Voice Recordings Research Articles

Predicting Transitions from Mild Cognitive Impairment to Dementia within 5‐years

AbstractBackgroundEarly detection of individuals at high risk of future progression toward dementia is in great need, particularly in the era of emerging treatments for Alzheimer’s disease. Given the easier administration of neuropsychological (NP) tests over invasive and expensive biomarker acquisition, we seek to predict the future cognitive status of an individual based on digital voice recordings of NP exams.MethodWe utilized automatic speech recognition and natural language processing techniques to create an automated tool that predicts if an individual with mild cognitive impairment (MCI) will develop dementia in a fixed follow‐up window (5 years). We characterize the predictive accuracy of this approach using data from the Framingham Heart Study. First, the automated transcription of the digital voice recordings were classified into 8 main NP sub‐tests including memory assessment, naming and language skill, verbal fluency, general questions, etc. Using a large language model, only the participants’ sentences were encoded into quantitative data. These data and the participants’ characteristics, such as age, sex, ApoE4 (+/‐) genotype, and education were employed to train and test a binary classifier, predicting dementia withing 5 years after a positive MCI diagnosis. We also provide a comparison between our approach and alternative models using demographics and ApoE genotype or the Mini‐Mental State Examination (MMSE) score.ResultWe evaluated the performance of the classification task on the data containing 111 stable MCI and 103 progressive MCI subjects. Our method achieved an average accuracy of 75.5% on the held‐out test data while a baseline model using age, sex, education, and ApoE has an average accuracy of 71.8%. The sensitivity and specificity of the proposed method versus the baseline model are 78.2% vs. 81.8% and 72.7% vs. 61.8%, respectively. Our approach also significantly outperforms the model based on the MMSE score by a margin of 14% in accuracy.ConclusionWe have analyzed digital voice recordings of NP exams for participants with MCI to predict their cognitive status in 5 years. The proposed approach offers a fully automated procedure with higher accuracy, providing an opportunity to develop an accessible and easy‐to‐administer screening tool that could be easily adapted to any language.

Topic Identification in Voice Recordings

Abstract The paper shows the understanding of a topic recognition problem like the speech recognition system based on Natural Language Processing (NLP) and the steps of its implementation of a rules-based approach, which is able to classify given audio materials based on predefined topics in real-time. During implementation, a statistical vocabulary was developed. Google Speech API (Application Programming Interface) was employed for subtitling audio materials, and the most ideal time frame for reception was identified through several experiments. The motivation of this work is based on the deficiency of similar simple systems for Hungarian topic recognition, even though numerous international languages already utilize multiple Automatic Sound Recognition (ASR) systems.

Natural speech as a digital biomarker in preclinical Alzheimer’s disease: Usability of a remote burst speech assessment

