Voice Pitch Research Articles

1A1-M08 人工喉頭を用いたSinger Robotの研究 : 発声の安定化

"Singing" is one of the most important and popular artistic human activities. The purpose of this research is to develop a mechanical Singer Robot with artificial larynx. We have already developed a singer robot that can generate Japanese 5 vowels [a], [i], [u], [e], and [o], and nasal plosive consonants. The robot has a mechanism to change the voice pitch by manipulating artificial vocal cord status. In this paper, a stable vocalization of vowel, improvements of the primary tone spectrum are presented.

A method of speech periodicity enhancement using transform-domain signal decomposition

Periodicity is an important property of speech signals. It is the basis of the signal’s fundamental frequency and the pitch of voice, which is crucial to speech communication. This paper presents a novel framework of periodicity enhancement for noisy speech. The enhancement is applied to the linear prediction residual of speech. The residual signal goes through a constant-pitch time warping process and two sequential lapped-frequency transforms, by which the periodic component is concentrated in certain transform coefficients. By emphasizing the respective transform coefficients, periodicity enhancement of noisy residual signal is achieved. The enhanced residual signal and estimated linear prediction filter parameters are used to synthesize the output speech. An adaptive algorithm is proposed for adjusting the weights for the periodic and aperiodic components. Effectiveness of the proposed approach is demonstrated via experimental evaluation. It is observed that harmonic structure of the original speech could be properly restored to improve the perceptual quality of enhanced speech.

Heroes and Villains: The Relationship between Pitch Tessitura and Sociability of Operatic Characters

ABSTRACT: Research in speech prosody and ethology suggests that pitch height indexes positive and negative social affects. For example, higher pitched voices are used to convey friendliness, whereas lower pitched voices are used to convey aggression (Bolinger, 1964). Research concerning animal calls suggests that this association generalizes to many species. In a study of the calls for 56 species, Morton (1977, 1994) proposed a sound-size model in which large size (and low pitch) is associated with aggression, whereas small size (and high pitch) is associated with friendliness, fear, or appeasement. We examine whether this association can be observed in music. Specifically, the results of three studies are reported in which the pitch-height of various voices is related to estimates of the sociability of the corresponding operatic characters. Results are consistent with lower-pitched voices being associated with less sociable characters, and higher-pitched voices being associated with more sociable characters. In addition, older male characters tend to exhibit lower-pitched voices, consistent with known physiological changes (Linville, 2004; Reubold, Harrington & Kleber, 2010).

Discrimination of voice gender in the human auditory cortex

Discerning a speaker's gender from their voice is a basic and crucial aspect of human communication. Voice pitch height, the perceptual correlate of fundamental frequency, is higher in females and provides a cue for gender discrimination. However, male and female voices are also differentiated by multiple other spectral and temporal characteristics, including mean formant frequency and spectral flux. The robust perceptual segregation of male and female voices is thought to result from processing the combination of discriminating features, which in neural terms may correspond to early sound object analysis occurring in non-primary auditory cortex. However, the specific mechanism for gender perception has been unclear. Here, using functional magnetic resonance imaging, we show that discrete sites in non-primary auditory cortex are differentially activated by male and female voices, with female voices consistently evoking greater activation in the upper bank of the superior temporal sulcus and posterior superior temporal plane. This finding was observed at the individual subject-level in all 24 subjects. The neural response was highly specific: no auditory regions were more activated by male than female voices. Further, the activation associated with female voices was 1) larger than can be accounted for by a sole effect of fundamental frequency, 2) not due to psychological attribution of female gender and 3) unaffected by listener gender. These results demonstrate that male and female voices are represented as distinct auditory objects in the human brain, with the mechanism for gender discrimination being a gender-dependent activation-level cue in non-primary auditory cortex.

