowels

In his 1989 review of vowel perception [J. Acoust. Soc. Am. 85, 2088–2113], Terry Nearey observed that the acoustic properties of vowels vary substantially across talkers due to differences in talker size. Taller speakers tend to have lower formant frequencies and lower fundamental frequencies than shorter speakers, and listeners appear to take these relationships into account in vowel identification. In this talk, we follow up some of the questions raised in that paper, reviewing recent findings on the perception of age, gender, and talker height in children’s voices. Children exhibit substantial age-related changes in fundamental and formant frequencies due to growth of the larynx and vocal tract, presenting an interesting opportunity to explore the perceptual consequences of these changes. Listeners’ perceptual judgments of age and gender reveal complex interdependencies, consistent with the idea that these indexical attributes are jointly estimated, in a manner that reflects their shared dependence on acoustic parameters related to perceived talker height.

Altered neural encoding of vowels in noise does not affect behavioral vowel discrimination in gerbils with age-related hearing loss.

Understanding speech in a noisy environment, as opposed to speech in quiet, becomes increasingly more difficult with increasing age. Using the quiet-aged gerbil, we studied the effects of aging on speech-in-noise processing. Specifically, behavioral vowel discrimination and the encoding of these vowels by single auditory-nerve fibers were compared, to elucidate some of the underlying mechanisms of age-related speech-in-noise perception deficits. Young-adult and quiet-aged Mongolian gerbils, of either sex, were trained to discriminate a deviant naturally-spoken vowel in a sequence of vowel standards against a speech-like background noise. In addition, we recorded responses from single auditory-nerve fibers of young-adult and quiet-aged gerbils while presenting the same speech stimuli. Behavioral vowel discrimination was not significantly affected by aging. For both young-adult and quiet-aged gerbils, the behavioral discrimination between /eː/ and /iː/ was more difficult to make than /eː/ vs. /aː/ or /iː/ vs. /aː/, as evidenced by longer response times and lower d' values. In young-adults, spike timing-based vowel discrimination agreed with the behavioral vowel discrimination, while in quiet-aged gerbils it did not. Paradoxically, discrimination between vowels based on temporal responses was enhanced in aged gerbils for all vowel comparisons. Representation schemes, based on the spectrum of the inter-spike interval histogram, revealed stronger encoding of both the fundamental and the lower formant frequencies in fibers of quiet-aged gerbils, but no qualitative changes in vowel encoding. Elevated thresholds in combination with a fixed stimulus level, i.e., lower sensation levels of the stimuli for old individuals, can explain the enhanced temporal coding of the vowels in noise. These results suggest that the altered auditory-nerve discrimination metrics in old gerbils may mask age-related deterioration in the central (auditory) system to the extent that behavioral vowel discrimination matches that of the young adults.

Effective Vowel Stimuli for Measuring Occlusion Effect in the Pediatric Population

Past studies have reported that there are higher sound-pressure levels for each vowel in a child’s ear canal than those in adults due to reduced volume and a shorter ear canal. Furthermore, longer vocal tracts are associated with lower formant frequencies, and vice versa. The structural differences in this regard may reflect the difference in the occlusion effect. Thus, the present study compares the sound pressure levels (SPLs) and first formant frequencies of children and adults with normal hearing and determines the best vowel stimulus to assess the occlusion effect. A repeated measures research design was utilized to investigate the best stimulus with which to measure the occlusion effect among children and adults. Group 1 included ten children, and Group 2 comprised ten adults with normal hearing. The SPLs at frequencies between 200 Hz and 1000 Hz for three uttered vowels, with steps of 100 Hz, were measured using a hearing aid analyzer. The recorded vowels were saved in a ‘.wav’ format for formant frequency analysis. Furthermore, a paired comparison method was used to identify the vowel stimulus that most effectively induced the occlusion effect. A significantly higher SPL was observed for children compared to adults for each vowel. The formant frequency F1 value was higher for children than adults for each vowel, constituting a significant finding. In the paired comparison, the occlusion effect was reported to be significantly greater with respect to the vowel /u/ among adults, while it was reported to be greater in relation to the vowel /i/ among children. The vowel /u/ was the best stimulus with which to assess the occlusion effect among adults. The vowel /i/ was the best stimulus with which to assess the occlusion effect among children.