AbstractBackgroundSpeech is a complex process, relying on multiple interacting cognitive functions. Fine‐grained speech analysis offers the potential for measuring subtle cognitive deficits in early‐stage Alzheimer’s disease (AD). As such, digital recordings of natural speech are a promising digital biomarker, particularly when measured in bursts (repeated short measurements). This pilot study focused on the usability of remote speech bursts to assess natural speech in cognitively normal (CN) adults with and without AD pathologic change. Additionally, the association between amyloid‐beta (Aβ)‐pathology and speech characteristics was examined.MethodWe included 49 CN older adults from AMYPAD‐PNHS(location VUmc), SCIENCe, and Amsterdam Dementia Cohort (Age mean(M) = 68±standard deviation(SD) = 6, n = 28(57%) female, n = 23(47%) Aβ‐positive, Table 1). Aβ‐status was based on local cut‐offs for Aβ1‐42‐concentrations in cerebrospinal‐fluid or visual inspection of amyloid positron emission tomography (PET)‐imaging. The remote burst assessment comprised seventeen tablet‐based a)picture description, b)journal‐prompt storytelling and c)verbal‐fluency subtasks, administered over five consecutive days in an unsupervised home‐setting. Daily administration time was approximately 10 minutes. To assess usability, participants completed the system usability scale (SUS). Various acoustic features, including silent pauses, pause‐to‐word‐ratio and mean intensity (i.e., perceived loudness), were extracted from the voice recordings, and mean burst scores were calculated over the five‐day period. The association between Aβ‐status and acoustic features was investigated using two‐sample t‐tests stratified by subtask.ResultsThe 49 participants rated usability with a mean SUS‐score of 86(SD = 10). For example, 46(94%) participants thought the assessment‐app was easy to use (Figure 1). Ratings were lower in Aβ‐positive (M = 83±10) than Aβ‐negative individuals (M = 88±9, p = 0.02), reflecting good and excellent usability respectively. Regarding the association between Aβ‐status and acoustic features, more silent pauses were observed for Aβ‐positive (M = 0.30±0.09) compared to Aβ‐negative individuals (M = 0.25±0.06, p = 0.025, Figure 2) in the storytelling subtask. No group differences for pause‐to‐word‐ratio or mean intensity were observed (p>0.05).ConclusionRemote burst assessment of speech is feasible in CN Aβ‐positive and Aβ‐negative adults. Additionally, preliminary results demonstrated that Aβ‐pathology was related to more silent pauses, indicating its potential as a digital AD biomarker. These findings support remote burst speech assessments in monitoring cognitive changes, for example in observational studies and decentralized clinical trials of preclinical AD.

Vocal interaction between Eurasian Eagle-Owl (Bubo bubo) and canines (Carnivora, Canidae)

Abstract We recorded vocal interaction in the natural environment of an Eurasian Eagle-Owl (Bubo bubo) with canines Gray Wolves (Canis lupus), Red Foxes (Vulpes vulpes) and domestic dogs (Canis familiaris). Vocalization was recorded using Olympus digital voice recorders. The calls of the male Eurasian Eagle-Owl were recorded by us in the frequency range of 200–420 Hz. The howl of a Gray Wolf was recorded in the frequency range from 300 to 1,100 Hz. Red Fox barking was recorded in the frequency range from 750 to 1,000 Hz. Barking of domestic dogs was recorded in the frequency range from 250 to 1,500 Hz. The vocalization of the Eurasian Eagle-Owl had an independent character inherent in the biology of the species. The Eurasian Eagle-Owl, with its cries, involuntarily provoked the entry of canines into joint vocal interaction, which can be explained by the high social activity of the latter. Co-vocalizations of the Eurasian Eagle-Owl and canines were noted in winter, spring and autumn, but mainly in spring (50%). The increased use of autonomous voice recorders, which record spontaneous vocalizations emitted by animals over long periods, will allow us to better document and study the importance of such interspecific interactions.

Preoperative Recording of Original Voices in Laryngectomized Patients : The Save the Voice Project

Preoperative Recording of Original Voices in Laryngectomized Patients : The Save the Voice Project

Validation of an AI-assisted Treatment Outcome Measure for Gender-Affirming Voice Care: Comparing AI Accuracy to Listener’s Perception of Voice Femininity

