Voice Users Research Articles

Acoustic-Perceptual Correlates of Voice Among Steam Train Engineers: Effects of Noise and Hearing Protection

Occupational voice users are at a higher risk for developing voice disorders due to their vocal demands, such as prolonged periods of work-related voice use and nonideal environmental factors, such as speaking above background noise. The current study focused on the effects of background noise and hearing protection on acoustic-perceptual correlates of voice among steam train engineers. Fourteen participants phonated vowel /a/, read a phrase, and described a map under different noise and hearing protection conditions. Relative sound pressure level, relative fundamental frequency, and perceived vocal effort and disturbance decreased in the presence of hearing protection for all noise conditions. In contrast, these acoustic measures increased in the absence of hearing protection supporting Lombard effect. Overall, results of the current study provide insight into possible risks to vocal health in workers exposed to high levels of background noise and use hearing protection.

Privacy‐preserving speaker verification system based on binary I‐vectors

Speaker verification is a key technology in many services and applications like smartphones and intelligent digital assistants. These applications usually require users to transmit their recordings, features, or models from their voices over untrusted public networks which stored and processed them on cloud-based infrastructure. Furthermore, the voice signal contains a great deal of the speaker's personal and private information which raises several privacy issues. Therefore, it is necessary to develop speaker verification systems that protect the user's voice against such threats. Herein, the cancellable biometric systems have been introduced as a privacy-preserving solution. A cancellable method for speaker verification systems is proposed using speaker i-vector embeddings. This method includes two stages: (i) i-vector binarisation and (ii) the protection of the binary i-vector with a shuffling scheme derived from a user-specific key. Privacy evaluation of this method according to the standards of biometric information protection (ISO/IEC 24745) shows that the proposed cancellable speaker verification system achieves the revocability, unlinkability, and irreversibility requirements. Moreover, the cancellable system improves biometric performance compared with the unprotected system and makes it resistant to different attack scenarios. Additionally, we demonstrate that this method can also operate to protect deep neural network speaker embeddings such as x-vectors.

Prevalence of voice disorders in tertiary care hospital

<p class="abstract"><strong>Background:</strong> Human voice is unique due to laryngeal configuration. However, change in voice is common with age and gender. It can also be altered by vocal abuse and misuse leading to different types of voice disorders. So, the present study aims to study the prevalence of voice disorder in patients visiting the voice clinic ENT-HNS department of Tribhuvan University Teaching Hospital tertiary care center. </p><p class="abstract"><strong>Methods:</strong> This was a retrospective study carried out among the patients visiting voice clinic in the department of ENT-HNS for a duration of one year. The patients were evaluated for voice disorder by a team of ENT doctors and Speech pathologists using perceptual and instrumental evaluation. The patients who required surgical intervention were excluded. A descriptive study was done among patients with voice disorders requiring voice therapy. </p><p class="abstract"><strong>Results:</strong> Out of 296 patients, the most common age group was 25-50 years with more females compared to males. Structural cause for voice disorder was most common in all age groups and gender in both professional and non-professional voice users but it was more prevalent in professional voice users.</p><p class="abstract"><strong>Conclusions:</strong> Voice disorders were more frequent in females than males and also in professional voice users as they tend to use voice more daily. The structural cause for voice disorder is the major cause of voice problems in all age groups. </p>

Do Voice-Over Artists Convey Emotion Better Than Untrained Voice Users?

ABSTRACT Voice is an invaluable asset for vocal professionals such as actors, singers, dubbing artists, etc. Among these, voice-over artists have rarely been the focus of research related to voice in India. These professionals need to convey emotions solely through the voice, and their vocal training must be suited to meet these demands. This study compares vocal capability of voice-over artists to those of untrained voice users. Each of 10 participants (five voice-over artists, five untrained vocal users) were asked to verbalize a target sentence to convey four different emotions. Their voice samples were recorded and given to 30 naive listeners to investigate whether the listeners perceived the emotion being conveyed. Results revealed that although voice-over artists were recognized by listeners as trained vocal users, there was no statistically significant difference in the number of sentences where the emotion was correctly conveyed through their voice, as outlined in the study. Among the emotions studied, “excitement” was maximally perceived correctly by listeners. While highlighting a lack of training by vocologists, the study also brings into focus a relatively neglected sub-population of vocal professionals in India.