Distinctive acoustic changes in speech in Parkinson's disease

Dysarthria occurs in the early stage of Parkinson's disease (PD), and the pronunciation of speech sounds is significantly different between PD patients and healthy subjects. However, the formant frequencies of different syllables as vocalized by Chinese PD patients remain unknown. In this study, we recruited 83 PD patients and 44 healthy controls to perform a syllable phonation test. The results revealed that all PD patients presented lower formant frequencies than the control group for various vocalized syllables, presenting a weak correlation with the severity of the disease.

Morphological disparity and evolutionary transformations in the primate hyoid apparatus

The hyoid apparatus plays an integral role in swallowing, respiration, and vocalization in mammals. Most placental mammals have a rod-shaped basihyal connected to the basicranium via both soft tissues and a mobile bony chain—the anterior cornu—whereas anthropoid primates have broad, shield-like or even cup-shaped basihyals suspended from the basicranium by soft tissues only. How the unique anthropoid hyoid morphology evolved is unknown, and hyoid morphology of nonanthropoid primates is poorly documented. Here we use phylogenetic comparative methods and linear morphometrics to address knowledge gaps in hyoid evolution among primates and their euarchontan outgroups. We find that dermopterans have variable reduction of cornu elements. Cynocephalus volans are sexually dimorphic in hyoid morphology. Tupaia and all lemuroids except Daubentonia have a fully ossified anterior cornu connecting a rod-shaped basihyal to the basicranium; this is the ancestral mammalian pattern that is also characteristic of the last common ancestor of Primates. Haplorhines exhibit a reduced anterior cornu, and anthropoids underwent further increase in basihyal aspect ratio values and in relative basihyal volume. Convergent with haplorhines, lorisoid strepsirrhines independently evolved a broad basihyal and reduced anterior cornua. While a reduced anterior cornu is hypothesized to facilitate vocal tract lengthening and lower formant frequencies in some mammals, our results suggest vocalization adaptations alone are unlikely to drive the iterative reduction of anterior cornua within Primates. Our new data on euarchontan hyoid evolution provide an anatomical basis for further exploring the form-function relationships of the hyoid across different behaviors, including vocalization, chewing, and swallowing.

Language effects on acoustic voice variation within and between talkers

Acoustic voice spaces for English speakers are characterized mainly by variability in F0, the balance between higher harmonic amplitudes and inharmonic energy, and higher formant frequencies [JASA 146(4), 3011 (2019)]. We extended this investigation to another language to test the hypothesis that a few biologically relevant measures will emerge commonly across languages, while remaining variance will depend on the structure of the language. This hypothesis was tested against sentence productions from 5 female and 5 male speakers of Seoul Korean. Like English, Korean does not have tone or phonation contrasts, but Seoul Korean exhibits specific phrase intonation patterns. PCAs were performed on scaled values of F0, formant frequencies, spectral noise, source spectral shape, and their variability, measured from vowels and approximants. Results revealed striking similarities between the acoustic voice spaces derived from Korean speakers and those for English speakers. For Korean voices, F0 and variability in lower formant frequencies (i.e., vowel quality) accounted for the most acoustic variance within and across talkers, presumably due to Seoul speakers' systematic use of these measures for phrasal/accentual information. These measures were insignificant for English voices. Our findings suggest that acoustic voice spaces are shaped by both biologically and phonologically relevant factors. [Work supported by NIH/NSF.]

