Voice Users Research Articles

Enhancing Accessibility: Assistive Devices Pairing Mobile Apps for Visually Impaired People

Vision impairment and blindness are significant health conditions that affect a considerable portion of the global population and have significant consequences, particularly in developed nations. Its harmful consequences can have a negative influence on a person's personal quality of life, as well as limit their ability to live independently. Navigation plays a crucial role in the activities that are impacted by vision loss. This is because it is almost impossible for the individual to move safely and independently. The goal of this project is to improve mobility for visually impaired people by employing Light Detection and Ranging (LiDAR) sensors to identify obstacles and alert users via haptic feedback if there are potential obstacles in their way. In addition, this study compares two sensors, LiDAR and ultrasonic sensors, besides the implementation of the Internet of Things (IoT), which was integrated into the small, portable robot by pairing a mobile phone. This is to ensure the user may be securely guided to their destination. I-Walk is a mobile app with a voice user interface (VUI) for GPS navigation systems and emergency calling. This app has been developed to allow users safe and independent travel. The simulation result shows that even though ultrasonic sensors are cheaper and serve the same function as LiDAR, their huge standard deviation and percentage error shown in this study cannot equal LiDAR's accuracy in which LiDAR has an average error of 0.63%. In contrast, ultrasonic sensors have an average error of 2.39%. Overall, this obstacle detection and navigation system comprises robust hardware and software to assist visually impaired people effectively.Keywords: Assistive technology, blind, mini guider robot, obstacle detection, navigation, IoT

Bridging the Gap Between Designers and Users

In this paper, we describe our approach to bridging the designer-user gap by engaging 20 older adults in codesign workshops to generatively re-imagine health-related voice agent (VA) design for voice user interfaces (VUI) for older adults, by older adults. Drawing on recent research in Library and Information Science (LIS), we describe codesign as a method, as well as how the 20 older adults perceived codesign in our two case studies of health-related VA design for health learning and exercise. Older adult codesigners perceived codesign as transformative, enjoyable, educational, and useful. We also share specific experiential recommendations of strategies for conducting codesign, including strategies for transitioning from expert to facilitator, creating experience-based questions, probing for deeper design ideation and priorities, and ways of approaching codesign through humility and care. We describe the codesign method’s generative benefits in our two case studies and encourage LIS researchers to consider including codesign methods in their research.

IoT based home automation with energy management

Automation of home appliances has become the prime utility of embedded systems. This automation method is inclusive of a sensor-based automated system that requires no human/conventional interventions. This paper proposes the usage of voice commands to have control over the entire appliances, which is easy to handle by old age/disabled people. The major aspect of this paper is to introduce a new system for disabled and normal people. This method involves a Raspberry-Pi-3 control board, which has a WiFi module termed Raspberry-Pi-3. For sending the voice instruction to the Raspberry-Pi-3 a mobile application is used. The function of the application is to record the voice and convert the voice note into a command for Raspberry-Pi-3. This process is facilitated by WiFi communication. This system also uses IoT for measuring the power consumed by the active appliance over the current sensor and the Webserver can preview the power consumption.

Metaphors in Voice User Interfaces: A Slippery Fish

We explore a range of different metaphors used for Voice User Interfaces (VUIs) by designers, end-users, manufacturers, and researchers using a novel framework derived from semi-structured interviews and a literature review. We focus less on the well-established idea of metaphors as a way for interface designers to help novice users learn how to interact with novel technology, and more on other ways metaphors can be used. We find that metaphors people use are contextually fluid, can change with the mode of conversation, and can reveal differences in how people perceive VUIs compared to other devices. Not all metaphors are helpful, and some may be offensive. Analyzing this broader class of metaphors can help understand, perhaps even predict problems. Metaphor analysis can be a low-cost tool to inspire design creativity and facilitate complex discussions about sociotechnical issues, enabling us to spot potential opportunities and problems in the situated use of technologies.