Real-time, Robust and Adaptive Universal Adversarial Attacks Against Speaker Recognition Systems

Voice user interface (VUI) has become increasingly popular in recent years. Speaker recognition system, as one of the most common VUIs, has emerged as an important technique to facilitate security-required applications and services. In this paper, we propose to design, for the first time, a real-time, robust, and adaptive universal adversarial attack against the state-of-the-art deep neural network (DNN) based speaker recognition systems in the white-box scenario. By developing an audio-agnostic universal perturbation, we can make the DNN-based speaker recognition systems to misidentify the speaker as the adversary-desired target label, with using a single perturbation that can be applied on arbitrary enrolled speaker’s voice. In addition, we improve the robustness of our attack by modeling the sound distortions caused by the physical over-the-air propagation through estimating room impulse response (RIR). Moreover, we propose to adaptively adjust the magnitude of perturbations according to each individual utterance via spectral gating. This can further improve the imperceptibility of the adversarial perturbations with minor increase of attack generation time. Experiments on a public dataset of 109 English speakers demonstrate the effectiveness and robustness of the proposed attack. Our attack method achieves average 90% attack success rate on both X-vector and d-vector speaker recognition systems. Meanwhile, our method achieves 100 × speedup on attack launching time, as compared to the conventional non-universal attacks.

Perancangan Voice User Interface(VUI) Aplikasi Presensi Karyawan Dengan Speech Recognition

As we all know, the very easy and fast spread of COVID-19 can threaten human life. This virus is easily spread from inanimate objects that are touched by sufferers. The use of fingerprint attendance machines in the workplace can be a source of the spread of COVID-19, this needs to be considered for other alternatives to record employee attendance, for example with speech recognition technology or known as automated Speech Recognition (ASR). This technology is needed as an important component to build a Voice user interface (VUI). ASR is a technology for translating speech from the user into text. In addition, the use of ASR can improve communication between human-computer interactions, that is, users can interact with computers using voice. This research is aimed at designing prototype employee attendance application with the VUI concept and speech recognition technology. To speed up design and development process, researchers used the RAD (Rapid Application Development) method. From the results of research using the RAD method, it can be concluded that the application can be designed quickly through prototype modeling together with the user.

VoLTE problem location method based on big data

This paper mainly introduces the key information extracting method from User Equipment (UE) test data, the synchronous algorithm of various interfaces X Detailed Record (XDR) data, the key information filling technique of multi interfaces X Detailed Record (XDR), the analytical procedure based on big data is descript in details. An example is given to explain that more accurate reason for Voice over Long-Term Evolution (VoLTE) call not connected can be found with big data from core network, compared with UE data only. With this method VoLTE problem can be quickly solved, it is very useful for wireless operator to improve the user voice service awareness rapidly.

서버관리 시스템을 위한 명령형 음성 사용자 인터페이스 설계 및 구현

Voice interface technology is a technology that enables dialogue between machines and humans by using ‘words’, one of the most natural means of communication for humans. One of the advantages of using the voice interface when entering commands is that it can be used in parallel with the keyboard interface. Therefore, usability can be improved by supporting a voice user interface in softwares running on computer servers or PCs. In this study, a command-and-control interface was designed and implemented for a multi-server management system that manages multiple servers. The command input mode used voice and keyboard, and the information output mode used the screen and beep interface. The voice recognizing software can be run in a web browser and uses the open source Annyang library. The results of this study are expected to be utilized in the design of voice user interfaces for application software, facility management software, and IoT device monitoring systems for professional users.

Gender differences in vocal doses among occupational voice users: a systematic review of literature and meta-analysis

Background Occupational voice users such as teachers, singers, and broadcasters have high vocal demands increasing the risk for developing voice disorders. Among occupational voice users, the literature has report gender differences in vocal doses as part of vocal demands. However, these differences have not been quantified. Objective To determine differences per gender on vocal doses among different groups of occupational voice users. Methods A systematic review of literature and meta-analysis were conducted on publications about vocal dose among occupational voice users. In total, 242 potential articles were found. After screening of titles and abstracts, 16 papers were included for full revision in the systematic review of literature. For the meta-analysis, 7 out of 16 papers were included. Results Females had higher phonation time percentage compared with males (mean difference = 1.44, p value = .16). Nevertheless, this difference was not statistically significant. Among call centre operators, there was no difference in time doses associated with gender. Sport teachers had high values in time dose percentage, dissipation, and radiated energies. Conclusions The meta-analysis shows variations in time doses, which could be associated with the vocal demand responses for females compared with males in occupational voice settings. Future investigations are required for establishing safe limits criteria for vocal doses, as well as vocal functioning in different working context.