A Preliminary Study on Malay Vowel Formant for Young Adult and Aged People

This paper presents a study on Malay vowel formants of different age group between young adults and aged people. About 12 of the available young adult speech (6 males and 6 females; mean age 23 years) and 4 of the recorded aged people speech (2 males and 2 females; mean 67.3 years) were examined for analysis of formants to study the effect of age and gender. This work is significant in studying the hypothesis that the formant of aged people is lower than young adult population. The findings obtained shows that the formant of aged group is not critically affected by their age and male groups produces lower formant frequencies compared to the female groups either young adult or aged people. Formant frequency values also subject to variation due to the influence of surrounding phonemes.

An investigation of the systematic use of spectral information in the determination of apparent-talker height

The perception of apparent-talker height is mostly determined by the fundamental frequency (f0) and spectral characteristics of a voice. Although it is traditionally thought that spectral cues affect apparent-talker height by influencing apparent vocal-tract length, a recent experiment [Barreda (2016). J. Phon. 55, 1-18] suggests that apparent-talker height can vary significantly within-talker on the basis of phonemically-determined spectral variability. In this experiment, listeners were asked to estimate the height of 10 female talkers based on manipulated natural productions of bVd words containing one of /i æ ɑ u ɝ/. Results indicate that although listeners appear to use vocal-tract length estimates in determining apparent-height, apparent-talker height also varies significantly within-talker based on the inherent spectral and source characteristics of different vowels, with vowels with lower formant-frequencies and f0 being associated with taller talkers overall. The use of spectral and f0 information in apparent-height estimation varied considerably between listeners, resulting in additional variation in the apparent-height of talkers. Although the use of acoustic information in the determination of apparent-height was highly systematic, it does not necessarily follow from the empirical relationship between speech acoustics and actual talker height.

Evolutionary Trade-Off between Vocal Tract and Testes Dimensions in Howler Monkeys

SummaryMales often face a trade-off between investments in precopulatory and postcopulatory traits [1], particularly when male-male contest competition determines access to mates [2]. To date, studies of precopulatory strategies have largely focused on visual ornaments (e.g., coloration) or weapon morphology (e.g., antlers, horns, and canines). However, vocalizations can also play an important role in both male competition and female choice [3, 4, 5]. We investigated variation in vocal tract dimensions among male howler monkeys (Alouatta spp.), which produce loud roars using a highly specialized and greatly enlarged hyoid bone and larynx [6]. We examined the relative male investment in hyoids and testes among howler monkey species in relation to the level of male-male competition and analyzed the acoustic consequences of variation in hyoid morphology. Species characterized by single-male groups have large hyoids and small testes, suggesting high levels of vocally mediated competition. Larger hyoids lower formant frequencies, probably increasing the acoustic impression of male body size and playing a role analogous to investment in large body size or weaponry. Across species, as the number of males per group increases, testes volume also increases, indicating higher levels of postcopulatory sperm competition, while hyoid volume decreases. These results provide the first evidence of an evolutionary trade-off between investment in precopulatory vocal characteristics and postcopulatory sperm production.

Identification of high-pass filtered male, female, and child vowels: The use of high-frequency cues.

Vowels are characteristically described according to low-frequency resonance characteristics, which are presumed to provide the requisite information for identification. Classically, the study of vowel perception has focused on the lowest formant frequencies, typically F1, F2, and F3. Lehiste and Peterson [Phonetica 4, 161-177 (1959)] investigated identification accuracy of naturally produced male vowels composed of various amounts of low- and high-frequency content. Results showed near-chance identification performance for vowel segments containing only spectral information above 3.5 kHz. The authors concluded that high-frequency information was of minor importance for vowel identification. The current experiments report identification accuracy for high-pass filtered vowels produced by two male, two female, and two child talkers using both between- and within-subject designs. Identification performance was found to be significantly above chance for the majority of vowels even after high-pass filtering to remove spectral content below 3.0-3.5 kHz. Additionally, the filtered vowels having the highest fundamental frequency (child talkers) often had the highest identification accuracy scores. Linear discriminant function analysis mirrored perceptual performance when using spectral peak information between 3 and 12 kHz.