OBJECTIVESThere is currently a lack of objective treatment outcome measures for transgender individuals undergoing gender-affirming voice care. Recently, Bensoussan et al. developed an AI model that is able to generate a voice femininity rating based on a short voice sample provided through a smartphone application. The purpose of this study was to examine the feasibility of using this model as a treatment outcome measure by comparing its performance to human listeners. Additionally, we examined the effect of two different training datasets on the model’s accuracy and performance when presented with external data. METHODS100 voice recordings from 50 cisgender males and 50 cisgender females were retrospectively collected from patients presenting at a university voice clinic for reasons other than dysphonia. The recordings were evaluated by expert and naïve human listeners, who rated each voice based on how sure they were the voice belonged to a female speaker (% voice femininity [R]). Human ratings were compared to ratings generated by (1) the AI model trained on a high-quality low-quantity dataset (voices from the Perceptual Voice Quality Database) (PVQD model), and (2) the AI model trained on a low-quality high-quantity dataset (voices from the Mozilla Common Voice database) (Mozilla model). Ambiguity scores were calculated as the absolute value of the difference between the rating and certainty (0 or 100%). RESULTSBoth expert and naïve listeners achieved 100% accuracy in identifying voice gender based on a binary classification (female >50% voice femininity [R]). In comparison, the Mozilla-trained model achieved 92% accuracy and the previously published PVQD model achieved 84% accuracy in determining voice gender (female > 50% AI voice femininity). While both AI models correlated with human ratings, the Mozilla-trained model showed a stronger correlation as well as lower overall rating ambiguity than the PVQD-trained model. The Mozilla model also appeared to handle pitch information in a similar way to human raters. CONCLUSIONSThe AI model predicted voice gender with high accuracy when compared to human listeners and has potential as a useful outcome measure for transgender individuals receiving gender-affirming voice training. The Mozilla-trained model performed better than the PVQD-trained model, indicating that for binary classification tasks, quantity of data may influence accuracy more than quality of the data used for training the voice AI models.

Small things—‘It felt like love’—The experience of being deeply moved in therapy: Clients' stories of the small things that matter in therapy

Small things—‘It felt like love’—The experience of being deeply moved in therapy: Clients' stories of the small things that matter in therapy

Collaboration Between Teachers and Learning Advisors to Improve Learner Outcomes

Qualitative research conducted at Toi Ohomai Institute of Technology investigated how kaiako (teachers) have been interacting with learning advisors (LA) and how they would like to engage with them to capitalise on their specialist skills to improve student outcomes. The research consisted of semi-structured interviews with 15 teachers from six campuses in the region, from a variety of teaching contexts including on-campus, online/distance, and mixed-mode learning. Teachers in a wide variety of subject areas were chosen to ascertain if there are different needs in any of those areas. The teachers were interviewed either in person or using Teams Meet software. Voice or video recordings were made and used for analysis and discussion. The results were tabulated and discussed by a panel of LA to find important implications of the study. The research determined that the experiences of teachers with LA are widely varied across the institute depending on factors such as geography, teaching styles, course duration, teacher and academic leader introductions to services, and other factors. All interviewees saw the value of LA but many were uncertain about how to engage with them and collaborate effectively at a classroom level. Where engagement exists at a classroom level, teachers are most satisfied with LA outcomes. The research indicates that closer collaboration can be achieved by increasing LA involvement in the classroom environment, particularly in areas identified as high needs. Critical to this collaboration is teachers’ facilitating contact with LA in the classroom. This collaboration will help identify student needs to enable timely interventions.

A machine learning method to process voice samples for identification of Parkinson’s disease

Machine learning approaches have been used for the automatic detection of Parkinson’s disease with voice recordings being the most used data type due to the simple and non-invasive nature of acquiring such data. Although voice recordings captured via telephone or mobile devices allow much easier and wider access for data collection, current conflicting performance results limit their clinical applicability. This study has two novel contributions. First, we show the reliability of personal telephone-collected voice recordings of the sustained vowel /a/ in natural settings by collecting samples from 50 people with specialist-diagnosed Parkinson’s disease and 50 healthy controls and applying machine learning classification with voice features related to phonation. Second, we utilize a novel application of a pre-trained convolutional neural network (Inception V3) with transfer learning to analyze the spectrograms of the sustained vowel from these samples. This approach considers speech intensity estimates across time and frequency scales rather than collapsing measurements across time. We show the superiority of our deep learning model for the task of classifying people with Parkinson’s disease as distinct from healthy controls.

Association Between Acoustic Features and Brain Volumes: the Framingham Heart Study.