Womens Discourse Dilemma from the Perspective of Social Media: Taking Reports and Comments on Social Events on Weibo as an Example

The emergence of the new media period has transformed social media into a public forum for the dissemination of public information, and the enhancement of user voice and demand has given feminism a new context and given women a free platform to express themselves. However, the ability of women to talk freely still faces significant restrictions as a result of the long-lasting effects of history and culture as well as the bias of the news media system. This essay attempts to explain the current situation of the absence of womens voices on the platform by using Weibo, the dominant force of Chinese social platforms, as the research object. The research method of case analysis is used to broaden the analysis based on the facts and material already available. The research results emphasize the plight of womens lack of voice in society, and put forward some relevant countermeasures from three different perspectives: media, women themselves, and social culture, to encourage the growth of womens voice on social platforms and ultimately achieve gender equality.

Vocal Cord Monitoring by Flexible Fiberoptic Laryngoscopy During Thyroid Radiofrequency Ablation.

Introduction: Thermal injury to recurrent laryngeal nerve (RLN) during radiofrequency ablation (RFA) can produce temporary or permanent vocal cord paralysis.1 Hydrodissection with cold 5% glucose of "danger triangle" protects RLN during RFA.2 When RFA is performed under local anesthesia, RLN function is monitored by patients producing vocal sounds.3 Large lesions requiring longer RFAs warrant general sedation where voice cannot be assessed, therefore, an additional technique for RLN protection is advisable. Observation of passive symmetrical vocal cord movements during breathing by laryngeal ultrasonography is useful in assessing vocal cord function4; however, flexible-fiberoptic fibrolaringoscopy (FFL) is gold standard for assessing vocal cord movements,5 anticipating potential RLN damage. We report FFL monitoring during RFA under general sedation on a large thyroid nodule. FFL during RFA may detect RLN irritation and dysfunction if asymmetry in passive vocal cord movements is noted. Should asymmetry appear, RFA operator stops delivering energy and repositions electrode needle. Materials and Methods: Thyroid function tests, blood glucose, creatinine, transaminase, International-Normalized-Ratio, and electrocardiogram were performed. Operating room (OR) layout created sufficient space for ear-nose-throat (ENT) and RFA operators. An examination with a fiberscope camera demonstrated normal vocal cord adduction during phonation and abduction during breathing. The procedure was assisted by an anesthetist administering fentanyl 50 mcg, midazolam 1.5 to 5.0 mg, and propofol infusion 2 mg/(kg·h). General sedation was conducted so that reflexes were attenuated but still observable. Incorporating in OR by an anesthetist who performs general sedation reduces side effects and complications.6 Ultrasonography showed a 34-mL right lobe nodule abutting on the RLN area. After sedation with propofol, the ENT specialist inserted an endoscope until the glottic plane. During calm breathing, vocal cords moved symmetrically. After obtaining anterior nodule hydrodissection from strap and sternocleidomastoid muscles with 10 mL of 2% lidocaine, posterior hydrodissection was achieved by ultrasound-guided administration of 30 mL of 5% cold glucose. Anterior and posterior hydrodissections merged, separating nodule from neck structures. The radiofrequency electrode needle was then inserted into the nodule, initially positioned in inferior nodule portion adjacent to danger triangle previously isolated by hydrodissection. Initial power was 30 watts. Moving-shot technique was used. Results: FFL was performed throughout thyroid RFA. Symmetric vocal cord movements during breathing demonstrated no RLN irritation. FFL monitoring allowed observation of natural reflexive phenomena, including swallowing. Complete nodule ablation was achieved. FFL performed post-RFA confirmed normal vocal cord motility. Conclusions: We report the first-time use of FFL for vocal cord monitoring during RFA. FFL was easily performed by the ENT specialist and well tolerated by the patient. Avoiding danger triangle and precise RFA needle positioning is key in preventing RLN injury. Benign nodules regrow if total ablation is not achieved7 and some authors propose additional procedures to complete ablation8 that obviously incurs costs. Total RFA nodule ablation-assisted FFL monitoring eliminates the need for repetitive RFAs, thus reducing overall treatment costs. Finally, FFL monitoring does not prolong procedure, as it is performed simultaneously with RFA. FFL is a valid technique when used in conjunction with hydrodissection to further prevent RLN thermal injury during RFA, especially indicated for large thyroid nodule ablation and professional voice users. Patient Consent and Permission: The patient provided written consent for FFL monitoring and permission to use his portrayals and ultrasonographic images during RFA. The study was completed in accordance with the Declaration of Helsinki as revised in 2013. Adherence to institutional review board protocols was granted. Disclaimer: Representation of any instrumentation within the video does not indicate any endorsement of the product and/or company by the publisher, the American Thyroid Association, or the authors. No competing financial interests exist. Runtime of video: 9 mins 39 secs.

