Voice Perturbation Measures Research Articles

Acoustic characterization and machine prediction of perceived masculinity and femininity in adults

Previous research has found that human voice can provide reliable information to be used for gender identification with a high level of accuracy. In social psychology, perceived masculinity and femininity (masculinity and femininity rated by humans) has often been considered an important feature when investigating the influence of vocal features on social behaviours. While previous studies have characterized the acoustic features that contributed to perceivers’ judgements of speakers’ masculinity or femininity, there is limited research on developing a machine masculinity/femininity scoring model and characterizing the independent acoustic factors that contribute to perceivers’ masculinity and femininity judgements. In this work, we first propose a machine scoring model of perceived masculinity/femininity based on the Extreme Random Forest and then characterize the independent and meaningful acoustic factors that contribute to perceivers’ judgements by using a correlation matrix based hierarchical clustering method. Our results show that the machine ratings of masculinity and femininity strongly correlated with the human ratings of masculinity and femininity when we used an optimal speech duration of 7 s, with a correlation coefficient of up to .63 for females and .77 for males. Nine independent clusters of acoustic measures were generated from our modelling of femininity judgements for female voices and eight clusters were found for masculinity judgements for male voices. The results revealed that, for both genders, the F0 mean is the most important acoustic measure affecting the judgement of acoustic-related masculinity and femininity. The F3 mean, F4 mean and VTL estimators were found to be highly inter-correlated and appeared in the same cluster, forming the second most significant factor in influencing the assessment of acoustic-related masculinity and femininity. Next, F1 mean, F2 mean and F0 standard deviation are independent factors that share similar importance. The voice perturbation measures, including HNR, jitter and shimmer, are of lesser importance in influencing masculinity/femininity judgements.

Does Self-Reported Childhood Trauma Relate to Vocal Acoustic Measures? Preliminary Findings at Trauma Recall

Nonverbal markers of childhood trauma have been identified in the literature. Yet, the relationship between childhood trauma and the voice is still largely unexplored. We report preliminary findings from a study (N = 48) that investigated whether self-reported childhood trauma is related to acoustic measures on sustained phonation before and after childhood trauma recall. Childhood trauma and minimization/denial of trauma were measured using the Childhood Trauma Questionnaire. Measures of voice (fundamental frequency, related to pitch, and perturbation, or “unsteadiness” in voice sustained phonation) were obtained using voice analysis software lingWAVES. Self-reported trait and state anxiety were also measured, using the State and Trait Anxiety Inventory. Results showed that voice fundamental frequency at baseline was related to self-reported minimization of childhood trauma. Childhood trauma and minimization of childhood trauma were significantly related to voice perturbation measures recorded after trauma recall, accounting for baseline voice measures and self-reported anxiety. These preliminary results indicate the importance of studying voice as a potential nonverbal marker of childhood trauma.

Feasibility of Galaxy Smartphone Recording as Portable Recorder for Acoustic Analysis of Voice

Background and Objectives Acoustic analysis of voice could be influenced so much by the quality of voice files which were recorded by recording device. In clinical practice, voice files that were recorded by analysis program directly or portable digital recording device were analyzed mostly. This study examined the feasibility of using Galaxy smartphone recordings for acoustic analysis of voice. Materials and Methods Acoustic measures were compared between voice signals recorded from 30 normal speakers (15 males and 15 females) through Galaxy smartphone, portable digital recording device and CSL. Fo, jitter, shimmer, NHR (Noise-Harmony ratio) and Formant frequencies were analyzed by MDVP. Results Fo, Jitter, Shimmer, NHR and formant frequencies from 3 devices were no significantly difference. The intraclass correlation coefficient (ICC) was higher between each of the voice perturbation measures. Conclusion The findings indicated that Galaxy smartphone recording system was useful device for acoustic analysis of voice. Furthermore, Galaxy smartphone can be applied widely in various way for acoustic analysis of voice. Key words: Galaxy smartphone ㆍ Recording devices ㆍ Acoustic analysis 핵심용어: 갤럭시 스마트폰 ㆍ ë ¹ìŒìž¥ë¹„ ㆍ 음향분석

Evaluating iPhone Recordings for Acoustic Voice Assessment

Aims: This study examined the viability of using iPhone recordings for acoustic measurements of voice quality. Methods: Acoustic measures were compared between voice signals simultaneously recorded from 11 normal speakers (6 females and 5 males) through an iPhone (model A1303, Apple, USA) and a comparison recording system. Comparisons were also conducted between the pre- and post-operative voices recorded from 10 voice patients (4 females and 6 males) through the iPhone. Participants aged between 27 and 79 years. Results: Measures from iPhone and comparison signals were found to be highly correlated. Findings of the effects of vowel type on the selected measures were consistent between the two recording systems and congruent with previous findings. Analysis of the patient data revealed that a selection of acoustic measures, such as vowel space area and voice perturbation measures, consistently demonstrated a positive change following phonosurgery. Conclusion: The present findings indicated that the iPhone device tested was useful for tracking voice changes for clinical management. Preliminary findings regarding factors such as gender and type of pathology suggest that intra-subject, instead of norm-referenced, comparisons of acoustic measures would be more useful in monitoring the progression of a voice disorder or tracking the treatment effect.

