Refining algorithmic estimation of relative fundamental frequency: Accounting for sample characteristics and fundamental frequency estimation method.

Relative fundamental frequency (RFF) is a promising acoustic measure for evaluating voice disorders. Yet, the accuracy of the current RFF algorithm varies across a broad range of vocal signals. The authors investigated how fundamental frequency (fo) estimation and sample characteristics impact the relationship between manual and semi-automated RFF estimates. Acoustic recordings were collected from 227 individuals with and 256 individuals without voice disorders. Common fo estimation techniques were compared to the autocorrelation method currently implemented in the RFF algorithm. Pitch strength-based categories were constructed using a training set (1158 samples), and algorithm thresholds were tuned to each category. RFF was then computed on an independent test set (291 samples) using category-specific thresholds and compared against manual RFF via mean bias error (MBE) and root-mean-square error (RMSE). Auditory-SWIPE' for fo estimation led to the greatest correspondence with manual RFF and was implemented in concert with category-specific thresholds. Refining fo estimation and accounting for sample characteristics led to increased correspondence with manual RFF [MBE = 0.01 semitones (ST), RMSE = 0.28 ST] compared to the unmodified algorithm (MBE = 0.90 ST, RMSE = 0.34 ST), reducing the MBE and RMSE of semi-automated RFF estimates by 88.4% and 17.3%, respectively.

Muscle Tension Dysphagia: An Expanded Investigation of Clinical Presentations and Swallowing Kinematics.

PURPOSE: This study retrospectively examined patient-reported symptoms, quality of life, and swallowing kinematics in individuals with presumed muscle tension dysphagia (MTDg). METHOD: Twenty-six individuals met the inclusion criteria. Data were gathered from patient-reported outcome measures (PROs), symptomology, clinician reports of palpation, and hyolaryngeal and hyoid movements measured on a 20-ml thin liquid bolus during videofluoroscopic swallowing studies. RESULTS: All PROs were outside of typical limits, except for the Voice Handicap Index-10. Mean hyoid excursion was 1.52 cm (SD = 0.46, range: 0.76-2.43), and hyolaryngeal excursion was 0.77 cm (SD = 0.44, range: -0.42-1.68). A minority of participants (4%-19%) demonstrated atypical hyoid and/or hyolaryngeal excursion compared to the available normative reference value sets. CONCLUSIONS: Individuals demonstrated abnormalities in the clinical evaluation of the areas of palpation and reported perilaryngeal discomfort and symptoms of laryngeal hyperresponsiveness, with a negative impact on their quality of life across various PROs. Atypical hyoid and/or hyolaryngeal excursion during swallowing was rare when compared to available normative reference values. The clinical evaluation of MTDg may be enhanced by including components related to muscle tension and laryngeal hyperresponsiveness in order to differentiate MTDg from idiopathic functional dysphagia and lead the patient to the otolaryngology/speech-language pathology clinic for intervention and management.

Normative Values of Cepstral Peak Prominence Measures in Typical Speakers by Sex, Speech Stimuli, and Software Type Across the Life Span.

PURPOSE: The purpose of this study was to determine normative values for cepstral peak prominence measures across the life span as a function of sex using clinically relevant stimuli (/ɑ/, /i/, and two sentences of The Rainbow Passage) and two commonly used software types: Praat (Version 6.0.50) and Analysis of Dysphonia in Speech and Voice (ADSV). METHOD: One hundred fifty speakers (75 males, 75 females; evenly distributed into three age groups) without voice disorders aged 18-91 years were recorded via headset microphone in a sound-treated booth. Cepstral measures were analyzed using common analysis methods in Praat and ADSV by sex, stimuli, and software type. Kruskal-Wallis tests and post hoc Mood's Median tests for significant factors were performed on cepstral measures to assess the effects of age group, sex, stimuli, and software type. RESULTS: The results revealed statistically significant effects of sex, stimuli, and software type on cepstral measures, but no statistical effect of age group on cepstral values. Females had lower average cepstral values compared to males. Across stimuli, the highest average cepstral measure was found for sustained /ɑ/, followed by sustained /i/, and then of the two sentences of The Rainbow Passage. Average cepstral measures in Praat were higher than those from ADSV. CONCLUSIONS: The current work did not find a statistical effect of age group on cepstral values; thus, normative cepstral values were reported by sex, stimuli, and software type. Future work should examine the applicability of these normative values for discriminating speakers with and without voice disorders.