Hoarseness of voice and accompanying symptoms among patients at a tertiary hospital, Tanzania

Introduction: Hoarseness of voice is a common complaint in otolaryngological practice and is an early manifestation of a large variety of conditions directly or indirectly affecting the larynx, ranging from inflammatory to malignant. Hoarseness has a prevalence of 6% in the general population, rising to 11% for professional voice users (30% of the patients whose work requires excessive voice use). Despite the high prevalence many people are unaware that persistent hoarseness may be a red flag indicating cancer. Any patient with hoarseness lasting longer than two weeks, in the absence of an apparent benign cause, requires a thorough evaluation of the larynx by direct or indirect laryngoscopy. The purpose of this study was to determine the prevalence of hoarseness among patients attending otorhinolaryngology services at a tertiary hospital in Tanzania.
 Method: A descriptive cross-sectional study was conducted. A total of 427 patients were recruited. Standard questionnaires were used to collect clinical and socio-demographic information. If hoarseness was identified, further evaluation was carried out including indirect and direct laryngoscopy. Biopsies were taken from the patients with laryngeal mass after direct laryngoscopy under general anaesthesia. Data were analysed using SPSS version 20.0. Confidentiality was maintained.
 Results: The prevalence of hoarseness was 65 (15.2%), which was more common in males 40 (17%) and the age group above 60 years. Common laryngoscopic findings among patients with hoarseness were laryngeal mass 47 (63%) and the leading accompanying symptom was difficulty in breathing 25(32%).
 Conclusion: Hoarseness of voice is a common complaint and accompanies various symptoms. It can also be a red flag for upper aero-digestive malignancies. Therefore, these patients should be evaluated with a high degree of suspicion.

Voice user interfaces in manufacturing logistics: a literature review

AbstractDue to the increasing digitalization in manufacturing logistics, devices to integrate the worker into the digital manufacturing system are necessary. A voice user interface (VUI) can be considered suitable for this purpose due to its flexibility and intuitive operability. Despite the popularity and acceptance of VUIs in everyday life, their use in industrial applications, especially in manufacturing logistics, is still rare. While VUIs have been successfully used in order picking for decades, hardly any other industrial fields of application exist. In this paper, we have identified various barriers to the use of VUI in industrial applications. We categorized them and identified four key barriers. We then conducted a systematic literature review to determine and compare already investigated application areas of VUIs, their characteristics, advantages and disadvantages. We found that in particular the operation of machines and industrial robots, as well as general data and information output on machine and system status, maintenance and employee training are frequently investigated. It is noticeable that VUIs are often used in combination with other user interfaces (UIs). Some challenges to VUI usage, such as high ambient noise levels, have already been solved through various approaches, while other challenges remain. Based on the results of the literature review, we put forward a research agenda regarding further suitable industrial application areas as well as general challenges for the use of VUIs in industrial environments.

Context Matters: Understanding the Effect of Usage Contexts on Users’ Modality Selection in Multimodal Systems

Users often select different modalities in multimodal systems based on context and function. This study aimed to investigate the impact of context on users’ modality selection in a multimodal system. A modality selection task was presented through the developed multimodal system, where the task’s difficulty was controlled based on the number of syllables and touches. The experiments were examined in four usage contexts (baseline, watching, reading, and driving), and the selected modality, physical and mental effort, and satisfaction were measured. A regression model for predicting modality selection was established using binomial logistic regression. The results showed that voice usage increased in the order of baseline, watching, reading, and driving contexts. It was revealed that these changes were attributable to differences in physical and mental interaction efforts of modalities according to contexts. Our finding provides valuable insights into users’ modality selection in different contexts, which could inform the design of more efficient multimodal systems.

