Vocal Intensity Research Articles (Page 1)

Prior studies of vocal auditory-motor control in people with hyperfunctional voice disorders (HVDs) have found evidence of unusually large responses to auditory feedback perturbations of fundamental frequency (F0) and more variable voice onset times in unperturbed speech. However, it is unknown whether people with HVDs perform similarly to people with typical voices when asked to make small changes in vocal parameters in volitional tasks. The purpose of this study was to compare performance on minimal movement tasks for F0 and intensity in people with and without HVDs. Twenty-six people with HVDs and 26 matched controls participated in tasks to assess the smallest volitional increases and decreases they could make in vocal F0 and intensity. Measures included the mean smallest change, variability of change, and accuracy of the direction of change. Group differences were tested with general linear models. No significant differences were found between people with and without HVDs on any of the measures. Singers produced significantly smaller mean smallest changes of both F0 and intensity than nonsingers. Our findings support the interpretation of prior studies of auditory-motor control in people with HVDs. Specifically, unusually large responses to perturbations of vocal auditory feedback cannot be explained by a broader impairment of the ability to make small changes in the vocal parameters F0 or intensity. The method devised to assess minimal movements for voice is sensitive to relevant group differences, such as singing experience. https://doi.org/10.23641/asha.30004969.

Perception of Vocal Tremor After Deep Brain Stimulation in Varying Speech Contexts.

Auditory-Motor Interaction for Vocal Rhythms and Resonance in Individuals With Normal and Pathological Voices.

BackgroundEmotional resilience (traditionally defined as the capacity to recover from adversity) and cognitive load (the mental effort for processing information) are critical aspects of mental health functioning. Traditional assessment methods, such as physiological sensors and post-task surveys, often disrupt natural behavior and fail to provide real-time insights. Speech prosody, encompassing pitch, intensity, loudness, and voice activity, offer a non-intrusive alternative for evaluating these psychological constructs. However, the relationship between speech prosody, emotional resilience, and cognitive load remains underexplored, particularly in conversational contexts.ObjectiveThis study proposes proxy measures for these constructs based on self-reported engagement, enjoyment, boredom, and cognitive effort during dyadic conversation. By leveraging the SEWA (Automatic Sentiment Estimation in the Wild) database, developed through a European research project on emotion recognition, the research seeks to develop machine learning models that correlate speech patterns with subjective self-reports of emotional and cognitive states.MethodsProsodic features, such as pitch variation, vocal intensity, and voice activity, were extracted from the SEWA database recordings. These features are then normalized to account for inter-speaker variability and used as predictors in machine learning models. Regression and classification models are employed to correlate speech features with subjective self-reports, which serve as ground truth for Positive Affective Engagement (as a proxy for emotional resilience) and Perceived Mental Strain (as a proxy for cognitive load). Data from English and German speakers are analyzed separately to account for linguistic and cultural differences.OutcomesThe study establishes a significant relationship between speech prosody and psychological states, demonstrating that Positive Affective Engagement (as a proxy for emotional resilience) and Perceived Mental Strain (as a proxy for cognitive load) can be effectively predicted through prosodic features. Higher emotional resilience is linked to more discernible prosodic patterns in German speech, such as higher loudness and greater voice probability consistency. In contrast, cognitive load prediction remains consistent across English and German datasets.ConclusionThis research introduces a novel approach for assessing Positive Affective Engagement (as a proxy for emotional resilience) and Perceived Mental Strain (as a proxy for cognitive load) through speech prosody, highlighting the significant impact of language-specific variations. By combining prosodic features with machine learning techniques, the study offers a promising alternative to traditional psychological assessments. The findings emphasize the need for tailored, multilingual models to accurately estimate psychological states, with potential applications in mental health monitoring, cognitive workload analysis, and human-computer interaction. This work lays the foundation for future innovations in speech-based psychological profiling, advancing our understanding of human emotional and cognitive states in diverse linguistic contexts.

This study examines how auditory spectral representations in the peripheral auditory system explain changes in vowel production under noisy conditions, especially when lower formants (F1 and F2) are masked. Ten adult male Japanese speakers produced sustained vowels /a/ and /i/ under quiet and noisy conditions involving three noise types (broadband, low-pass, and high-pass) at 75 and 85 dB. We analyzed vocal intensity and the amplitudes and frequencies of the F1 and F2. Auditory spectral representations, simulated using a loudness model, were used to estimate excitation patterns in the auditory periphery. Most noise conditions significantly increased vocal intensity and the amplitude of both formants. F1 frequency consistently shifted upward under high-intensity broadband noise, while F2 shifts depended on vowel and noise type, shifting upward for /a/ and downward for /i/. Some patterns could not be explained by power spectra alone. Instead, they were better accounted for by frequency-specific masking effects, reflected in overlapping excitation patterns in the auditory spectral representation. These overlaps indicated reduced self-audibility in specific frequency bands, triggering compensatory adjustments. The findings highlight how auditory masking influences speech production, supporting a perceptually grounded model of auditory-motor control in noisy environments.