Effects of Sidetone Amplification on Vocal Function During Telecommunication.

PURPOSE: Society has become increasingly dependent on telecommunication, which has been shown to negatively impact vocal function. This study explores the use of sidetone regulation during audio-visual communication as one potential technique to alleviate the effects of telecommunication on the voice. METHOD: The speech acoustics of 18 participants with typical voices were measured during conversational tasks during three conditions of sidetone amplification: baseline (no sidetone amplification), low sidetone amplification, and high sidetone amplification. Vocal intensity, vocal quality (estimated using acoustic measures of the low-high ratio and the smoothed cepstral peak prominence), and self-perceived vocal effort were used to measure the impacts of sidetone amplification on vocal function. RESULTS: Compared to baseline, there were statistically significant decreases in vocal intensity and increases in low-high ratio in the high level of sidetone amplification condition. Changes in these measures were not significantly correlated. When asked to rank conditions based on their perceived vocal effort, participants most often ranked the high level of sidetone amplification as least effortful; however, the visual-analog ratings of vocal effort were not significantly different between conditions. The smoothed cepstral peak prominence did not change with varying levels of sidetone amplification. CONCLUSIONS: Vocal intensity decreased with high levels of sidetone amplification. High levels of sidetone amplification also resulted in increases in the low-high ratio, which were shown to be more than just a byproduct of decreased vocal intensity. The impact of sidetone amplification on vocal effort was less clear, but results suggested that participants generally decreased their vocal effort with increased levels of sidetone amplification. This was a preliminary study and future work is warranted in a population of participants with voice complaints and in a more noisy, realistic environments.

Automated Creak Differentiates Adductor Laryngeal Dystonia and Muscle Tension Dysphonia.

OBJECTIVE: The purpose of this study was to determine whether automated estimates of vocal creak would differentiate speakers with adductor laryngeal dystonia (AdLD) from speakers with muscle tension dysphonia (MTD) and speakers without voice disorders. METHODS: Sixteen speakers with AdLD, sixteen speakers with MTD, and sixteen speakers without voice disorders were recorded in a quiet environment reading aloud a standard paragraph. An open-source creak detector was used to calculate the percentage of creak (% creak) in each of the speaker's six recorded sentences. RESULTS: A Kruskal-Wallis one-way analysis of variance revealed a statistically significant effect of group on the % creak with a large effect size. Pairwise Wilcoxon tests revealed a statistically significant difference in % creak between speakers with AdLD and controls as well as between speakers with AdLD and MTD. Receiver operating characteristic curve analyses indicated that % creak differentiated AdLD from both controls and speakers with MTD with high sensitivity and specificity (area under the curve statistics of 0.94 and 0.86, respectively). CONCLUSION: Percentage of creak as calculated by an automated creak detector may be useful as a quantitative indicator of AdLD, demonstrating the potential for use as a screening tool or to aid in a differential diagnosis. LEVEL OF EVIDENCE: 3 Laryngoscope, 133:2687-2694, 2023.

Voice and Speech Changes in Transmasculine Individuals Following Circumlaryngeal Massage and Laryngeal Reposturing.

PURPOSE: The purpose of this study was to measure the short-term effects of circumlaryngeal massage and laryngeal reposturing on acoustic and perceptual characteristics of voice in transmasculine individuals. METHOD: Fifteen transmasculine individuals underwent one session of sequential circumlaryngeal massage and laryngeal reposturing with a speech-language pathologist. Voice recordings were collected at three time points-baseline, postmassage, and postreposturing. Fundamental frequency (f o), formant frequencies, and relative fundamental frequency (RFF; an acoustic correlate of laryngeal tension) were measured. Estimates of vocal tract length (VTL) were derived from formant frequencies. Twelve listeners rated the perceived masculinity of participants' voices at each time point. Repeated-measures analyses of variance measured the effect of time point on f o, estimated VTL, RFF, and perceived voice masculinity. Significant effects were evaluated with post hoc Tukey's tests. RESULTS: Between baseline and end of the session, f o decreased, VTL increased, and participant voices were perceived as more masculine, all with statistically significant differences. RFF did not differ significantly at any time point. Outcomes were highly variable at the individual level. CONCLUSION: Circumlaryngeal massage and laryngeal reposturing have short-term effects on select acoustic (f o, estimated VTL) and perceptual characteristics (listener-assigned voice masculinity) of voice in transmasculine individuals. SUPPLEMENTAL MATERIAL: https://doi.org/10.23641/asha.19529299.