Keeping Control: The Importance of Nonverbal Expressions of Power by Organizational Spokespersons in Times of Crisis

This paper examines the impact of nonverbal expressions of power by organizational spokespersons during different crisis stages. Study 1 investigates how vocal nonverbal cues express power during crises and how this affects perceptions of spokespersons. The results illustrate that a spokesperson who speaks with a lowered voice pitch, which expresses power, appears more competent than one with a raised voice pitch. Study 2 examines the moderating influence of crisis stage on the impact of visual nonverbal cues. During a crisis, powerful nonverbal behaviors minimize reputational damage through an increase in perceived competence of the spokesperson. In the aftermath of a crisis, powerless nonverbal behaviors positively affect the organizational reputation through an intermediate effect on perceived sincerity of the spokesperson.

On the Apparent Propagation Speed in Transmission Line Matrix Uniform Grid Meshes

Numerical models consisting of two-dimensional (2D) and three-dimensional (3D) uniform grid meshes for the transmission line matrix method (TLM) currently use 2 and 3, respectively, to compensate for the apparent sound speed. In this paper, new compensation factors are determined from a priori simulations, performed without compensation, in 2D and 3D TLM one-section tube models. The frequency values of the first mistuned resonance peaks, obtained from these simulations, are substituted in the corresponding equations for the resonance frequencies in one-section tubes to find the apparent sound propagation speed in the mesh environment and, thus, the necessary compensation. The new factors have been tested in more complex models like a two-tube concatenation model and a realistic magnetic resonance imaging (MRI)-reconstructed human vocal tract (VT) model. Important VT modeling results confirm the improvement over the conventional compensation factors, particularly for frequencies above 4 kHz. Among these results are the identification of the spectral trough at about 5200 Hz caused by the piriform fossa and the application of a pitch extraction algorithm to the 3D TLM output signal, finding a difference smaller than 0.66% relatively to human voice pitch.

Event related potentials study of aberrations in voice control mechanisms in adults with attention deficit hyperactivity disorder

ObjectiveThe present study was designed to test for neural signs of impulsivity related to voice motor control in young adults with ADHD using EEG recordings in a voice pitch perturbation paradigm. MethodsTwo age-matched groups of young adults were presented with brief pitch shifts of auditory feedback during vocalization. Compensatory behavioral and corresponding bioelectrical brain responses were elicited by the pitch-shifted voice feedback. ResultsThe analysis of bioelectrical responses showed that the ADHD group had shorter peak latency and onset time of motor-related bioelectrical brain responses as compared to the controls. ConclusionsThese results were interpreted to suggest differences in executive functions between ADHD and control participants. SignificanceWe hypothesize that more rapid motor-related bioelectrical responses found in the present study may be a manifestation of impulsiveness in adults with ADHD at the involuntary level of voice control.

Deficits in the pitch sensitivity of cochlear-implanted children speaking English or Mandarin.

Sensitivity to complex pitch is notoriously poor in adults with cochlear implants (CIs), but it is unclear whether this is true for children with CIs. Many are implanted today at a very young age, and factors related to brain plasticity (age at implantation, duration of CI experience, and speaking a tonal language) might have strong influences on pitch sensitivity. School-aged children participated, speaking English or Mandarin, having normal hearing (NH) or wearing a CI, using their clinically assigned settings with envelope-based coding strategies. Percent correct was measured in three-interval three-alternative forced choice tasks, for the discrimination of fundamental frequency (F0) of broadband harmonic complexes, and for the discrimination of sinusoidal amplitude modulation rate (AMR) of broadband noise, with reference frequencies at 100 and 200 Hz to focus on voice pitch processing. Data were fitted using a maximum-likelihood technique. CI children displayed higher thresholds and shallower slopes than NH children in F0 discrimination, regardless of linguistic background. Thresholds and slopes were more similar between NH and CI children in AMR discrimination. Once the effect of chronological age was extracted from the variance, the aforementioned factors related to brain plasticity did not contribute significantly to the CI children's sensitivity to pitch. Unless different strategies attempt to encode fine structure information, potential benefits of plasticity may be missed.

