Search

This study compares the use of electroglottograms (EGGs) and glottal area waveforms (GAWs) to study phonation in different vibratory states as produced by professionally trained singers. Six western classical tenors were asked to phonate pitch glides from modal to falsetto phonation, or from modal to their stage voice above the passaggio (SVaP). For each pitch glide the sample entropy (SampEn) of the EGG signal was calculated to detect the occurrence of phonatory instabilities and establish a "ground truth" for the performed phonation type. The cycles before the maximum SampEn were labeled as modal, and the cycles after the peak were labeled as either falsetto, or SVaP. Three automatic categorizations of vibratory state were performed using clustering: one based only on the EGG, one based on the GAW, and one based on their combination. The error rate (clustering vs ground truth) was, on average, lower than 10% for all of the three settings, revealing no special advantage of the GAW over EGG, and vice versa. Modal voice cycles exhibited a larger contact quotient, larger normalized derivative peak ratio, and lower rise time, compared to SVaP and falsetto. The GAW-based normalized maximum area declination rate was larger in SVaP compared to modal voice.

This study investigates the effect of vowel context (excerpted from speech versus sustained) on two voice quality measures: the cepstral peak prominence smoothed (CPPS) and sample entropy (SampEn). Thirty-one dysphonic subjects with different types of organic dysphonia and thirty-one controls read a phonetically balanced text and phonated sustained [a:] vowels in comfortable pitch and loudness. All the [a:] vowels of the read text were excerpted by automatic speech recognition and phonetic (forced) alignment. CPPS and SampEn were calculated for all excerpted vowels of each subject, forming one distribution of CPPS and SampEn values per subject. The sustained vowels were analyzed using a 41 ms window, forming another distribution of CPPS and SampEn values per subject. Two speech-language pathologists performed a perceptual evaluation of the dysphonic subjects’ voice quality from the recorded text. The power of discriminating the dysphonic group from the controls for SampEn and CPPS was assessed for the excerpted and sustained vowels with the Receiver-Operator Characteristic (ROC) analysis. The best discrimination in terms of Area Under Curve (AUC) for CPPS occurred using the mean of the excerpted vowel distributions (AUC=0.85) and for SampEn using the 95th percentile of the sustained vowel distributions (AUC=0.84). CPPS and SampEn were found to be negatively correlated, and the largest correlation was found between the corresponding 95th percentiles of their distributions (Pearson, r=−0.83, p < 10−3). A strong correlation was also found between the 95th percentile of SampEn distributions and the perceptual quality of breathiness (Pearson, r=0.83, p < 10−3). The results suggest that depending on the acoustic voice quality measure, sustained vowels can be more effective than excerpted vowels for detecting dysphonia. Additionally, when using CPPS or SampEn there is an advantage of using the measures’ distributions rather than their average values. QC 20180129

To determine if in singing there is an effect of lung volume on the electroglottographic waveform, and if so, how it varies over the voice range.Eight trained female singers sang the tune "Frère Jacques" in 18 conditions: three phonetic contexts, three dynamic levels, and high or low lung volume. Conditions were randomized and replicated.The audio and EGG signals were recorded in synchrony with signals tracking respiration and vertical larynx position. The first 10 Fourier descriptors of every EGG cycle were computed. These spectral data were clustered statistically, and the clusters were mapped by color into a voice range profile display, thus visualizing the EGG waveform changes under the influence of fIn most subjects, EGG waveforms varied considerably over the voice range. Within subjects, reproducibility was high, not only across the replications, but also across the phonetic contexts. The EGG waveforms were quite individual, as was the nature of the EGG shape variation across the range. EGG metrics were significantly correlated to changes in lung volume, in parts of the range of the song, and in most subjects. However, the effect sizes of the relative lung volume were generally much smaller than the effects of fMost subjects exhibited small, reproducible effects of the relative lung volume on the EGG waveform. Some hypothesized influences of tracheal pull were seen, mostly at the lowest SPLs. The effects were however highly variable, both across the moderately wide f