Automated Relative Fundamental Frequency Algorithms for Use With Neck-Surface Accelerometer Signals.

OBJECTIVE: Relative fundamental frequency (RFF) has been suggested as a potential acoustic measure of vocal effort. However, current clinical standards for RFF measures require time-consuming manual markings. Previous semi-automated algorithms have been developed to calculate RFF from microphone signals. The current study aimed to develop fully automated algorithms to calculate RFF from neck-surface accelerometer signals for ecological momentary assessment and ambulatory monitoring of voice. METHODS: Training a set of 2646 /vowel-fricative-vowel/ utterances from 317 unique speakers, with and without voice disorders, was used to develop automated algorithms to calculate RFF values from neck-surface accelerometer signals. The algorithms first rejected utterances with poor vowel-to-noise ratios, then identified fricative locations, then used signal features to determine voicing boundary cycles, and finally calculated corresponding RFF values. These automated RFF values were compared to the clinical gold-standard of manual RFF calculated from simultaneously collected microphone signals in a novel test set of 639 utterances from 77 unique speakers. RESULTS: Automated accelerometer-based RFF values resulted in an average mean bias error (MBE) across all cycles of 0.027 ST, with an MBE of 0.152 ST and -0.252 ST in the offset and onset cycles closest to the fricative, respectively. CONCLUSION: All MBE values were smaller than the expected changes in RFF values following successful voice therapy, suggesting that the current algorithms could be used for ecological momentary assessment and ambulatory monitoring via neck-surface accelerometer signals.

Clinical Cutoff Scores for Acoustic Indices of Vocal Hyperfunction That Combine Relative Fundamental Frequency and Cepstral Peak Prominence.

PURPOSE: This study examined the discriminative ability of acoustic indices of vocal hyperfunction combining smoothed cepstral peak prominence (CPPS) and relative fundamental frequency (RFF). METHOD: Demographic, CPPS, and RFF parameters were entered into logistic regression models trained on two 1:1 case-control groups: individuals with and without nonphonotraumatic vocal hyperfunction (NPVH; n = 360) and phonotraumatic vocal hyperfunction (PVH; n = 240). Equations from the final models were used to predict group membership in two independent test sets (n = 100 each). RESULTS: Both CPPS and RFF parameters significantly improved model fits for NPVH and PVH after accounting for demographics. CPPS explained unique variance beyond RFF in both models. RFF explained unique variance beyond CPPS in the PVH model. Final models included CPPS and RFF offset parameters for both NPVH and PVH; RFF onset parameters were significant only in the PVH model. Area under the receiver operating characteristic curve analysis for the independent test sets revealed acceptable classification for NPVH (72%) and good classification for PVH (86%). CONCLUSIONS: A combination of CPPS and RFF parameters showed better discriminative ability than either measure alone for PVH. Clinical cutoff scores for acoustic indices of vocal hyperfunction are proposed for assessment and screening purposes.

Relative Fundamental Frequency in Individuals with Globus Syndrome and Muscle Tension Dysphagia.

OBJECTIVE: Relative fundamental frequency (RFF) has been investigated as an acoustic measure to assess for changes in laryngeal tension. This study aimed to assess RFF in individuals with globus syndrome, individuals with muscle tension dysphagia (MTDg), and individuals with typical voices. METHODS: RFF values were calculated from the speech acoustics of individuals with globus syndrome (n = 12), individuals with MTDg (n = 12), and age- and sex-matched controls with typical voices (n = 24). An analysis of variance was performed on RFF values to assess the effect of group. RESULTS: There was no statistically significant effect of group on RFF values, with similar values for individuals with globus syndrome, individuals with MTDg, and control participants. CONCLUSIONS: These results suggest that individuals with these disorders do not appear to possess paralaryngeal muscle tension in a locus and/or manner that directly impacts voice production.