The goal of this study was to understand the interaction between the voice source spectral shape, formant tuning, and fundamental frequency in determining the vocal tract contribution to vocal intensity. Computational voice simulations were performed with parametric variations in both vocal fold and vocal tract configurations. The vocal tract contribution to vocal intensity was quantified as the difference in the A-weighted sound pressure level between the radiated sound pressure and the sound pressure at the glottis. The results from the simulations showed that the vocal tract contribution to vocal intensity depends strongly on the amplitude of the first vocal tract resonance. Two strategies to increase vocal tract contribution to vocal intensity were identified. The first strategy was to increase vocal fold adduction, which increases the relative prominence of the harmonics near the first vocal tract resonance. The second strategy was to bring a vocal tract resonance and a nearby harmonic closer in frequency, often known as formant tuning. In this study, increasing vocal fold adduction was the primary strategy at low fundamental frequencies, whereas formant tuning was more effective at high fundamental frequencies, particularly when formant tuning involves the strongest harmonic in the voice source spectrum.

Videolaryngostroboscopic Findings in Vocally HealthyIndividuals After Vocal LoadingActivity.

Automatic Measurement and Classification of Pressure Peaks When Estimating Subglottal Pressure by Oral Pressure During p-Occlusions.

ABSTRACT Cyberbullying through voice-based communication on platforms such as WhatsApp, Facebook, and voice messaging applications has emerged as a growing concern, yet it remains underexplored in the current research landscape. This study proposes a novel hybrid deep learning approach for automatic cyberbullying user profile detection using voice features, specifically targeting the emotional and psychological cues present in audio data. Unlike previous works that primarily rely on textual or image-based cues, our method extracts and analyzes audio signals using the Librosa Python library to capture vocal cord variations and emotional tone – key indicators of negative behavior and distress in both perpetrators and victims. The proposed system incorporates a hybrid architecture combining Convolutional Neural Networks (CNN) for feature extraction and Long Short-Term Memory (LSTM) networks for temporal pattern recognition. The model is trained on real-world voice datasets, and preprocessing involves noise reduction, voice activity detection, and MFCC feature extraction. Emotional attributes such as pitch variation, spectral contrast, and vocal intensity are used to infer behavioral patterns. Our experimental results demonstrate a significant performance improvement compared to state-of-the-art approaches in terms of accuracy, precision, recall, and F1-score, showing that voice features can be highly effective in detecting cyberbullying tendencies. While earlier works achieved accuracy ranging from 66% to 92% using models such as Logistic Regression, XGBoost, CR-based CNN-RNN, and other traditional deep learning methods, our proposed hybrid deep learning model (CNN+LSTM) achieves superior accuracy of up to 98%.

Parkinson's Disease (PD) is the second most common neurodegenerative disorder. Hypokinetic dysarthria, a motor speech disorder affects approximately 90% of PD patients. In Morocco, linguistic features of Moroccan Arabic may influence how voice impairment appear in PD. This study investigates the acoustic characteristics of Moroccan PD patients' voices. This was a case-control study involving 30 Moroccan PD patients, both medicated and unmedicated) and 30healthy control participants. Voice recordings were collected from participants before and after dopaminergic treatment. Four tasks were performed: sustained vowel production, spontaneous speech, diadochokinesis, and reading tasks in both Arabic and French. The Unified Parkinson's Disease Rating Scale was computed for each state and data were analyzed using PRAAT software (v6.2), for phonatory and articulatory parameters. PD patients exhibited significant alterations in shimmer, jitter, and Harmonics-to-Noise and Noise-to-Harmonic Ratios, reflecting pitch and vocal intensity instability. A "hoarse" and "breathy" vocal quality was noted. Rhythm disturbances were marked by an increased number of long pauses (≥500 ms), indicating difficulty in sustaining speech flow. No adverse events were reported. Moroccan PD patients exhibit distinctive voice and rhythm abnormalities, particularly in vowel articulation, phonation, and timing. These markers may serve as reliable indicators for PD diagnosis and progression.