Speaker identification based on artificial neural networks. Case study: the Polish vowel (pilot study)

Abstract Current statistical methods and technologies used for speaker identification via dynamic formant frequency often involve classic multivariate analyses that must meet a number of criteria in order to be considered trustworthy. The authors propose more advanced classification techniques, including artificial neural networks. Owing to iterative learning algorithms, neural networks can be trained to detect highly complex, nonlinear relations hidden in input data. This study specifically considers feed-forward multilayer perceptron and radial basic function network models. The investigation involves an analysis of the Polish vowel (stressed or unstressed) in selected contexts described by the four lowest formant frequencies. Results indicate high accuracy of neural networks as a speaker identification tool reaching up to 100%. In addition, the authors have determined that the accuracy of classification is similar when based on a single context to when input data are aggregated over several different contexts.

Joint Source-Filter Optimization for Accurate Vocal Tract Estimation Using Differential Evolution

In this work, we present a joint source-filter optimization approach for separating voiced speech into vocal tract (VT) and voice source components. The presented method is pitch-synchronous and thereby exhibits a high robustness against vocal jitter, shimmer and other glottal variations while covering various voice qualities. The voice source is modeled using the Liljencrants-Fant (LF) model, which is integrated into a time-varying auto-regressive speech production model with exogenous input (ARX). The non-convex optimization problem of finding the optimal model parameters is addressed by a heuristic, evolutionary optimization method called differential evolution. The optimization method is first validated in a series of experiments with synthetic speech. Estimated glottal source and VT parameters are the criteria used for comparison with the iterative adaptive inverse filter (IAIF) method and the linear prediction (LP) method under varying conditions such as jitter, fundamental frequency (f0) as well as environmental and glottal noise. The results show that the proposed method largely reduces the bias and standard deviation of estimated VT coefficients and glottal source parameters. Furthermore, the performance of the source-filter separation is evaluated in experiments using speech generated with a physical model of speech production. The proposed method reliably estimates glottal flow waveforms and lower formant frequencies. Results obtained for higher formant frequencies indicate that research on more accurate voice source models and their interaction with the VT is necessary to improve the source-filter separation. The proposed optimization approach promises to be a useful tool for future research addressing this topic.

Sex-biased sound symbolism in english-language first names.

Sexual selection has resulted in sex-based size dimorphism in many mammals, including humans. In Western societies, average to taller stature men and comparatively shorter, slimmer women have higher reproductive success and are typically considered more attractive. This size dimorphism also extends to vocalisations in many species, again including humans, with larger individuals exhibiting lower formant frequencies than smaller individuals. Further, across many languages there are associations between phonemes and the expression of size (e.g. large /a, o/, small /i, e/), consistent with the frequency-size relationship in vocalisations. We suggest that naming preferences are a product of this frequency-size relationship, driving male names to sound larger and female names smaller, through sound symbolism. In a 10-year dataset of the most popular British, Australian and American names we show that male names are significantly more likely to contain larger sounding phonemes (e.g. “Thomas”), while female names are significantly more likely to contain smaller phonemes (e.g. “Emily”). The desire of parents to have comparatively larger, more masculine sons, and smaller, more feminine daughters, and the increased social success that accompanies more sex-stereotyped names, is likely to be driving English-language first names to exploit sound symbolism of size in line with sexual body size dimorphism.

Musicians have fine-tuned neural distinction of speech syllables

One of the benefits musicians derive from their training is an increased ability to detect small differences between sounds. Here, we asked whether musicians’ experience discriminating sounds on the basis of small acoustic differences confers advantages in the subcortical differentiation of closely related speech sounds (e.g., /ba/ and /ga/), distinguishable only by their harmonic spectra (i.e., their second formant trajectories). Although the second formant is particularly important for distinguishing stop consonants, auditory brainstem neurons do not phase-lock to its frequency range (above 1000Hz). Instead, brainstem neurons convert this high-frequency content into neural response timing differences. As such, speech tokens with higher formant frequencies elicit earlier brainstem responses than those with lower formant frequencies. By measuring the degree to which subcortical response timing differs to the speech syllables /ba/, /da/, and /ga/ in adult musicians and nonmusicians, we reveal that musicians demonstrate enhanced subcortical discrimination of closely related speech sounds. Furthermore, the extent of subcortical consonant discrimination correlates with speech-in-noise perception. Taken together, these findings show a musician enhancement for the neural processing of speech and reveal a biological mechanism contributing to musicians’ enhanced speech perception in noise.