Hey Siri: How Effective are Common Voice Recognition Systems at Recognizing Dysphonic Voices?

OBJECTIVES/HYPOTHESIS: Interaction with voice recognition systems, such as Siri™ and Alexa™, is an increasingly important part of everyday life. Patients with voice disorders may have difficulty with this technology, leading to frustration and reduction in quality of life. This study evaluates the ability of common voice recognition systems to transcribe dysphonic voices. STUDY DESIGN: Retrospective evaluation of "Rainbow Passage" voice samples from patients with and without voice disorders. METHODS: Participants with (n = 30) and without (n = 23) voice disorders were recorded reading the "Rainbow Passage". Recordings were played at standardized intensity and distance-to-dictation programs on Apple iPhone 6S™, Apple iPhone 11 Pro™, and Google Voice™. Word recognition scores were calculated as the proportion of correctly transcribed words. Word recognition scores were compared to auditory-perceptual and acoustic measures. RESULTS: Mean word recognition scores for participants with and without voice disorders were, respectively, 68.6% and 91.9% for Apple iPhone 6S™ (P < .001), 71.2% and 93.7% for Apple iPhone 11 Pro™ (P < .001), and 68.7% and 93.8% for Google Voice™ (P < .001). There were strong, approximately linear associations between CAPE-V ratings of overall severity of dysphonia and word recognition score, with correlation coefficients (R2 ) of 0.609 (iPhone 6S™), 0.670 (iPhone 11 Pro™), and 0.619 (Google Voice™). These relationships persisted when controlling for diagnosis, age, gender, fundamental frequency, and speech rate (P < .001 for all systems). CONCLUSION: Common voice recognition systems function well with nondysphonic voices but are poor at accurately transcribing dysphonic voices. There was a strong negative correlation with word recognition scores and perceptual voice evaluation. As our society increasingly interfaces with automated voice recognition technology, the needs of patients with voice disorders should be considered. LEVEL OF EVIDENCE: 4 Laryngoscope, 131:1599-1607, 2021.

Reliability and Accuracy of Expert Auditory-Perceptual Evaluation of Voice via Telepractice Platforms.

Purpose This study assessed the reliability and accuracy of auditory-perceptual voice evaluations by experienced clinicians via telepractice platforms. Method Voice samples from 20 individuals were recorded after transmission via telepractice platforms. Twenty experienced clinicians (10 speech-language pathologists, 10 laryngologists) evaluated the samples for dysphonia percepts (overall severity, roughness, breathiness, and strain) using a modified Consensus Auditory-Perceptual Evaluation of Voice. Reliability was calculated as the mean of squared differences between repeated ratings (intrarater agreement), and between individual and group mean ratings (interrater agreement). Repeated measures analyses of variance were constructed to measure effects of transmission condition (e.g., original recording, WebEx, Zoom), dysphonia percept, and their interaction on intrarater agreement, interrater agreement, and average ratings. Significant effects were evaluated with post hoc Tukey's tests. Results There were significant effects of transmission condition, percept, and their interaction on average ratings, and a significant effect of percept on interrater agreement. Post hoc testing revealed statistically, but not clinically, significant differences in average roughness ratings across transmission conditions, and significant differences in interrater agreement for several percepts. Overall severity had the highest agreement and strain had the lowest. Conclusion Telepractice transmission does not substantially reduce reliability or accuracy of auditory-perceptual voice evaluations by experienced clinicians.

Acoustic Identification of the Voicing Boundary during Intervocalic Offsets and Onsets based on Vocal Fold Vibratory Measures.

Methods for automating relative fundamental frequency (RFF)-an acoustic estimate of laryngeal tension-rely on manual identification of voiced/unvoiced boundaries from acoustic signals. This study determined the effect of incorporating features derived from vocal fold vibratory transitions for acoustic boundary detection. Simultaneous microphone and flexible nasendoscope recordings were collected from adults with typical voices (N=69) and with voices characterized by excessive laryngeal tension (N=53) producing voiced-unvoiced-voiced utterances. Acoustic features that coincided with vocal fold vibratory transitions were identified and incorporated into an automated RFF algorithm ("aRFF-APH"). Voiced/unvoiced boundary detection accuracy was compared between the aRFF-APH algorithm, a recently published version of the automated RFF algorithm ("aRFF-AP"), and gold-standard, manual RFF estimation. Chi-square tests were performed to characterize differences in boundary cycle identification accuracy among the three RFF estimation methods. Voiced/unvoiced boundary detection accuracy significantly differed by RFF estimation method for voicing offsets and onsets. Of 7721 productions, 76.0% of boundaries were accurately identified via the aRFF-APH algorithm, compared to 70.3% with the aRFF-AP algorithm and 20.4% with manual estimation. Incorporating acoustic features that corresponded with voiced/unvoiced boundaries led to improvements in boundary detection accuracy that surpassed the gold-standard method for calculating RFF.