Contribution of auditory working memory to speech understanding in mandarin-speaking cochlear implant users.

PurposeTo investigate how auditory working memory relates to speech perception performance by Mandarin-speaking cochlear implant (CI) users.MethodAuditory working memory and speech perception was measured in Mandarin-speaking CI and normal-hearing (NH) participants. Working memory capacity was measured using forward digit span and backward digit span; working memory efficiency was measured using articulation rate. Speech perception was assessed with: (a) word-in-sentence recognition in quiet, (b) word-in-sentence recognition in speech-shaped steady noise at +5 dB signal-to-noise ratio, (c) Chinese disyllable recognition in quiet, (d) Chinese lexical tone recognition in quiet. Self-reported school rank was also collected regarding performance in schoolwork.ResultsThere was large inter-subject variability in auditory working memory and speech performance for CI participants. Working memory and speech performance were significantly poorer for CI than for NH participants. All three working memory measures were strongly correlated with each other for both CI and NH participants. Partial correlation analyses were performed on the CI data while controlling for demographic variables. Working memory efficiency was significantly correlated only with sentence recognition in quiet when working memory capacity was partialled out. Working memory capacity was correlated with disyllable recognition and school rank when efficiency was partialled out. There was no correlation between working memory and lexical tone recognition in the present CI participants.ConclusionsMandarin-speaking CI users experience significant deficits in auditory working memory and speech performance compared with NH listeners. The present data suggest that auditory working memory may contribute to CI users' difficulties in speech understanding. The present pattern of results with Mandarin-speaking CI users is consistent with previous auditory working memory studies with English-speaking CI users, suggesting that the lexical importance of voice pitch cues (albeit poorly coded by the CI) did not influence the relationship between working memory and speech perception.

Infant-Directed Visual Prosody: Mothers' Head Movements and Speech Acoustics.

Acoustical changes in the prosody of mothers' speech to infants are distinct and near universal. However, less is known about the visible properties mothers' infant-directed (ID) speech, and their relation to speech acoustics. Mothers' head movements were tracked as they interacted with their infants using ID speech, and compared to movements accompanying their adult-directed (AD) speech. Movement measures along three dimensions of head translation, and three axes of head rotation were calculated. Overall, more head movement was found for ID than AD speech, suggesting that mothers exaggerate their visual prosody in a manner analogous to the acoustical exaggerations in their speech. Regression analyses examined the relation between changing head position and changing acoustical pitch (F0 ) over time. Head movements and voice pitch were more strongly related in ID speech than in AD speech. When these relations were examined across time windows of different durations, stronger relations were observed for shorter time windows (< 5 sec). However, the particular form of these more local relations did not extend or generalize to longer time windows. This suggests that the multimodal correspondences in speech prosody are variable in form, and occur within limited time spans.

Coordinated plasticity in brainstem and auditory cortex contributes to enhanced categorical speech perception in musicians.

Musicianship is associated with neuroplastic changes in brainstem and cortical structures, as well as improved acuity for behaviorally relevant sounds including speech. However, further advance in the field depends on characterizing how neuroplastic changes in brainstem and cortical speech processing relate to one another and to speech-listening behaviors. Here, we show that subcortical and cortical neural plasticity interact to yield the linguistic advantages observed with musicianship. We compared brainstem and cortical neuroelectric responses elicited by a series of vowels that differed along a categorical speech continuum in amateur musicians and non-musicians. Musicians obtained steeper identification functions and classified speech sounds more rapidly than non-musicians. Behavioral advantages coincided with more robust and temporally coherent brainstem phase-locking to salient speech cues (voice pitch and formant information) coupled with increased amplitude in cortical-evoked responses, implying an overall enhancement in the nervous system's responsiveness to speech. Musicians' subcortical and cortical neural enhancements (but not behavioral measures) were correlated with their years of formal music training. Associations between multi-level neural responses were also stronger in musically trained listeners, and were better predictors of speech perception than in non-musicians. Results suggest that musicianship modulates speech representations at multiple tiers of the auditory pathway, and strengthens the correspondence of processing between subcortical and cortical areas to allow neural activity to carry more behaviorally relevant information. We infer that musicians have a refined hierarchy of internalized representations for auditory objects at both pre-attentive and attentive levels that supplies more faithful phonemic templates to decision mechanisms governing linguistic operations.