Comparison between Voice Disorders among Female Professional and Nonprofessional Voice Users Dealing with Early School Aged Children

Abstract Background Voice disorders are common among females; a main cause is their continuous communication with children as their caregivers. Teachers are exposed to voice abuse and misuse and so also housewives, both of them use their voice excessively to deal with children. Aim of the Work The aim of this study is to compare between voice disorders among female professional and non-professional voice users, in order to identify voice problems affecting nonprofessional voice users for prevention, early detection and proper management to avoid serious vocal health injuries. Study Design a comparative cross-sectional study. Patients and Methods This study included 100 adult female patients who complained of symptoms of voice disorders and were subjected for voice evaluation at the Phoniatrics Unit, Ain Shams University Hospitals. The participated patients were aged between 20 years to 50 years and were divided into 2 groups 50 in each group, group I (teachers) and group II (housewives). Detailed history taking, auditory perceptual assessment of voice (APA), video-laryngoscopic examination, documentation of the auditory perceptual assessment, acoustic analysis and application of Arabic VHI (voice handicap index) were done for all cases. Results The duration of the voice complaint was longer in teachers than housewives. There was nonsignificant difference between both groups regarding the presence of phonasthenic symptoms. The impact of compliant regarding patients own rating of the severity and effect on daily life was higher` in housewives than in teachers, but there was non-significant difference between both groups regarding the listener reaction. There was non-significant difference between teachers and housewives as regard the APA in the overall grade of dysphonia and voice quality. The most prevalent diagnosis of voice disorders among teachers was the vocal fold nodules, while phonasthenia was the most prevalent in housewives. There was a significant difference between the teachers and housewives as regard the total score, functional and emotional subscales of VHI. Results of the acoustic analysis revealed a significant difference between teachers and housewives as regard average F0, jitter % where teachers had higher scores than housewives. While there was non-significant difference as regard RAP, shimm%, APQ and NHR between both groups. A significant positive correlation was found between number of children and functional, emotional and total VHI score in housewives group. There was a significant positive correlation between number of students and patient’s own rating of severity and years of teaching in teachers group. Moreover, there was a highly significant positive relation between having children and emotional subscale of VHI in teachers group. Also, there was a significant positive relation between number of children and functional subscale of VHI in housewives group. Conclusion Voice disorders are common among housewives (Nonprofessional voice users) the same as in teachers (Professional voice users) especially when dealing with early school aged children. Recommendation There is need to increase awareness among the housewives about the importance of voice in their daily life, risk factors for voice problems, how to protect themselves from developing vocal pathologies and when to consult the medical professional.

Digital Enabling Assistive Technology Through the Provision Lens - A Global Perspective in a Nordic Context.

The Nordic healthcare system is well-established, taxation-based, and locally administered, featuring person-centered care as a social welfare pillar. Public provision of assistive technology and welfare technology within healthcare systems means every citizen has equal access to services. But how well are policies and procedures keeping pace with demographic changes and technology developments?. This study critically analyses qualitative data from 24 stakeholders involved in municipal-level procurement and allocation of assistive and welfare technology in Sweden with a specific focus on emerging digital technology. An extant analysis framework was used: the World Health Organization-GATE 5P framework for strengthening access to AT. Recommendations are made for agile procurement and an outcome-based decision frame. The voice of the AT user may be a valuable addition to inform policy.

A Pre-Training Framework Based on Multi-Order Acoustic Simulation for Replay Voice Spoofing Detection.

Voice spoofing attempts to break into a specific automatic speaker verification (ASV) system by forging the user's voice and can be used through methods such as text-to-speech (TTS), voice conversion (VC), and replay attacks. Recently, deep learning-based voice spoofing countermeasures have been developed. However, the problem with replay is that it is difficult to construct a large number of datasets because it requires a physical recording process. To overcome these problems, this study proposes a pre-training framework based on multi-order acoustic simulation for replay voice spoofing detection. Multi-order acoustic simulation utilizes existing clean signal and room impulse response (RIR) datasets to generate audios, which simulate the various acoustic configurations of the original and replayed audios. The acoustic configuration refers to factors such as the microphone type, reverberation, time delay, and noise that may occur between a speaker and microphone during the recording process. We assume that a deep learning model trained on an audio that simulates the various acoustic configurations of the original and replayed audios can classify the acoustic configurations of the original and replay audios well. To validate this, we performed pre-training to classify the audio generated by the multi-order acoustic simulation into three classes: clean signal, audio simulating the acoustic configuration of the original audio, and audio simulating the acoustic configuration of the replay audio. We also set the weights of the pre-training model to the initial weights of the replay voice spoofing detection model using the existing replay voice spoofing dataset and then performed fine-tuning. To validate the effectiveness of the proposed method, we evaluated the performance of the conventional method without pre-training and proposed method using an objective metric, i.e., the accuracy and F1-score. As a result, the conventional method achieved an accuracy of 92.94%, F1-score of 86.92% and the proposed method achieved an accuracy of 98.16%, F1-score of 95.08%.