Training listeners to report fundamental frequency and formant range information independently.

The vowels of speakers of different sizes vary in terms of their average f0 and formant frequencies. In general, larger speakers produce vowels with lower formant frequencies and lower f0s. Listeners have demonstrated the ability to estimate the approximate size of a speaker and previous experiments have shown that these judgments are based on the joint consideration of f0 and formant range information. Thus both lower f0s and lower formant frequency ranges are associated with larger speakers by listeners. Studies which have asked listeners to evaluate voices have focused on the extraction of apparent speaker characteristics (which are informed by f0 and formant frequencies) rather than asking speakers to report f0 and formant range information directly. The current study consists of a training procedure by which participants will learn to report the f0 and formant range of voices independently. The training consists of a voice matching game in which participants hear a pair of vowels produced by a voice and are asked to indicate which of the candidate voices they just heard. Results will be analyzed in light of current theories of vowel perception and normalization.

The effect of phonation into a straw on the vocal tract adjustments and formant frequencies. A preliminary MRI study on a single subject completed with acoustic results

Phonation into a straw is used in voice training and therapy. This study investigated its’ effects on the vocal tract adjustments. Magnetic resonance images were taken from one female phonating first on [a:], then into a plastic drinking straw and then again on [a:]. Acoustic samples were recorded from her and 10 other females. The samples were analyzed for fundamental frequency, sound pressure level (SPL), five lowest formant frequencies ( F 1– F 5) and SPL of the speaker's formant cluster range (2.5–4.5 kHz). During and after phonation into a straw, the midsagittal area of the vocal tract increased and the velar closure improved. The ratio of the transversal area of the lower pharynx over that of the epilarynx increased both during (27%) and after the straw (20%). The overall SPL and the SPL of the speaker's formant cluster increased. The frequencies of F 2, F 4 and F 5 decreased, while F 3 increased. The distances between F 4 and F 3 and F 5 and F 4 decreased by 49%. For the 10 subjects, F 3 increased, F 4 decreased and the distance between F 3 and F 4 decreased significantly. The results suggest that exercising with a straw helps establishing a speaker's formant cluster, which increases loudness and thus improves vocal economy.

Darwinius masillae is a strepsirrhine—a reply to Franzen et al. (2009)

The hyoid apparatus plays an integral role in swallowing, respiration, and vocalization in mammals. Most placental mammals have a rod-shaped basihyal connected to the basicranium via both soft tissues and a mobile bony chain—the anterior cornu—whereas anthropoid primates have broad, shield-like or even cup-shaped basihyals suspended from the basicranium by soft tissues only. How the unique anthropoid hyoid morphology evolved is unknown, and hyoid morphology of nonanthropoid primates is poorly documented. Here we use phylogenetic comparative methods and linear morphometrics to address knowledge gaps in hyoid evolution among primates and their euarchontan outgroups. We find that dermopterans have variable reduction of cornu elements. Cynocephalus volans are sexually dimorphic in hyoid morphology. Tupaia and all lemuroids except Daubentonia have a fully ossified anterior cornu connecting a rod-shaped basihyal to the basicranium; this is the ancestral mammalian pattern that is also characteristic of the last common ancestor of Primates. Haplorhines exhibit a reduced anterior cornu, and anthropoids underwent further increase in basihyal aspect ratio values and in relative basihyal volume. Convergent with haplorhines, lorisoid strepsirrhines independently evolved a broad basihyal and reduced anterior cornua. While a reduced anterior cornu is hypothesized to facilitate vocal tract lengthening and lower formant frequencies in some mammals, our results suggest vocalization adaptations alone are unlikely to drive the iterative reduction of anterior cornua within Primates. Our new data on euarchontan hyoid evolution provide an anatomical basis for further exploring the form-function relationships of the hyoid across different behaviors, including vocalization, chewing, and swallowing.