The effect of temperature on basal tension and thyroarytenoid muscle contraction in an isolated rat glottis model.

The pitch of voice is closely related to the vocal fold tension, which is the end result of coordinated movement of the intralaryngeal muscles, and especially the thyroarytenoid muscle. It is known that vocal quality may be affected by surrounding temperature; however, the effect of temperature on vocal fold tension is mostly unknown. Thus, the aim of this study was to evaluate the effect of temperature on isolated rat glottis and thyroarytenoid muscle contraction induced by electrical field stimulation. In vitro isometric tension of the glottis ring from 30 Sprague-Dawley rats was continuously recorded by the tissue bath method. Electrical field stimulation was applied to the glottis ring with two wire electrodes placed parallel to the glottis and connected to a direct-current stimulator. The tension changes of the rat glottis rings that were either untreated or treated with electrical field stimulation were recorded continuously at temperatures from 37 to 7 °C or from 7 to 37 °C. Warming from 7 to 37 °C increased the basal tension of the glottis rings and decreased the electrical field stimulation-induced glottis ring contraction, which was chiefly due to thyroarytenoid muscle contraction. In comparison, cooling from 37 to 7 °C decreased the basal tension and enhanced glottis ring contraction by electrical field stimulation. We concluded that warming increased the basal tension of the glottis in vitro and decreased the amplitude of electrical field stimulation-induced thyroarytenoid muscle contraction. Thus, vocal pitch and the fine tuning of vocal fold tension might be affected by temperature in vivo.

Social dialect and men's voice pitch influence women's mate preferences

Low male voice pitch may communicate potential benefits for offspring in the form of heritable health and/or dominance, whereas access to resources may be indicated by correlates of socioeconomic status, such as sociolinguistic features. Here, we examine if voice pitch and social dialect influence women's perceptions of men's socioeconomic status and attractiveness. In Study 1, women perceived lower pitched male voices as higher in socioeconomic status than higher pitched male voices. In Study 2, women independently perceived lower pitched voices and higher status sociolinguistic dialects as higher in socioeconomic status and attractiveness. We also found a significant interaction wherein women preferred lower pitched men's voices more often when dialects were lower in sociolinguistic status than when they were higher in sociolinguistic status. Women also perceived lower pitched voices as higher in socioeconomic status more often when dialects were higher in sociolinguistic status than when lower in sociolinguistic status. Finally, women's own self-rated socioeconomic status was positively related to their preferences for voices with higher status sociolinguistic dialects, but not to their preferences for voice pitch. Hence, women's preferences for traits associated with potentially biologically heritable benefits, such as low voice pitch, are moderated by the presence of traits associated with resource accrual, such as social dialect markers. However, women's preferences for language markers of resource accrual may be functionally independent from preferences for potential biological indicators of heritable benefits, such as voice pitch.

Cortical pitch response components index stimulus onset/offset and dynamic features of pitch contours