Automatic classification of vocal intensity categories from amplitude-normalized speech signals by comparing acoustic features and classifier models

Objectives: This study examines age-related differences in acoustic characteristics of emotional speech, focusing on pitch, intensity, and utterance duration across four emotional types (happiness, sadness, anger, and neutral) in older and younger female adults. Methods: The participants consisted of 16 young female adults and 16 elderly female individuals, all of whom scored within normal ranges on the Korean Mini-Mental State Examination-2 and the Geriatric Depression Scale. Four emotional episodes were visually presented, and recordings were conducted with each episode being uttered three times. Emotional scenarios were developed to elicit empathetic and expressive speech, which was analyzed using Praat software (version 6.4.23). Results: Significant age-related differences were observed in minimum fundamental frequency and its range. Additionally, all intensity measures (mean intensity, minimum intensity, maximum intensity, and intensity range) differed significantly between the two groups. Utterance duration was longer in older adults than in younger adults, indicating a slower speech rate. Conclusion: These findings suggest age-related changes in vocal intensity and speech rate while preserving emotional expression capabilities, contributing to our understanding of aging and communication. This study goes beyond simply comparing the acoustic features of emotional speech by enabling a more precise analysis of the differences in emotional speech between the elderly and young adults. These findings provide important implications for future research on emotional speech and expand our understanding of emotional expression in older adults.

The aim of the study is to determine the role of emotive speech acts in cross-cultural language learning environments, revealing the complex interplay between universal emotional markers and culturally specific expression patterns. In the course of the research, data analysis methods were applied (acoustic analysis, facial expression analysis using the Facial Action Coding System (FACS), lexical analysis, correlational and regression analysis). Through comprehensive analysis of acoustic features, facial expressions, and lexical patterns, the research demonstrates that emotional expression follows dual patterns: universal elements remain consistent across languages while others undergo significant cultural adaptation. Results indicate that language learners develop an “emotional interlanguage” that synthesizes native expression strategies with target language norms. Spanish learners exhibited greater facial expressiveness when expressing happiness, suggesting adoption of the target culture’s more overt emotional display rules. Anger was more explicitly verbalized across all language learning groups, indicating that different emotions utilize distinct channels of expression. Principal component analysis and hierarchical clustering revealed discrete emotional expression profiles across language groups, while multiple regression models identified predictive relationships between linguistic proficiency, cultural exposure, and emotional adaptation. Our findings support a nuanced theoretical model that integrates universalist and relativist perspectives on emotional expression, suggesting that language learners navigate a dynamic space between these poles. The research confirms that certain aspects of emotional expression — such as increased vocal intensity for anger and decreased speech rate for sadness — remain relatively consistent across language groups, supporting the universality hypothesis. However, other aspects — particularly facial expressiveness for happiness and lexical choices for emotional states — show significant adaptation to target language norms, supporting the cultural relativity perspective. Our data reveals that language learners develop what might be termed an “emotional interlanguage” — a dynamic system of emotional expression that incorporates elements from both their native emotional repertoire and the target language’s cultural norms. This emotional interlanguage evolves with increased language proficiency and cultural exposure, but the adaptation process varies across different channels of emotional expression and across different emotions. The finding that cultural familiarity mediates the relationship between language proficiency and emotional expressiveness suggests that emotional adaptation in language learning is not simply a function of linguistic knowledge, but requires deeper cultural learning and engagement.

The vocalization behaviour of birds is influenced by multiple factors, such as reproductive status, photoperiod and climatic conditions, which drive adjustments in their calls to respond to environmental changes. This study employed passive acoustic monitoring and automated signal recognition technology to analyse the daily and seasonal variations in the vocal rhythms of the red junglefowl Gallus gallus jabouillei over a complete annual cycle, examining the impact of climatic factors on their calling activity. Results revealed significant diurnal and seasonal variations in the vocalization behaviour of red junglefowls. The daily activity patterns showed distinct morning and evening peaks, occurring one hour before sunrise and sunset, respectively. During the breeding season, calling activity progressively increased, reaching its peak during the incubation and brooding periods. The vocal intensity of red junglefowls fluctuated with environmental changes, showing significant associations with daily precipitation and morning temperatures but was unaffected by daily maximum wind speeds. During the full moon, the morning peak was observed to occur earlier than during other lunar phases, whereas the evening peak consistently coincided with sunset. This study further suggested the optimal sampling periods and minimum monitoring durations required to track red junglefowl vocal behaviour, offering guidance for call-based population surveys and passive acoustic monitoring.