Although brain magnetic resonance imaging (MRI) is a valuable tool for investigating structural changes in the brain associated with neurodegeneration, the development of non-invasive and cost-effective alternative methods for detecting early cognitive impairment is crucial. The human voice has been increasingly used as an indicator for effectively detecting cognitive disorders, but it remains unclear whether acoustic features are associated with structural neuroimaging. This study aims to investigate the association between acoustic features and brain volume and compare the predictive power of each for mild cognitive impairment (MCI) in a large community-based population. The study included participants from the Framingham Heart Study (FHS) who had at least one voice recording and an MRI scan. Sixty-five acoustic features were extracted with the OpenSMILE software (v2.1.3) from each voice recording. Nine MRI measures were derived according to the FHS MRI protocol. We examined the associations between acoustic features and MRI measures using linear regression models adjusted for age, sex, and education. Acoustic composite scores were generated by combining acoustic features significantly associated with MRI measures. The MCI prediction ability of acoustic composite scores and MRI measures were compared by building random forest models and calculating the mean area under the receiver operating characteristic curve (AUC) of 10-fold cross-validation. The study included 4,293 participants (age 57 ± 13 years, 53.9% women). During 9.3±3.7 years follow-up, 106 participants were diagnosed with MCI. Seven MRI measures were significantly associated with more than 20 acoustic features after adjusting for multiple testing. The acoustic composite scores can improve the AUC for MCI prediction to 0.794, compared to 0.759 achieved by MRI measures. We found multiple acoustic features were associated with MRI measures, suggesting the potential for using acoustic features as easily accessible digital biomarkers for the early diagnosis of MCI.

Confounding Factor Analysis for Vocal Fold Oscillations.

This paper provides a methodology to better understand the relationships between different aspects of vocal fold motion, which are used as features in machine learning-based approaches for detecting respiratory infections from voice recordings. The relationships are derived through a joint multivariate analysis of the vocal fold oscillations of speakers. Specifically, the multivariate setting explores the displacements and velocities of the left and right vocal folds derived from recordings of five extended vowel sounds for each speaker (/aa/, /iy/, /ey/, /uw/, and /ow/). In this multivariate setting, the differences between the bivariate and conditional interactions are analyzed by information-theoretic quantities based on transfer entropy. Incorporation of the conditional quantities reveals information regarding the confounding factors that can influence the statistical interactions among other pairs of variables. This is demonstrated on a vector autoregressive process where the analytical derivations can be carried out. As a proof of concept, the methodology is applied on a clinically curated dataset of COVID-19. The findings suggest that the interaction between the vocal fold oscillations can change according to individuals and presence of any respiratory infection, such as COVID-19. The results are important in the sense that the proposed approach can be utilized to determine the selection of appropriate features as a supplementary or early detection tool in voice-based diagnostics in future studies.

THE ROLE OF ACCENTS IN BATU BARA MALAY COMMUNICATION

This research aims to explore the role of accent in Batu Bara Malay communication. Batu Bara Malay is a dialect of Malay spoken in Batu Bara area of North Sumatra, Indonesia and has unique phonological characteristics. This research is also to identify the characteristics of the accent in Batu Bara Malay, understand its impact in daily communication, and investigate the role of accent in communication. This research uses a combination approach of case study and phonological analysis. The data were collected through interviews with native Batu Bara Malay speakers, observations of daily communication, and analysis of voice recordings. The research results show that Batu Bara Malay has distinctive accent characteristics, including phonological variations in the use of phonemes and intonation that are different from the standard Malay or Indonesian. In the context of daily communication, the use of accents in Batu Bara Malay affects understanding and expression. Differences in accents in informal conversation and formal communication were also identified. Social and cultural factors play a role in accent variation. This research contributes to further understanding of Batu Bara Malay and the importance of maintaining this dialect. The results can be used as a basis for efforts to preserve language and culture in Batu Bara Malay community.