Futuristic applications of voice user interference on child language development

Voice User Interface (VUI) is an Artificial Intelligence tool that enables children to access a computing device and complete tasks through speech instead of using learning methods. VUI, a form of AI (Artificial Intelligence), takes a sound that children articulate in a spoken statement and use intent recognition to understand the action required to fulfill the child’s spoken request. The design and features of VUI have been developed to increase the interpersonal level of communication with users and, to some degree, make voice assistants behave like humans. The features that have been created, have been shaped in such a way as to improve learning efficacy and ease of use for early childhood learning development. The current available VUIs in the market have been geared to provide children with a simpler way to interact with access to educational technology learning tools. The research posits that there are two primary uses of VUI in childhood learning development exploration, whereby children use VUI as a form of entertainment and information seeking, and children use VUI to develop various knowledge facets. For children in the early language stages currently using language to communicate, VUI language stimulation can help children to engage in continuous communication processes, use and understand various words, and successfully complete more complex sentences. The research seeks to state the problems associated with VUI and the standard opinions based on research associated with the problem. Moreover, the study seeks to articulate the hypothesis that VUI is an effective tool for early childhood language learning through the use of peer-reviewed evidence and examples, to the hypothesis, to generate new and innovative perspectives.

An Evaluation of Voice Hygiene in High School Choir Singers

Voice hygiene, including proper hydration and limiting phonotraumatic behaviors and refluxogenic foods when appropriate, is essential to voice care for singers. The purpose of this study was to evaluate voice care practices, knowledge, and factors correlating with improved voice hygiene and knowledge in adolescents, an understudied demographic within the singing population. 214 high school choir singers, ages 14 through 18, responded to a voice health questionnaire, which included questions about phonotraumatic behaviors and other lifestyle measures regarding voice care, perceived knowledge of the voice, and potential correlational factors such as past singing experience, having a friend or family member when is a professional voice user or possesses an in-depth knowledge of the voice, use of the Internet to sing better or improve understanding of the voice, and other factors. These data were analyzed using descriptive statistics and multivariable linear regressions. Results showed poor voice hygiene. Phonotraumatic behaviors were frequent and the mean daily water intake was only 5.24 cups. However, 94.4% of the respondents warmed up their voices before rehearsals or performances. The mean total knowledge score, from 0 to 6, was only 2.09. In addition, feeling that one took care of his/her voice well, and use of Internet to sing better or advance knowledge of the voice, were correlated with improved knowledge or measures of voice health. Data suggest that improved instruction regarding voice health in the high school setting is necessary and that the Internet or social media may be valuable resources for reaching adolescent singers.

Extent and Effect of Covering Laryngeal Structures with Synthetic Laryngeal Mucus via Two Different Administration Techniques