This study explored the genetics of the levels of stress hormones (cortisol, cortisone, DHEA, and DHEA-S) in hair of 863 clinically healthy Yorkshire × Landrace male pigs at ∼40 days of age and evaluated their potential as biomarkers of innate stress response by estimating genetic correlations with responses to a 30 s backtest performed at ∼27 days of age. Backtest responses included the number and intensity of vocalizations (VN and VI) and struggles (SN and SI). With pigs genotyped using a 50 K single nucleotide polymorphism (SNP) panel that was imputed to 650 K SNPs, heritability estimates for the levels of cortisol, cortisone, DHEA, and DHEA-S were 0.33, 0.04, 0, and 0.31, respectively, while those for backtest responses ranged from 0.26 to 0.57. Litter effects accounted for 9 to 16% of the phenotypic variance for stress hormone levels and none for backtest responses. Genetic correlation estimates among stress hormone levels were strongest between cortisol and cortisone (0.99 ± 0.12), while those among backtest responses ranged from 0.60 to 0.99. Cortisol was estimated to have moderate genetic correlations with VN (0.24 ± 0.19) and VI (0.50 ± 0.24) but not with SN and SI. Genome-wide association studies identified a major quantitative trait locus (QTL) for hair cortisol levels near the glucocorticoid receptor gene (NR3C1) that explained 45.3% of the genetic variance and that may be different than a causative mutation that was previously identified in this gene for cortisol levels in porcine blood. An extra copy of the minor allele (frequency = 9%) at the lead SNP for this QTL, rs341258564 originated from both parental breeds and reduced levels of cortisol by 30 ± 6% and of cortisone by 17 ± 4%, and increased VN by 5 ± 2%. Additional QTL with smaller effects (1.0 to 11.1% of genetic variance) were identified for DHEA-S, cortisol/DHEA-S, cortisone/DHEA-S, VI, and VN. Ranked gene set enrichment analyses of 0.25 Mb windows based on genetic variance explained showed that windows associated with glucocorticoid levels were enriched for biological terms related to energy production and suppression of inflammation. In contrast, those associated with DHEA-S were enriched for biological processes related to immunity activation and gene transcriptional and post-transcriptional regulation. These findings establish the genetic basis of stress response in young and clinically healthy pigs, identify the genomic location of a major QTL for hair cortisol levels, and show that cortisol levels in hair of young and healthy pigs are potential genetic biomarkers for the innate coping response style of pigs to noninfectious stressors. These results open avenues that can facilitate selection of pigs that cope better with noninfectious stressors.

Purpose: Extensive literature exists about the effects of LSVT LOUD®, SPEAK OUT!®, and SPEAK OUT!® Therapy Group programs on voice and speech outcomes for individuals living with Parkinson Disease (PD). Prior studies have reported increased speech intelligibility and vocal intensity among participants with PD immediately upon completing one-on-one individual programs (including LSVT LOUD® and SPEAK OUT!® therapy). However, few researchers have reported longitudinal data on the specific effects of the SPEAK OUT!® Therapy Group program on participants’ cognitive, self-reported swallowing, depression, and voice-related quality of life (QoL) status. Therefore, the current study examined the effects of SPEAK OUT!® Therapy Group participation for fourteen participants with PD in multiple domains, including cognition, self-reported swallowing status, depression, and voice-related QoL across two years.Methods: Data were collected at six-month intervals over two years. Participants included a treatment (TX) group that attended weekly SPEAK OUT!® Therapy Group sessions and a control (CN) group that did not receive any speech therapy services for their PD symptoms.Results: No statistically significant group differences were found for the different measures. However, descriptive statistics showed maintenance, rather than decline, in the TX group for cognition and other domains compared to the CN group at the end of two years. In contrast, the CN group demonstrated a slight cognitive decline during the same period.Conclusions: The current study provides evidence of possible slow progression of non-voice domains such as cognition, self-reported swallowing, and depression after participating in the SPEAK OUT!® Therapy Group program. However, additional studies are warranted.

This case report investigates the effects of semi-occluded vocal tract exercises (SOVTEs) on voice improvement in elderly female patients with presbyphonia. Elderly female patients with presbyphonia commonly present with symptoms such as hoarseness, decreased vocal intensity, and phonatory difficulties. These symptoms are often associated with age-related vocal fold atrophy, leading to compensatory muscle tension and inefficient phonation, which necessitate targeted therapeutic interventions. In this study, two elderly female patients, aged 73 years and 71 years, participated in a voice therapy program centered on SOVTEs designed to promote vocal fold vibration efficiency and reduce compensatory tension. Case 1 underwent five therapy sessions, while Case 2 completed sixteen sessions. Pre- and post-treatment voice assessments revealed notable improvements in vocal quality, suggesting that SOVTEs may be an effective therapeutic approach for managing presbyphonia in elderly women.