A Comparative Analysis of Pronunciation Proficiency among Arabic and English Department Students Using Accent Software

One way to improve pronunciation skills is to use sophisticated technology, namely software that can help reduce accent errors. In this study we used ELSA Speak software which is an application (app) for moderating a non-native English accent. This study aims to determine whether accent reduction software can improve students’ pronunciation abilities. It uses quantitative methods with an experimental research approach. The research was conducted at one of the Islamic universities in Indonesia. We recruited 40 students – 20 from the English department and 20 from the Arabic language department. Their ages are between 19 and 20, and they have been studying English and Arabic within their departments for four semesters. Ten students from each department were allocated to the experimental class, and ten other students from both majors were placed in the control class. The experimental class was taught by researchers using accent reduction software, while the control class was taught by an English teacher using conventional techniques. We used a voice recorder to capture students’ pronunciation. The data were analysed using descriptive analysis and t-test. The results of the study show that the application of accent reduction software as a medium for teaching English pronunciation can improve students’ skills in English pronunciation (t = 3.538, tcalculate> ttable). This software helps students produce English words clearly and easily because they can hear and imitate the sounds it generates. The implication of using the software can help students to imitate native speaker pronunciation and teachers can reduce their efforts in pronouncing native English sounds because the software functions as a pronunciation learning aid

Menggali Akar Permasalahan: Konflik Antara Tukang Ojek Konvensional Dan Tukang Ojek Online Di Stasiun Jember

Ojek is a transportation service that has long been the choice of people to travel. Along with the development of technology, currently ojek develops into two types, namely conventional motorcycle taxis and online motorcycle taxis.However, in recent times conventional motorcycle taxis have begun to face a decline in popularity along with the emergence of online motorcycle taxis. One of them is a conventional motorcycle taxi in front of Jember station. So this causes conflicts due to passengers preferring to use online motorcycle taxis compared to conventional motorcycle taxis. The reason they choose online motorcycle taxis is because what is offered is considered more economical and sometimes offers promos that make online motorcycle taxi fares cheaper than conventional motorcycle taxis. This certainly makes conventional motorcycle taxis experience a decrease in income. This study used qualitative research methods, where data were collected through field observations, interviews, and voice recordings, and refer to previous studies. This study uses Dahrendorf conflict theory to analyze conflicts that arise between conventional motorcycle taxis and online motorcycle taxis in the Jember station area, where differences in status and access rights set by local authorities are the main focus in this study.

Using Wearable Devices and Speech Data for Personalized Machine Learning in Early Detection of Mental Disorders: Protocol for a Participatory Research Study.

Early identification of mental disorder symptoms is crucial for timely treatment and reduction of recurring symptoms and disabilities. A tool to help individuals recognize warning signs is important. We posit that such a tool would have to rely on longitudinal analysis of patterns and trends in the individual's daily activities and mood, which can now be captured through data from wearable activity trackers, speech recordings from mobile devices, and the individual's own description of their mental state. In this paper, we describe such a tool developed by our team to detect early signs of depression, anxiety, and stress. This study aims to examine three questions about the effectiveness of machine learning models constructed based on multimodal data from wearables, speech, and self-reports: (1) How does speech about issues of personal context differ from speech while reading a neutral text, what type of speech data are more helpful in detecting mental health indicators, and how is the quality of the machine learning models influenced by multilanguage data? (2) Does accuracy improve with longitudinal data collection and how, and what are the most important features? and (3) How do personalized machine learning models compare against population-level models? We collect longitudinal data to aid machine learning in accurately identifying patterns of mental disorder symptoms. We developed an app that collects voice, physiological, and activity data. Physiological and activity data are provided by a variety of off-the-shelf fitness trackers, that record steps, active minutes, duration of sleeping stages (rapid eye movement, deep, and light sleep), calories consumed, distance walked, heart rate, and speed. We also collect voice recordings of users reading specific texts and answering open-ended questions chosen randomly from a set of questions without repetition. Finally, the app collects users' answers to the Depression, Anxiety, and Stress Scale. The collected data from wearable devices and voice recordings will be used to train machine learning models to predict the levels of anxiety, stress, and depression in participants. The study is ongoing, and data collection will be completed by November 2023. We expect to recruit at least 50 participants attending 2 major universities (in Canada and Mexico) fluent in English or Spanish. The study will include participants aged between 18 and 35 years, with no communication disorders, acute neurological diseases, or history of brain damage. Data collection complied with ethical and privacy requirements. The study aims to advance personalized machine learning for mental health; generate a data set to predict Depression, Anxiety, and Stress Scale results; and deploy a framework for early detection of depression, anxiety, and stress. Our long-term goal is to develop a noninvasive and objective method for collecting mental health data and promptly detecting mental disorder symptoms. DERR1-10.2196/48210.