Vocal Fatigue Index in Teachers Using Mokken Analysis

Vocal fatigue (VF) is a primary vocal symptom experienced by professional voice users, such as teachers, whose voice is an occupational tool. The study determines the utilization of the Vocal Fatigue Index (VFI), a 19-item scale in identifying symptoms of VF and its severity in teachers. Using responses of 695 teachers who completed the VFI, Mokken scaling was conducted on the items to identify the experiences of VF and its associated hierarchical nature of VF symptoms in teachers. Mokken scaling was completed on a total of four groups: (a) Total teachers group, (b) No VF group, (c) Low VF group, and (d) High VF group. Results revealed differences in item hierarchies between total teachers and across the separate groups of VF severity. Item hierarchy for teachers highlighted items from physical discomfort at the mild end of the hierarchy to items from symptom improvement with rest at the severe end of the hierarchy. Items related to avoidance presented as a separate scale for teachers presenting with high VF. Mokken scaling in teachers provides insight into the underlying complexity of the experience of VF symptoms and reliance on differential behavioral strategies in its management, suggesting the heterogenous nature of latent trait for VF in this specific population.

Integrating User Voice in Hearing Care With Focus on Off-Duty Warfighter.

Noise-Induced Hearing Loss (NIHL) is a growing public health concern in the USA and globally because of the emergence of lifestyle preferences and environmental exposures to sound levels exceeding safe listening limits for extended periods of time. In the military, hearing and communication are important for survival and the Department of Defense's Hearing Center of Excellence highlights the importance of protection from NIHL. Issuance of the World health organization safe audio listening standards along with existing U.S. federal and military standards provides a framework for developing an accessible tool for promoting safe listening. An App is being proposed for an aggregated assessment of a user's daily sound exposure, through personal audio system and ambient sources and providing tools for managing personal sound exposures. The key elements for development of App features were integration of user voice for functionalities, computations based on global standards, including military, for safe listening and alignment with regulatory standards for Apps. Features of the App are "Track" for audio and ambient exposures, "Compute" for real-time and cumulative exposures vs. safe listening standards, "Alert" for unsafe exposures using visual and haptic, "Inform" for personal exposures vs. safe dose, associated NIHL risk and education, "Choice" for alternatives that can be relatable to personal lifestyle, and "Share" for informing others as determined by the user. App safety is ensured by privacy and cybersecurity regulatory standards. The overall goals of the App are to increase NIHL awareness and to empower users to improve personal listening behaviors that can potentially reduce the risk of NIHL.

Effects of Mechanical Vibration Stimulation of the Larynx on Voice Production

Combining mechanical vibration stimulation with resistance training to improve neuromuscular performance is a method that has gained popularity in the field of sports sciences. To evaluate the immediate and mid-term effects of mechanical vibration stimulation (G-Life NL3000) of the larynx associated to vocal exercises. Fourteen professional voice users divided into: Experimental Group - Seven participants who performed a 20-minute exercise protocol with humming applying mechanical vibration stimulation (G-Life NL3000) of the larynx once a day for 30 days; Control Group - Seven participants who performed only the 20-minute exercise protocol with humming. For comparison purposes, the evaluation was carried out pre-stimulation, immediately after simulation (post) and 30 days after stimulation (late-post). The Experimental Group showed a statistically significant progress in the post-immediate evaluation and in the post-treatment evaluation for maximum phonation time (*P=< 0.001), Jitter (*P=0.008), Shimmer (*P=0.005), noise-to-harmonics ratio (*P=0.005). In the GRBASI scale, the Experimental Group showed significant results in the post-treatment evaluation for Grade (*P=0.005), Roughness (*P=0.008), Breathiness (*P=0.018) and Instability (*P=0.018). A self-perception voice questionnaire showed better voice quality (*P= 0.031) and easier phonation (*P=0.005) in the Experimental Group. Mechanical vibration stimulation of the larynx associated to humming produced positive changes in the vocal pattern and significantly improvement in short and mid-term vocal quality for participants in this study.

Testing Hirschman’s exit, voice, and loyalty model: citizen and provider responses to decline in public health services

Dissatisfied users of public services may choose to voice or exit. But when does voice emerge? To answer this question, we deploy Hirschman’s exit, voice, and loyalty (EVL) model. We set out an ‘available alternatives’ hypothesis—increasing the number of exit options reduces voice—in contrast to an ‘effective voice’ hypothesis where voice is lower when there is no choice. We expect the exit-voice relationship to be moderated by loyalty; and providers that respond to user voice improve satisfaction and reduce exit. We evaluate these hypotheses in a survey experiment on publicly funded doctors’ services in the UK. We find no effects of the number of exit options on voice, nor evidence for loyalty as a moderator. A response by the provider results in higher satisfaction and lower intention to exit, strengthened by loyalty. Providers can promote satisfaction and discourage users moving to alternative providers by responding to voice.