This study investigated the characteristic changes of brain function in children with Tourette syndrome (TS) based on resting-state functional MRI (rs-fMRI) and their correlation with the Yale Global Tic Severity Scale (YGTSS) scores. A total of 19 children diagnosed as TS at the Department of Pediatrics in China-Japan Friendship Hospital from October 2021 to April 2023 were enrolled prospectively as the TS group, and 20 healthy children matched for age and gender were field recruited from the community as the control group. All participants underwent head rs-fMRI scans, and regional homogeneity (ReHo) and amplitude of low-frequency fluctuations (ALFF) values were calculated. YGTSS scores were assessed for the TS group. Pearson correlation analysis and Spearman correlation analysis were used to evaluate the correlation between ReHo, ALFF and YGTSS scores. Receiver operating characteristic (ROC) curves were plotted and area under the curve (AUC) was calculated to evaluate the efficacy of brain function parameters related to YGTSS score in diagnosing TS. The TS group included 19 patients, and aged (8.5±2.6) years, including 13 males (68.4%). The control group included 20 children, and aged (5.5±1.8) years, including 7 males (35.0%). In the TS group, the ReHo values in the left lentiform nucleus gray matter, left occipital lobe fusiform gyrus, right occipital lobe medial gyrus, and right frontal lobe paracentral lobule white matter were lower than those in the control group. The ReHo values in the pons and right cerebellar tonsil region, as well as the ALFF values in the right lingual gyrus, were higher than those in the control group (all P<0.01). The ReHo values in the right frontal lobe paracentral lobule, right occipital lobe medial gyrus, and left occipital lobe fusiform gyrus were positively correlated with vocal frequency (r=0.466, 0.576, 0.491) and intensity (r=0.498, 0.593, 0.609); the ReHo values in the right occipital lobe medial gyruss (r=0.615) and left occipital lobe fusiform gyrus (r=0.661) were positively correlated with the number of vocal tics (all P<0.05). The AUC (95%CI) for discriminating TS based on the ReHo values in the right frontal lobe paracentral lobule, right occipital lobe medial gyrus, and left occipital lobe fusiform gyrus were 0.890 (0.774-1.000), 0.863 (0.745-0.982), and 0.858 (0.760-0.988), respectively, with no statistically significant differences in the efficacy of these 3 ReHo values for discriminating TS (all P>0.05). Children with TS have characteristic brain function changes, and there is a correlation between the characteristic brain regions of MRI brain function changes and vocal tic symptoms.

Formant tuning refers to matching one of the lower vocal tract resonances to a harmonic of the voice source to maximize the radiated vocal intensity. While it is often observed in sopranos singing at high notes, there have been conflicting reports for its use in male voices. Formant tuning can have large impact on vocal intensity at high pitches, where the harmonics are far apart and their amplitudes are sensitive to frequency spacing between the harmonic and the nearest vocal tract resonance, whereas at low pitches it may become less effective. In this study, the effect of formant tuning on vocal intensity is investigated in computational simulations with parametric variations in both the vocal fold and vocal tract properties. The effect of formant tuning is quantified by the difference between the A-weighted radiated sound pressure level (SPL) and the source SPL. The results showed that formant tuning significantly increases SPL at high pitches but has only small effect at low pitches. With sufficient formant tuning, the radiated SPL also increases with increasing pitch, indicating that formant tuning is more effective in increasing SPL when vocal tract resonance is tuned to a lower harmonic.

Vocal responses to elevated background noise (Lombard effect) show promise for clinical voice assessment for occupational voice users, but implementation is limited by impractical loudspeaker setups in clinical environments. Additionally, the relationship between communication-related stress and voice use suggests that different noise types may impact vocal behavior differently. This study investigated how individuals adjust their vocal patterns in response to different types of background noise (pink noise versus babble noise) and noise delivery methods (loudspeakers versus open-air headphones). Participants simulated grade-school lectures under the four noise conditions calibrated at 75 dBA. Vocal responses were measured using sound pressure levels (SPL), self-perceived vocal effort through the Borg CR-100 scale, and subjective task difficulty was assessed using the NASA Task Load Index (NASA-TLX). Participants demonstrated equivalent changes in vocal loudness and perceived task difficulty regardless of noise delivery method, suggesting that open-air headphones can effectively replace traditional loudspeaker setups. Babble noise was consistently rated as more challenging than pink noise across both delivery methods. These findings validate the use of open-air headphones for Lombard effect studies in clinical environments and highlight the differential impact of noise types on communication stress. This research provides valuable insights for clinical voice assessment protocols.