Excavation of the Suspect's Identity by Analysis of the Speaker Profiling Case Evidence of Fake Voice Recordings

Technological developments, especially in AI applications, allow someone to commit criminal acts by changing other people's voices. This study implements forensic linguistics by comparing data sources on the audio of Lesti (L) and Rizky Billar (RB) to ensure the originality of the sounds. The Speaker Profiling method with aural-perception analysis is used because there is no valid comparative evidence, so the audio comparison is taken from the L and RB platforms published to the public. This research resulted in two different aspects of each audio: phonetics and acoustics. These two aspects show the two audio sources' differences in pronunciation, emphasis, frequency, and tone. As a result, the vote is identified as a hoax. This research can demonstrate the use of software to help identify individual language profiles; it contributes to the development of applied linguistics, especially as a legal aid tool. Researchers hope that there will be further research from other linguists that will address more benefits of using AI / Corpus technology as a linguistic analysis tool, both with spoken and written data.

Familiarity and understanding of assistive technology among visually impaired users of libraries of special institutes of Haryana, India

PurposeAssistive technology (AT) is like the eyes of visually impaired users (VIUs) and plays a vital role in making them independent in accessing a wide range of library resources and services. It is a revolutionary way to improve the academic performance of VIUs and amalgamate them into mainstream library services. This paper aims to explore the familiarity and understanding of AT among VIUs in accessing library resources and services of three special institutes of Haryana state, India.Design/methodology/approachThe study used a survey method, and data were collected using an online questionnaire. The questionnaire consisted of closed-ended questions focused on participants’ familiarity, understanding and key barriers VIUs face in accessing existing AT facilities. Scope of the study is to include three special institutions’ libraries serving to VIUs in Haryana (India). Purposive sampling technique was adopted for data collection. The data are tabulated and analyzed with descriptive statistics methods using Statistical Package for Social Sciences (SPSS) software version 20.0, and the results are presented in tables, diagrams and graphs.FindingsThe study found that most of the VIUs are highly familiar and reported an expert understanding of Daisy player and digital voice recorder, whereas refreshable Braille display, Braille printer/embosser and scanning and reading devices are least familiar with and reported poor understanding. In the software category, it is noted that VIUs are more familiar with screen reading and speech synthesizer software than other software, and participants confirmed a good understanding of the said software.Practical implicationsThe literature review revealed that the majority of library staff in the institutions being investigated had limited knowledge about the availability of the most suitable AT tools. So, the study will be beneficial for them to select the most popular AT tools for their libraries to amalgamate VIUs. The study also highlights some barriers VIUs face in using the existing AT facility of their respective libraries that may be helpful for librarians to understand the root cause of low utilization of available AT tools and how to address this problem effectively. The study can also help policymakers in formulating AT policies for academic libraries. Findings will also help to develop a cost-effective AT workstation suiting the needs of VIUs.Originality/valueThis study will provide an idea to library professionals about which AT hardware and software are more popular among VIUs and useful for developing/improving cutting-edge AT facilities in their libraries to amalgamate VIUs in the mainstream of the library.