Physical simulation of laryngeal disorders using a multiple-mass vocal fold model

Abstract This paper reports on the use of a high-dimensional discrete vocal fold model for the simulation of voice production under the presence of laryngeal disorders. Specifically, the effect of increases in mass and stiffness, both unilateral and bilateral, has been analysed independently for both magnitudes. The glottal flow waveform and the mass displacement have been studied and the obtained results are coherent with clinical observations that relate mass increments with lowered fundamental frequencies and mass and stiffness increments with reduced vibratory amplitudes of vocal folds. The reported results also indicate that asymmetries in the physical properties of vocal folds result in asymmetries in their vibratory patterns, including phase, amplitude and behaviour on collision. These are also correlated with voice perturbation measures such as jitter, shimmer and normalised noise energy.

Comparison of results in two acoustic analysis programs: Praat and MDVP

To compare acoustic analysis results obtained by 2 computer programs, Praat and the Multi-Dimensional Voice Program (MDVP). Different voice analysis programs use similar descriptions to define voice perturbation measures. Materials and methods: A total of 47 voice samples reflecting a spectrum of normal and pathological voices were randomly selected from a database, and the same voice samples were used to obtain mean fundamental frequency, jitter, shimmer, and noise-to-harmonics ratio results from 2 acoustic analysis programs. Results: The results obtained for mean fundamental frequency and shimmer were not significantly different between the 2 computer programs. The results for jitter and noise-to-harmonics ratio, however, were significantly different between Praat and MDVP (P < 0.001). There was a strong correlation for mean fundamental frequency and jitter values. The correlations for shimmer values and the noise-to-harmonics ratio were moderate. Conclusion: The numerical values obtained for mean fundamental frequency were comparable between the 2 computer programs. The values obtained for shimmer were not significantly different, but the correlation was moderate. The jitter values and noise-to-harmonics ratio were not comparable between the 2 acoustic analysis programs.

Clinical value of acoustic voice measures: a retrospective study

Within this study a retrospective analysis of clinical voice perturbation measures, Dysphonia Severity Index and subjective perceived hoarseness was performed to determine their value under clinical aspects. The study included the data of 580 healthy and 1,700 pathologic voices, which were investigated under the following aspects. The relevant parameters were identified and their interrelation determined. Group differences between healthy and pathologic voices were figured out and investigated if voice quality measures allowed an automatic diagnosis of voice disorders. The analysis revealed significant changes between the clinical groups, which indicate the diagnostic relevance of voice quality measures. However, an individual diagnosis of the underlying voice disorder failed due to a vast spread of the parameter values within the respective groups. Classification accuracies of 75-90% were achieved. The high misclassification rate of up to 25% implied that in voice disorder diagnosis, the individual interpretation of the parameter values has to be done carefully.

The effects of vowels on voice perturbation measures

This study examines voice perturbation parameters of the sustained [ɑ] in English and of the eight vowels in Turkish to discover whether any difference exists between these languages, and whether a correlation exists between voice perturbation parameters and articulatory and acoustic properties of the Turkish vowels. Eight Turkish vowels uttered by 26 healthy nonsmoker volunteer males who are native Turkish speakers were compared with a voice database that includes samples of normal and disordered voices belonging to American English speakers. Fundamental frequencies, the first and second formants, and perturbation parameters, such as jitter percent, pitch perturbation quotient, shimmer percent, and amplitude perturbation quotient of the sustained vowels, were measured. Also, the first and second formants of the sustained [ɑ] in English were measured, and other parameters have been obtained from the database. When the voice perturbation parameters in Turkish and English were compared, statistically significant differences were not found. However, when Turkish vowels compared with each other, statistically significant differences were found among perturbation values. Categorical comparisons of the Turkish vowels like high-low, rounded-unrounded, and front-back revealed significant differences in perturbation values. In correlation analysis, a weak linear inverse relation between jitter percent and the first formant ( r = −0.260, p<0.05) was found.

Effects of head extension and tongue protrusion on voice perturbation measures

Head extension with protruded tongue is the position for videolaryngoscopy and simultaneous glottographic recordings including photoglottographic signals. This study investigated the effect of head extension and tongue protrusion on the measures of fundamental frequency, frequency perturbation (jitter), and amplitude perturbation (shimmer). Acoustic signals recorded during sustained vowels were obtained from 49 women and 66 men with no speech or voice disorders in different head-tongue positions. Head extension was associated with increased fundamental frequency and decreased shimmer. In men, head extension did not appear to affect jitter. When the tongue was protruded, head extension tended to lower jitter. For both genders, tongue protrusion was associated with decreased fundamental frequency with head extension. In the men, tongue protrusion tended to increase shimmer when the head was in the neutral position. In the women, tongue protrusion was associated with increased jitter and increased shimmer and was most evident in the head-neutral position. These findings supported a physical linkage hypothesis of the relationship between vocal tract configuration and vocal fold vibration, suggesting that head-tongue position must be taken into account when comparing voice measures.