Recent advances in machine learning and time series analysis techniques have brought new perspectives to a great number of scientific fields. This thesis contributes applications of such techniques to voice analysis, in an attempt to extract information on the vibration of the vocal folds as such, as well as on the radiated acoustic signal. The data that was analyzed in this work are acoustic recordings, electroglottographic (EGG) signals and transnasal high- speed videoendoscopic images. The data analysis techniques are primarily based on clustering, i.e., grouping of data based on similarity, and sample entropy analysis, i.e., quantifying the degree of irregularity in a given signal. The experiments were conducted so as to provide data for different types of vibratory behaviors (or vibratory states) of the vocal folds. Clustering was used in order to categorize in an unsupervised fashion these different vi- bratory states, based solely on the electroglottographic signal, or the glottal area waveform, or both. Sample entropy was utilized as an indicator of in- stabilities, when subjects produced voiced sounds using irregular vibratory patterns, such as register breaks, intermittent diplophonia, and other types of irregularities. The prominent role of sound pressure level and fundamental frequency motivated further study of the relationship between them and the shape of the electroglottographic waveform. Graphical representations were created to visualize the relationship between different vibratory behaviors with fundamental frequency and sound pressure level. The EGG waveform shape was seen to depend strongly on sound pressure level and somewhat less on fundamental frequency. In very soft phonation, the almost sinusoidal waveform of the EGG suggests that studying the EGG using clusters may give a better representation compared to conventional time-domain metrics. The paradigm of the clustering was later applied in synchronous recordings of electroglottogram and glottal area waveforms in professional tenor singers. Different vibratory states were classified successfully using clustering, and the electroglottogram was seen to be as good as the glottal area waveform for such a classification task. The last part of this work concerns voices from subjects with organic dysphonia. A study was dedicated to investigate how vowel context (sustained versus excerpted from speech) can affect the power of quantitative acoustic measures to discriminate dysphonic subjects from controls. Two acoustic voice quality measures were used: the cepstral peak prominence (smoothed) and sample entropy. The cepstral peak prominence (smoothed) showed better discriminatory power with excerpted vowels, while sample entropy with sustained vowels. Additionally, it was found that sample entropy was strongly correlated with cepstral peak prominence (smoothed) and with the perceptual quality of breathiness. QC 20180126 Phonatory dynamics and states

There is a growing interest in Cepstral and Entropy analyses of voice samples for defining a vocal health indicator, due to their reliability in investigating both regular and irregular voice signals. The purpose of this study is to determine whether the Cepstral Peak Prominence Smoothed (CPPS) and Sample Entropy (SampEn) could differentiate dysphonic speakers from normal speakers in vowels excerpted from readings and to compare their discrimination power. Results are reported for 33 patients and 31 controls, who read a standardized phonetically balanced passage while wearing a head mounted microphone. Vowels were excerpted from recordings using Automatic Speech Recognition and, after obtaining a measure for each vowel, individual distributions and their descriptive statistics were considered for CPPS and SampEn. The Receiver Operating Curve analysis revealed that the mean of the distributions was the parameter with the highest discrimination power for both CPPS and SampEn. CPPS showed a higher diagnostic precision than SampEn, exhibiting an Area Under Curve (AUC) of 0.85 compared to 0.72. A negative correlation between the parameters was found (Spearman; = −0.61), with higher SampEn corresponding to lower CPPS. The automatic method used in this study could provide support to voice monitorings in clinic and during individual's daily activities.

Introduction Due to a lack of empirical data, the current understanding of the laryngeal mechanics in the passaggio regions (i.e., the fundamental frequency ranges where vocal registration events usually occur) of the female singing voice is still limited. Material and methods In this study the first and second passaggio regions of 10 professionally trained female classical soprano singers were analyzed. The sopranos performed pitch glides from A3 (ƒo = 220 Hz) to A4 (ƒo = 440 Hz) and from A4 (ƒo = 440 Hz) to A5 (ƒo = 880 Hz) on the vowel [iː]. Vocal fold vibration was assessed with trans-nasal high speed videoendoscopy at 20,000 fps, complemented by simultaneous electroglottographic (EGG) and acoustic recordings. Register breaks were perceptually rated by 12 voice experts. Voice stability was documented with the EGG-based sample entropy. Glottal opening and closing patterns during the passaggi were analyzed, supplemented with open quotient data extracted from the glottal area waveform. Results In both the first and the second passaggio, variations of vocal fold vibration patterns were found. Four distinct patterns emerged: smooth transitions with either increasing or decreasing durations of glottal closure, abrupt register transitions, and intermediate loss of vocal fold contact. Audible register transitions (in both the first and second passaggi) generally coincided with higher sample entropy values and higher open quotient variance through the respective passaggi. Conclusions Noteworthy vocal fold oscillatory registration events occur in both the first and the second passaggio even in professional sopranos. The respective transitions are hypothesized to be caused by either (a) a change of laryngeal biomechanical properties; or by (b) vocal tract resonance effects, constituting level 2 source-filter interactions.

The vocal folds can oscillate in several different ways, manifest to practitioners and clinicians as “registers” or “mechanisms,” of which the two most often considered are modal voice and falsetto voice. Here these will be taken as instances of different “vibratory states,” i.e., distinct quasi-stationary patterns of vibration of the vocal folds. State transitions are common in biomechanical nonlinear oscillators, and they are often abrupt and impossible to predict exactly. Therefore, vibratory states are a source of confounding variation, for instance when acquiring a voice range profile (VRP). In the quest for a state-based, non-invasive VRP, a semi-automatic method based on the short-term spectrum of the electroglottographic (EGG) signal was developed. The method identifies rapid vibratory state transitions, such as the modal-falsetto switch, and clusters the EGG data based on their similarities in the relative levels and phases of the lower frequency components. Productions of known modal and falsetto voice were accurately clustered by a Gaussian mixture model. When mapped into the VRP, this EGG-based clustering revealed connected regions of different vibratory sub-regimes in both modal and falsetto.