An Acoustic Study of the Stop Consonants in Lettered-Words Produced by Chinese Mandarin Speakers

This paper presents key findings from an acoustic study on Chinese Mandarin Speakers’ realization of the stop consonants in lettered-words and the stop consonants in Chinese phrases. The research was conducted through voice recording and analysis of data collected from Chinese Mandarin speakers. Recordings of 2000 overall tokens were collected and analyzed using Praat phonetic software. The mean VOT values obtained from both lettered-words and Chinese phrases were compared and statistically tested using the Independent Samples t-test. The research findings show that first, the stop consonants in lettered-words produced by Chinese Mandarin speakers are all voiceless stop consonants, distinguished as being [+aspirated] and [-aspirated]. Second, the VOT values of [p] [t] [k] in lettered-words are significantly lower than their counterparts in Chinese phrases. Third, the stop consonants in lettered-words have the same stop pattern as those in Chinese phrases. The research presents proof that where there is phonemic similarity (but phonetic dissimilarity) across Chinese Mandarin and English, L1 phonetic properties are strong for Chinese Mandarin speakers to produce lettered-words in L1 environment, which further confirms former scholars’ conviction of mother tongue interference in L2 learning. Similar phonemes in L1 and L2 are realized identically to L1 sounds, and there is significant interference from the speaker’s native language on phonetic properties produced in L2. In addition, the research presents implication for future research that might explore other acoustic features of lettered-words in Chinese Mandarin.

An Online Attachment Style Recognition System Based on Voice and Machine Learning.

Attachment styles are known to have significant associations with mental and physical health. Specifically, insecure attachment leads individuals to higher risk of suffering from mental disorders and chronic diseases. The aim of this study is to develop an attachment recognition model that can distinguish between secure and insecure attachment styles from voice recordings, exploring the importance of acoustic features while also evaluating gender differences. A total of 199 participants recorded their responses to four open questions intended to trigger their attachment system using a web-based interrogation system. The recordings were processed to obtain the standard acoustic feature set eGeMAPS, and recursive feature elimination was applied to select the relevant features. Different supervised machine learning models were trained to recognize attachment styles using both gender-dependent and gender-independent approaches. The gender-independent model achieved a test accuracy of 58.88%, whereas the gender-dependent models obtained 63.88% and 83.63% test accuracy for women and men respectively, indicating a strong influence of gender on attachment style recognition and the need to consider them separately in further studies. These results also demonstrate the potential of acoustic properties for remote assessment of attachment style, enabling fast and objective identification of this health risk factor, and thus supporting the implementation of large-scale mobile screening systems.

Smartphone-Based Voice Wellness Index Application for Dysphonia Screening and Assessment: Development and Reliability.

This study aimed to develop a Voice Wellness Index (VWI) application combining the acoustic voice quality index (AVQI) and glottal function index (GFI) data and to evaluate its reliability in quantitative voice assessment and normal versus pathological voice differentiation. Cross-sectional study. A total of 135 adult participants (86 patients with voice disorders and 49 patients with normal voices) were included in this study. Five iOS and Android smartphones with the "Voice Wellness Index" app installed were used to estimate VWI. The VWI data obtained using smartphones were compared with VWI measurements computed from voice recordings collected from a reference studio microphone. The diagnostic efficacy of VWI in differentiating between normal and disordered voices was assessed using receiver operating characteristics (ROC). With a Cronbach's alpha of 0.972 and an ICC of 0.972 (0.964-0.979), the VWI scores of the individual smartphones demonstrated remarkable inter-smartphone agreement and reliability. The VWI data obtained from different smartphones and a studio microphone showed nearly perfect direct linear correlations (r=0.993-0.998). Depending on the individual smartphone device used, the cutoff scores of VWI related to differentiating between normal and pathological voice groups were calculated as 5.6-6.0 with the best balance between sensitivity (94.10-95.15%) and specificity (93.68-95.72%), The diagnostic accuracy was excellent in all cases, with an area under the curve (AUC) of 0.970-0.974. The "Voice Wellness Index" application is an accurate and reliable tool for voice quality measurement and normal versus pathological voice screening and has considerable potential to be used by healthcare professionals and patients for voice assessment.