How governments manage personal assistance schemes in response to the United Nations Convention on the Rights of Persons with Disabilities: A Scoping Review

Governments are obligated to safeguard social inclusion for disabled people through user-led personal assistance (PA) under Article 19 of the United Nations Convention on the Rights of Persons with Disabilities (UNCRPD). This scoping review was carried out to map and explore current knowledge on how governments internationally have managed PA schemes in response to the UNCRPD. The review examined 99 documents, and categorised the literature into the following themes; legislation, funding, model of service provision, governance and regulation, and the COVID-19 pandemic response. We include recommendations to co-design legislation and quality improvement policies to ensure that PA schemes are underpinned by a social model of disability mindset. Further research needs to be undertaken to guarantee that policymakers include the voice of PA users in the management of PA schemes. Points of interest This article looked at 99 documents to find out how governments are managing personal assistance (PA). It found out that governments can often decide to spend less money on a PA scheme rather than protect our rights. To overcome this problem the documents recommended that legislation for PA schemes must be designed with disabled people. Governments must redirect their money from institutional services to community-based services. Eligibility criteria to control access and the costs of PA should be removed. This paper suggests that we need to have the voice of the PA user to direct the design and delivery of PA schemes. Supplemental data for this article is available online at https://doi.org/10.1080/09687599.2021.1877114.

What does safety in mental healthcare transitions mean for service users and other stakeholder groups: An open-ended questionnaire study.

BackgroundHistorically, safety mental health research has tended to focus on risks of homicide, suicide and deaths. Although wider safety issues are now recognized in regards to mental health services, the safety of mental health transitions, a key research and policy priority according to World Health Organisation, has not been explored.ObjectiveThe purpose of this study was to investigate perceptions of safety in mental health transitions (hospital to community) amongst five stakeholder groups.Design and settingAn online, international cross‐sectional, open‐ended questionnaire.ParticipantsThere were five stakeholder participant groups: service users; families/carers; mental health‐care professionals; researchers; and end users of research.ResultsNinety‐three participants from 12 different countries responded. Three overarching themes emerged: ‘individual/clinical’, ‘systems/services’ and ‘human, behavioural and social’ elements of safe mental health transitions. Whilst there was a great focus on clinical elements from researchers and healthcare professionals, service users and carers considered safety in terms of human, behavioural and social elements of transitional safety (ie loneliness, emotional readiness for discharge) and systems/services (ie inter‐professional communication).DiscussionSafety in mental health‐care transitions is perceived differently by service users and families compared to healthcare professionals and researchers. Traditional safety indicators for care transitions such as suicide, self‐harm and risk of adverse drug events are raised as important. However, service users and families in particular have a much wider perception of transitions safety.ConclusionFuture quality and safety research and policy should consider including a service user voice and consider integration of psychosocial elements in discharge interventions.

Post-lingual deaf adult cochlear implant users' speech and voice characteristics: cochlear implant turned-on versus turned-off

Background A time-period of turning off cochlear implant (CI) might affect the speech and voice features of the CI users due to loss of auditory feedback monitoring. Objectives To evaluate Persian-speaking CI users' speech and voice parameters in a device-on/off experiment. Material and methods Forty-eight Persian-speaking adults with post-linguistic deafness who had received a unilateral CI and 50 normally-hearing counterparts were recruited. All CI users had to fulfill three phases of speech sampling, including the device-on, device-off (a 5-hourly period), and once again device-on phase. Their Fundamental frequency, Jitter, Shimmer, Harmonics-to-noise ratio, hyper-nasality score, and the percentage of speech intelligibility were assessed for both groups. Results The findings revealed that the CI-on/off conditions significantly changed cochlear implanted participants' F0, Shimmer, HNR, and hyper-nasality values. However, no statistical differences were observed in Jitter and the percentage of speech intelligibility between the CI and control groups. Conclusions A short-term auditory deprivation in CI recipients could degrade the speech and voice quality in CI users, so that could change the quality of their speech resonance into hyper-nasality. However, when the CIs were turned on again, the quality of their speech and voice returned to normal-sounding.