The human voice is a product of an intricate biophysical system. The complexity of this system enables a rich variety of possible sounds, but at the same time poses great challenges for quantitative voice analysis. For example, the vocal folds can vibrate in several different ways, leading to variations in the acoustic output. Because the vocal folds are relatively inaccessible, such variations are often difficult to account for. This work proposes a novel method for extracting non-invasively information on the vibratory state of the human vocal folds. Such information is important for creating a more complete voice analysis scheme. Invasive methods are undesirable because they often disturb the subjects and/or the studied phenomena, and they are also impractical in terms of accessibility and cost. A useful frame of reference for voice analysis is the Voice Range Profile (VRP). The 3 dimensional form of the VRP can be used to depict any phonatory metric over the 2 dimensional plane defined by the fundamental frequency of phonation (x-axis) and the sound pressure level (y-axis). The primary goal of this work was to incorporate information on the vibratory state of the vocal folds into the Voice Range Profile (e.g., as a color change). For this purpose, a novel method of analysis of the electroglottogram (EGG) was developed, using techniques from machine learning (clustering) and nonlinear time series analysis (sample entropy estimation). The analysis makes no prior assumptions on the nature of the EGG signal and does not rely on its absolute amplitude or frequency. Unlike time-domain methods, which typically define thresholds for quantifying EGG cycle metrics, the proposed method uses information from the entire cycle of each period. The analysis was applied in a variety of experimental conditions (constant vowel with different vibratory states, constant vibratory state and different vowels, constant vowel and vibratory state with varying lung volume) and the magnitude of effect on the EGG short-term spectrum was estimated for each of these conditions. It was found that the short-term spectrum of the EGG signal sufficed to discriminate between different phonatory configurations, such as modal and falsetto voice. It was found also that even supposedly purely articulatory changes could be traced in the spectrum of the EGG signal. Finally, possible pedagogical and clinical applications of the method are discussed. QC 20140609 FonaDyn

International audience; Although the electroglottographic (EGG) signal is not acoustic, one might expect that some of the source-induced variation in the spectrum slope of the airborne signal would be present also in the EGG signal. Wideband spectra of EGG signals were evaluated from sustained and intermittent phonation under various conditions, including changes of electrode position, vowel, subglottal pressure and SPL. Recordings were made of subjects producing /pV/ utterances with simultaneous acquisition of EGG and intraoral pressure. The EGG spectrum envelope was found to be quite linear in dB/octave, with the exception of the fundamental partial. The EGG spectrum effects of vowel changes were negligible. EGG spectrum slope change with SPL was large at phonation onset and small in loud phonation. Additionally, recordings from an existing database of 8 trained male singers were analysed for EGG spectrum variation with SPL. The singers performed crescendo tasks on sustained tones, with a typical SPL variation of up to 20 dB from soft to loud. The corresponding EGG spectra had slopes of -14 to ‑9 dB/octave. The variation in EGG spectrum slope was again small, on the order of one quarter of the slope variation in the airborne spectrum. Occasionally, ripple in the EGG spectrum envelope was present, due to double peaks in the time derivative of the closing part of the EGG waveform. We conclude that the EGG spectrum slope appears to offer a convenient contacting criterion, but will be harder to use for judging vocal effort beyond contacting.

The importance of the interaction between the acoustic impedance of the vocal tract with the flow across the vocal cords is well established. In this paper we are investigating the changes in vocal tract impedance when using the different modes of phonation according to Sadolin [1], going from the soft levels of the Neutral mode to the high levels of the fully ‘metallic’ Edge mode. The acoustic impedance of vocal tract as seen from the mouth opening is measured via a microphone placed close to the mouth when exciting the system with a volume velocity source [2]. At the same time a Laryngograph frontend is used to measure the electroglottograph signal which reflects the opening and closing pattern of the vocal folds. The measurements were carried out for all four modes (Neutral, Curbing, Overdrive and Edge) for the vowel [a] in three different pitches: C3(131 Hz), G3 (196 Hz) and C4 (262Hz) . The results show changes in the resonance frequencies of the vocal tract with increasing pitch, whereas the changes between the modes are less clear due to the measurement signal being weak in comparison to the louder modes, especially at high pitches. The electroglottograph shows a very different waveform for the Neutral mode compared to the other, so-called metallic modes. The differences in waveform between Curbing, Overdrive and Edge modes are minor. However, the spectrum of the Overdrive mode shows stronger 2nd harmonic and weaker 4th and 6th harmonic compared to Curbing and Edge. Finally the Overdrive mode, which is the mode that is most limited in pitch range, was tested at its pitch limit C5 (523 Hz) under normal conditions and when the singer has inhaled Helium. When inhaling Helium the acoustic impedance of the vocal tract is reduced in magnitude and the resonances are scaled upwards in frequency due to different density and speed of sound in Helium. The electroglottograph shows a change in waveform when the singer inhales helium. The percentage of the glottal cycle when the vocal cords are open, the so-called open quotient, increases from 40 to 55%. When inhaling helium the male singer was able reach Eb5, a minor third over the normal limit for males, this seems to indicate that the vocal tract impedance is at least partially responsible for the pitch limit in ‘Overdrive’.