Voice pitch is an important information-bearing component of language that is subject to experience dependent plasticity at both early cortical and subcortical stages of processing. We have already demonstrated that pitch onset component (Na) of the cortical pitch response (CPR) is sensitive to flat pitch and its salience. In regards to dynamic pitch, we do not yet know whether the multiple pitch-related transient components of the CPR reflect specific temporal attributes of such stimuli. Here we examine the sensitivity of the multiple transient components of CPR to changes in pitch acceleration associated with the Mandarin high rising lexical tone. CPR responses from Chinese listeners were elicited by three citation forms varying in pitch acceleration and duration. Results showed that the pitch onset component (Na) was invariant to changes in acceleration. In contrast, Na–Pb and Pb–Nb showed a systematic increase in the interpeak latency and decrease in amplitude with increase in pitch acceleration that followed the time course of pitch change across the three stimuli. A strong correlation with pitch acceleration was observed for these two components only – a putative index of pitch-relevant neural activity associated with the more rapidly-changing portions of the pitch contour. Pc–Nc marks unambiguously the stimulus offset. We therefore propose that in the early stages of cortical sensory processing, a series of neural markers flag different temporal attributes of a dynamic pitch contour: onset of temporal regularity (Na); changes in temporal regularity between onset and offset (Na–Pb, Pb–Nb); and offset of temporal regularity (Pc–Nc). At the temporal electrode sites, the stimulus with the most gradual change in pitch acceleration evoked a rightward asymmetry. Yet within the left hemisphere, stimuli with more gradual change were indistinguishable. These findings highlight the emergence of early hemispheric preferences and their functional roles as related to sensory and cognitive properties of the stimulus.

Unfamiliar voice identification: Effect of post-event information on accuracy and voice ratings

This study addressed the effect of misleading post-event information (PEI) on voice ratings, identification accuracy, and confidence, as well as the link between verbal recall and accuracy. Participants listened to a dialogue between male and female targets, then read misleading information about voice pitch. Participants engaged in verbal recall, rated voices on a feature checklist, and made a lineup decision. Accuracy rates were low, especially on target-absent lineups. Confidence and accuracy were unrelated, but the number of facts recalled about the voice predicted later lineup accuracy. There was a main effect of misinformation on ratings of target voice pitch, but there was no effect on identification accuracy or confidence ratings. As voice lineup evidence from earwitnesses is used in courts, the findings have potential applied relevance.

Pre-babbling infants prefer listening to infant speech: Implications for vocal learning in humans

For human infants to engage in vocal learning, they must effectively monitor and assess their own self-produced speech, which entails perceiving speech produced by an infant. Yet, little is known about how infants respond to infant-produced speech. Here, we demonstrate that pre-babbling infants prefer listening to infant speech. Across four experiments, 3-to-6-month-olds were tested in a preferential listening procedure, using vowels synthesized to emulate productions by female adults and infants. In experiment 1, infants listened longer to vowels produced by infant than adult speakers. However, in experiment 2, infants failed to show any listening preference for infant versus adult vowels synthesized with matching, infant-appropriate pitch values, suggesting that infants were either attracted to higher voice pitch per se or to infant-like voice pitch. Failing to support a bias of the first type, infants in experiment 3 showed no listening preference when presented infant vowels with higher and lower infant-appropriate pitch values. Moreover, in experiment 4, infants showed a preference for infant versus adult vowels when synthesized with pitch values that are appropriate for a female using adult-directed speech; this suggests that infants are also attracted to infant vocal resonance properties. The implications of these results for speech development are discussed.

The role of pitch strength in extracting speech from complex backgrounds

How do people extract a target speech signal from a chorus of other sounds including other speech sounds? And what are the central and peripheral auditory processes that make this possible, and that may fail in people with damaged auditory systems? These critical issues in speech and hearing science were of continuing interest to Sid Bacon throughout his career. As Sid has noted, the problem is multifaceted, and he chose to study many of the individual factors, as well as their interactions. In keeping with that approach, I will discuss how normal-hearing and hearing-impaired listeners use pitch to extract a target from a complex noise background. In one study, auditory stream segregation of iterated rippled noises (IRN) with varying pitch strengths was explored to understand the limits of tonality for separating two patterns of sounds. In a second study, voice pitch strength was investigated as a means to support perceptual separation of target and background speech, with a focus on either spectral or temporal characteristics of the speech sounds. Interactions between degree of tonality in speech and other factors related to perception of speech in background sound will be examined. [Work supported by NIH.]