Vocal fatigue in teachers and non-teachers in a Turkish population

Objectives: The aim of this study was to determine the voice use and associated vocal fatigue in teachers and to define their differences with other professional voice users and non-vocal non-professionals. Patients and Methods: Between May 2020 and October 2020, a total of 187 participants (41 males, 146 females; mean age: 32.6±10.5 years; range, 23 to 65 years) were administered the Vocal Fatigue Index (VFI) questionnaire. Of the participants, 93 were teachers and 94 were non-teachers recruited from Istanbul province of Turkey. The relationship between the VFI and sociodemographic characteristics, professional activity, talkativeness, duration of active vocal use, and active vocal complaints were analyzed. Results: The mean VFI score was 35.5±16.2. Those with a vocal complaint had significantly higher mean total VFI scores than those without any vocal complaints (p&lt;0.001). Duration of voice use and total VFI scores were significantly higher in teachers than other professions (p&lt;0.001 and p&lt;0.001, respectively). The mean total VFI scores of teachers with or without vocal complaints did not significantly differ (p=0.065). Conclusion: Our study results suggest that voice disorders are more common among teachers, compared to non-professional voice users. It is important to reveal the etiology of voice problems in teachers, particularly for early diagnosis and immediate treatment.

A scoping review on dental clinic accessibility for people using wheelchairs.

We aimed to explore the scientific literature on accessible dental clinics for wheelchair users. More specifically, we sought out literature addressing how the human environment and physical space shape the dental services of accessible dental clinics. We conducted a scoping review (May 2019) in Embase, PubMed, Web of Science, and the Avery index of architectural Periodicals (3994 articles). We followed Arksey and O'Malley's recommended procedures; after screening, we retained 17 articles. We performed a critical appraisal, followed by thematic content analyses of extracted data. The articles originated mainly from the United States and United Kingdom. Only three reported original research. We illustrated the results within a three-step dental care pathway cycle. In each step, the interaction between accessibility of the physical and human environments (ie, the layout/design of the clinic and the attitudes and skills of the dental professional, respectively) contributed to the overall accessibility. We also found that empirical evidence on clinics' accessibility was lacking: many articles resorted to broad "one size fits all" recommendations and fragmented information on accessibility. Finally, the voice of wheelchair users was missing. There are knowledge gaps in terms of dental clinics' accessibility. We thus invite researchers to conduct original studies with dental professionals, wheelchair users, and their caregivers.

Design Space for Voice-Based Professional Reporting

Speech technology has matured so that voice-based reporting utilizing speech-to-text can be applied in various domains. Speech has two major benefits: it enables efficient reporting and speech input improves the quality of the reports since reporting can be done as a part of the workflow without delays between work and reporting. However, designing reporting voice user interfaces (VUIs) for professional use is challenging, as there are numerous aspects from technology to organization and language that need to be considered. Based on our experience in developing professional reporting VUIs with different stakeholders representing both commercial and public sector, we define a design space for voice-based reporting systems. The design space consists of 28 dimensions grouped into five categories: Language Processing, Structure of Reporting, Technical Limitations in the Work Domain, Interaction Related Aspects in the Work Domain, and Organization. We illustrate the design space by discussing four voice-based reporting systems, designed and implemented by us, and describing a design process that utilizes it. The design space enables designers to identify critical aspects of professional reporting VUIs and optimize those for their target domain. The design space can be used as a practical tool especially by designers with limited experience on speech technologies.

In Pursuit of Sustainable Mobile Policy: A Study of Consumer Tariff Preferences under Uncertainty

Literature suggests that consumers expect disutility not only from payment uncertainties but also from reference uncertainties embedded in mobile plans. This paper develops a model of mobile plan choice incorporating both reference and payment uncertainties and uses this model to derive testable implications. The paper argues that consumer choice reflects those uncertainties more than could be justified by rational choice theory. Such patterns, the paper hypothesizes, would be more salient in the choice of data plan than voice plan because consumers tend to perceive data usage to be less controllable than voice usage, thus preferring the plan that reduces uncertainty. The paper tests the predictions with data from a laboratory study analyzing a series of choices between plans with different tariff structures—flat-rate, two-part, and three-part tariffs. As predicted, the results suggest that payment and reference uncertainties create significant disutility for consumers, especially when they perceive high uncertainty about their usage. Such understanding of consumer preference and underlying psychological biases is important in the sense that it provides an essential basis for the development of sustainable mobile policy.