Acoustic Model of Perceived Overall Severity of Dysphonia in Adductor-Type Laryngeal Dystonia.

Purpose This study is a secondary analysis of existing data. The goal of the study was to construct an acoustic model of perceived overall severity of dysphonia in adductory laryngeal dystonia (AdLD). We predicted that acoustic measures (a) related to voice and pitch breaks and (b) related to vocal effort would form the primary elements of a model corresponding to auditory-perceptual ratings of overall severity of dysphonia. Method Twenty inexperienced listeners evaluated the overall severity of dysphonia of speech stimuli from 19 individuals with AdLD. Acoustic features related to primary signs of AdLD (hyperadduction resulting in pitch and voice breaks) and to a potential secondary symptom of AdLD (vocal effort, measures of relative fundamental frequency) were computed from the speech stimuli. Multiple linear regression analysis was applied to construct an acoustic model of the overall severity of dysphonia. Results The acoustic model included an acoustic feature related to pitch and voice breaks and three acoustic measures derived from relative fundamental frequency; it explained 84.9% of the variance in the auditory-perceptual ratings of overall severity of dysphonia in the speech samples. Conclusions Auditory-perceptual ratings of overall severity of dysphonia in AdLD were related to acoustic features of primary signs (pitch and voice breaks, hyperadduction associated with laryngeal spasms) and were also related to acoustic features of vocal effort. This suggests that compensatory vocal effort may be a secondary symptom in AdLD. Future work to generalize this acoustic model to a larger, independent data set is necessary before clinical translation is warranted.

Transmasculine Voice Modification: A Case Study.

This case study measured the effects of manual laryngeal therapy on the fundamental frequency (fo), formant frequencies, estimated vocal tract length, and listener perception of masculinity of a 32-year-old transmasculine individual. The participant began testosterone therapy 1.5 years prior to the study. Two therapy approaches were administered sequentially in a single session: (1) passive circumlaryngeal massage and manual laryngeal reposturing, and (2) active laryngeal reposturing with voicing. Acoustic recordings were collected before and after each treatment and 3 days after the session. Speaking fo decreased from 124 Hz to 120 Hz after passive training, and to 108 Hz after active training. Estimated vocal tract length increased from 17.0 cm to 17.3 cm after passive training, and to 19.4 cm after active training. Eight listeners evaluated the masculinity of the participant's speech; his voice was rated as most masculine at the end of the training session. All measures returned to baseline at follow-up. Overall, both acoustic and perceptual changes were observed in one transmasculine individual who participated in manual laryngeal therapy, even after significant testosterone-induced voice changes had already occurred; however, changes were not maintained in the follow-up. This study adds to scant literature on effective approaches to and proposed outcome measures for voice masculinization in transmasculine individuals.

Budgie in the mirror: An exploratory analysis of social behaviors and mirror use in the Budgerigar (Melopsittacus undulatus).

Despite their prevalent use amongst avian species, particularly in Budgerigars, the relationship between mirror use and social behaviors is in need of elucidation. After establishing social behavior measures (e.g., agonistic and pair-bond data) in a colony of Budgerigars, mirrors were introduced and subsequent use was documented. Results illustrated a significant relationship between mirror use and pair-bonds, such that mirror use increased with increasing pair-bond strength. The authors posit that individual differences in gregariousness may best account for the obtained results, with certain birds seeking more social stimulation than others, whether it be from a pair-bond or from its own reflected mirror image. Additionally, the trends observed within agonistic behaviors were consistent with previous avian literature such that those who initiated more encounters also won more encounters and that there was a preference to be closer to birds that win more encounters.