The neural changes in connectivity of the voice network during voice pitch perturbation

Voice control is critical to communication. To date, studies have used behavioral, electrophysiological and functional data to investigate the neural correlates of voice control using perturbation tasks, but have yet to examine the interactions of these neural regions. The goal of this study was to use structural equation modeling of functional neuroimaging data to examine network properties of voice with and without perturbation. Results showed that the presence of a pitch shift, which was processed as an error in vocalization, altered connections between right STG and left STG. Other regions that revealed differences in connectivity during error detection and correction included bilateral inferior frontal gyrus, and the primary and pre motor cortices. Results indicated that STG plays a critical role in voice control, specifically, during error detection and correction. Additionally, pitch perturbation elicits changes in the voice network that suggest the right hemisphere is critical to pitch modulation.

Speech Enhancement for Listeners With Hearing Loss Based on a Model for Vowel Coding in the Auditory Midbrain

A novel signal-processing strategy is proposed to enhance speech for listeners with hearing loss. The strategy focuses on improving vowel perception based on a recent hypothesis for vowel coding in the auditory system. Traditionally, studies of neural vowel encoding have focused on the representation of formants (peaks in vowel spectra) in the discharge patterns of the population of auditory-nerve (AN) fibers. A recent hypothesis focuses instead on vowel encoding in the auditory midbrain, and suggests a robust representation of formants. AN fiber discharge rates are characterized by pitch-related fluctuations having frequency-dependent modulation depths. Fibers tuned to frequencies near formants exhibit weaker pitch-related fluctuations than those tuned to frequencies between formants. Many auditory midbrain neurons show tuning to amplitude modulation frequency in addition to audio frequency. According to the auditory midbrain vowel encoding hypothesis, the response map of a population of midbrain neurons tuned to modulations near voice pitch exhibits minima near formant frequencies, due to the lack of strong pitch-related fluctuations at their inputs. This representation is robust over the range of noise conditions in which speech intelligibility is also robust for normal-hearing listeners. Based on this hypothesis, a vowel-enhancement strategy has been proposed that aims to restore vowel encoding at the level of the auditory midbrain. The signal processing consists of pitch tracking, formant tracking, and formant enhancement. The novel formant-tracking method proposed here estimates the first two formant frequencies by modeling characteristics of the auditory periphery, such as saturated discharge rates of AN fibers and modulation tuning properties of auditory midbrain neurons. The formant enhancement stage aims to restore the representation of formants at the level of the midbrain by increasing the dominance of a single harmonic near each formant and saturating that frequency channel. A MATLAB implementation of the system with low computational complexity was developed. Objective tests of the formant-tracking subsystem on vowels suggest that the method generalizes well over a wide range of speakers and vowels.

The influence of vocal training and acting experience on measures of voice quality and emotional genuineness

Vocal training through singing and acting lessons is known to modify acoustic parameters of the voice. While the effects of singing training have been well documented, the role of acting experience on the singing voice remains unclear. In two experiments, we used linear mixed models to examine the relationships between the relative amounts of acting and singing experience on the acoustics and perception of the male singing voice. In Experiment 1, 12 male vocalists were recorded while singing with five different emotions, each with two intensities. Acoustic measures of pitch accuracy, jitter, and harmonics-to-noise ratio (HNR) were examined. Decreased pitch accuracy and increased jitter, indicative of a lower “voice quality,” were associated with more years of acting experience, while increased pitch accuracy was associated with more years of singing lessons. We hypothesized that the acoustic deviations exhibited by more experienced actors was an intentional technique to increase the genuineness or truthfulness of their emotional expressions. In Experiment 2, listeners rated vocalists’ emotional genuineness. Vocalists with more years of acting experience were rated as more genuine than vocalists with less acting experience. No relationship was reported for singing training. Increased genuineness was associated with decreased pitch accuracy, increased jitter, and a higher HNR. These effects may represent a shifting of priorities by male vocalists with acting experience to emphasize emotional genuineness over pitch accuracy or voice quality in their singing performances.