Low Frequency Groans Indicate Larger and More Dominant Fallow Deer (Dama dama) Males

BackgroundModels of honest advertisement predict that sexually selected calls should signal male quality. In most vertebrates, high quality males have larger body sizes that determine higher social status and in turn higher reproductive success. Previous research has emphasised the importance of vocal tract resonances or formant frequencies of calls as cues to body size in mammals. However, the role of the acoustic features of vocalisations as cues to other quality-related phenotypic characteristics of callers has rarely been investigated.Methodology/Principal FindingsWe examined whether the acoustic structure of fallow deer groans provides reliable information on the quality of the caller, by exploring the relationships between male quality (body size, dominance rank, and mating success) and the frequency components of calls (fundamental frequency, formant frequencies, and formant dispersion). We found that body size was not related to the fundamental frequency of groans, whereas larger males produced groans with lower formant frequencies and lower formant dispersion. Groans of high-ranking males were characterised by lower minimum fundamental frequencies and to a lesser extent, by lower formant dispersions. Dominance rank was the factor most strongly related to mating success, with higher-ranking males having higher mating success. The minimum fundamental frequency and the minimum formant dispersion were indirectly related to male mating success (through dominance rank).Conclusion/SignificanceOur study is the first to show that sexually selected vocalisations can signal social dominance in mammals other than primates, and reveals that independent acoustic components encode accurate information on different phenotypic aspects of male quality.

Human listeners attend to size information in domestic dog growls

The acoustic features of vocalizations have the potential to transmit information about the size of callers. Most acoustic studies have focused on intraspecific perceptual abilities, but here, the ability of humans to use growls to assess the size of adult domestic dogs was tested. In a first experiment, the formants of growls were shifted to create playback stimuli with different formant dispersions (Deltaf), simulating different vocal tract lengths within the natural range of variation. Mean fundamental frequency (F0) was left unchanged and treated as a covariate. In a second experiment, F0 was resynthesized and Deltaf was left unchanged. In both experiments Deltaf and F0 influenced how participants rated the size of stimuli. Lower formant and fundamental frequencies were rated as belonging to larger dogs. Crucially, when F0 was manipulated and Deltaf was natural, ratings were strongly correlated with the actual weight of the dogs, while when Deltaf was varied and F0 was natural, ratings were not related to the actual weight. Taken together, this suggests that participants relied more heavily on Deltaf, in accordance with the fact that formants are better predictors of body size than F0.

Fourth formant dip as a correlate of American English flaps

Unlike the lower formant frequencies, the fourth formant frequency (F4) is rarely an important acoustic correlate of speech. This study shows evidence of a depression in F4 for some tokens of the American English flap. F4 may be lowered during American English /r/ (Espy‐Wilson et al., 2000) and during retroflex sounds of some languages (e.g., Pima, as found by Avelino and Kim, 2002), but other cues are likely to be more prominent for such sounds. The current project uses a corpus of acoustic measurements of flapped English /t, d/ (e.g., pretty, order) and shows an inconsistent but large drop in F4 at the flap consonant. This occurs predominantly near an /r/, but the F4 drop is timed to the flap, not the /r/. In some tokens, this F4 drop occurs even if the flap is reduced to an approximant, leaving few other acoustic cues to its presence. The F4 drop likely reflects a sublingual cavity resonance caused by an interaction of the tongue positions for /r/ and for a flap. This study presents results on the distribution, frequency of occurrence, and magnitude of the F4 drop. Future research will investigate the perceptual role of the F4 drop.