Comparison of Fo extraction methods for high-precision voice perturbation measurements.

Voice perturbation measures, such as jitter and shimmer, depend on accurate extraction of fundamental frequency (Fo) and amplitude of various waveform types. The extraction method directly affects the accuracy of the measures, particularly if several waveform types (with or without formant structure) are under consideration and if noise and modulation are present in the signal. For frequency perturbation, high precision is defined here as the ability to extract Fo to +/- 0.01% under conditions of noise and modulation. Three Fo-extraction methods and their software implementations are discussed and compared. The methods are cycle-to-cycle waveform matching, zero-crossing and peak-picking. Interpolation between samples is added to make the extractions more accurate and reliable. The sensitivity of the methods to different parameters such as sampling frequency, mean Fo, signal-to-noise ratio, frequency modulation, and amplitude modulation are explored.

Effect of Microphone Type and Placement on Voice Perturbation Measurements

This study was conducted to explore the effects of microphone type (dynamic vs. condenser) and pattern (omnidirectional vs. cardioid) on the extraction of voice perturbation measures for sustained phonation. Also of interest were the effects of distance and angle between the source and the microphone. Four professional-grade and two consumer-grade microphones were selected for analysis. Synthesized phonation with different amplitude and frequency modulations at fundamental frequencies of 100 Hz and 300 Hz were presented over a loudspeaker. Human phonation was also included to test the validity of loudspeaker presentations. Three microphone distances (4 cm, 30 cm, 1 m) and three angles (0 degree, 45 degrees, 90 degrees) were used for microphone placement. Among the professional grade microphones, the cardioid condenser type had the smallest effect on perturbation measures. In general, condenser types gave better results than dynamic types. Microphones with an unbalanced output did not perform as well as those with balanced outputs. Microphone sensitivity and distance had the largest effect on perturbation measures, making it difficult to resolve normal vocal jitter at anything but a few centimeters from the mouth. Angle had little effect for short distances, but a greater effect for longer distances. These conclusions are preliminary because the sampling of microphones, distances, and signal types was very coarse. The study serves only to chart the course for future work.

Separation of trends and subharmonic structures from random variations in fundamental frequency and amplitude.

An implicit assumption in voice perturbation analysis is that one is dealing with small random variations in fundamental frequency and amplitude. However, many irregularities in voiced speech are not necessarily small or random (e.g., subharmonics, amplitude modulations, frequency modulations, linear trends). In this study, various processing schemes are employed to detect/remove trends and subharmonics from fundamental frequency and amplitude contours. Voice perturbation measures are calculated before and after application of the detection/removal techniques. Separating trends and subharmonics from random perturbations in voice analysis may prove useful in classifying and identifying voice disorders. Results of analysis on a variety of subjects will be presented.

Comparison of Acoustic Voice Perturbation Measures Among Three Independent Voice Laboratories

The purpose of this study was to compare jitter and shimmer data measured with three different analysis systems, the Visi-Pitch PC system (Pine Brook) and two systems based on minicomputers (Chicago and Denver), as a preliminary step toward establishing recording and analysis standards. The results show that, although similar hardware and software used at independent laboratories can yield similar findings, differences in recording hardware as well as recording and analysis procedures can result in important differences in perturbation findings. Jitter measurements obtained with the Visi-Pitch were not consistently in good agreement with jitter measurements obtained from the minicomputer systems due, in part, to an interaction between the Visi-Pitch internal filter selected during the recording process and the novel method of pitch period determination used in the Visi-Pitch. Magnitude of shimmer measurements differed between the two minicomputer systems, in part because of differences in amplitude resolution of the A/D converters and recording noise. The correlation between the two shimmer data sets was relatively high, however, indicating that relative changes across utterances were comparable in spite of magnitude differences.

Unification of perturbation measures in speech signals

Voice perturbation measures, formerly defined in a somewhat ad hoc fashion, are discussed within the framework of signal theory. An attempt is made to unify a variety of existing jitter, shimmer, and noise measures on the basis of common underlying perturbation functions and their derivatives. Some simple modulations (sinusoid, Gaussian noise, linear trend) are imposed on cyclic parameters in phonation (e.g., amplitude and fundamental frequency) to test the ability of perturbation measures to detect or reject these types of modulations. It is expected that this systematic approach to perturbation analysis will be helpful in identifying the sources of irregularity in the voice and, thereby, in the detection of laryngeal disorders.