The first goal of this study was to investigate the coverage of laryngeal structures using two potential administration techniques for synthetic mucus: inhalation and lozenge ingestion. As a second research question, the study investigated the potential effects of these techniques on standardized voice assessment parameters. Fluorescein was added to throat lozenges and to an inhalation solution to visualize the coverage of laryngeal structures through blue light imaging. The study included 70 vocally healthy subjects. Fifty subjects underwent administration via lozenge ingestion and 20 subjects performed the inhalation process. For the first research question, the recordings from the blue light imaging system were categorized to compare the extent of coverage on individual laryngeal structures objectively. Secondly, a standardized voice evaluation protocol was performed before and after each administration to determine any measurable effects of typical voice parameters. The administration via inhalation demonstrated complete coverage of all laryngeal structures, including the vocal folds, ventricular folds, and arytenoid cartilages, as visualized by the fluorescent dye. In contrast, the application of the lozenge predominantly covered the pharynx and laryngeal surface toward the aryepiglottic fold, but not the inferior structures. All in all, the comparison before and after administration showed no clear effect, although a minor deterioration of the acoustic signal was noted in the shimmer and cepstral peak prominence after the inhalation. Our findings indicate that the inhalation process is a more effective technique for covering deeper laryngeal structures such as the vocal folds and ventricular folds with synthetic mucus. This knowledge enables further in vivo studies on the role of laryngeal mucus in phonation in general, and how it can be substituted or supplemented for patients with reduced glandular activity as well as for heavy voice users.

Direct and Indirect Voice Therapy Approaches for Unloading the Laryngeal Mechanism in Professional Voice Users

Direct and Indirect Voice Therapy Approaches for Unloading the Laryngeal Mechanism in Professional Voice Users

Acoustic Parameters in the Evaluation of Voice Quality of Choral Singers. Prototype of Mobile Application for Voice Quality Evaluation

Choral singers are among intensive voice users whose excessive vocal effort puts them at risk of developing voice disorders. The aim of the work was to assess voice quality for choral singers in the choir at the Polish-Japanese Academy of Information Technology. This evaluation was carried out using the acoustic parameters from the COVAREP (A Collaborative Voice Analysis Repository For Speech Technologies) repository. A prototype of a mobile application was also prepared to allow the calculation of these parameters. The study group comprised 6 male and 19 female choir singers. The control group consisted of healthy non-singing individuals, 50 men and 39 women. Auditory perceptual assessment (using the RBH scale) as well as acoustic analysis were used to test the voice quality of all the participants. The voice quality of the female choir singers proved to be normal in comparison with the control group. The male choir singers were found to have tense voice in comparison with the controls. The parameters which proved most effective for voice evaluation were Peak Slope and Normalized Amplitude Quotient.

Service user experiences of participating in a Recovery and Collaborative Care Planning Café framed with CHIME:

The Recovery and Collaborative Care Planning Café (RCCP) used World Café principles[i] to foster inquiry and learning about recovery by talking about important questions in a safe environment that allowed supportive relationships to form. Changing conversations towards recovery and living well with conditions by applying CHIME[ii] as a framework was an important part of the method. CHIME incorporates Connectedness, Hope, Identity, Meaning and Empowerment to help shape conversations. The monthly cafés began with a masterclass exploring each concept in turn and this was then followed by a World Café conversation. Through undertaking a reflective story process with four active service user participants, it became clear that these concepts resonated strongly; their feedback was that the cafe had led to growth and positive health changes and had helped them to develop supportive peer networks and increased stability. This article describes the methodology and explores the impact through service user voices.

Spasmodic Dysphonia Patients' Perception of Pain with Botulinum Toxin Injections.

To evaluate spasmodic dysphonia patients' perception of pain associated with laryngeal botulinum toxin (BTX) injections and to determine factors associated with higher pain scores relative to other included patients. Prospective cohort study. Adult patients with adductor spasmodic dysphonia that presented to a tertiary laryngology practice for BTX injections were recruited from March to July 2022. Patients completed the visual analog scale (VAS) pre-procedure to quantify predicted pain. Ten minutes post-procedure they completed VAS and the short form McGill Pain Questionnaire (SF-MPQ). Factors that may affect pain were extracted from charts. Descriptive statistics, univariate, and multivariate analyses were conducted (alpha = 0.05). One hundred and nineteen patients were included (63 ± 14 yo, 26% Male). SF-MPQ reported mild pain (4.12 ± 4.05 out of 45) with a pain intensity of none to mild (0.70 ± 0.89 out of 5). Bilateral injections yielded significantly higher SF-MPQ scores (5.19 ± 4.66) than unilateral injections (3.30 ± 3.30) (p = 0.012). There was a significant VAS reduction from pre 28.9 ± 24.6 mm (out of 10 mm) to post 24.5 ± 22.3 (p < 0.001). On multiple regression analyses, receiving a bilateral injection significantly (p < 0.05) contributed to a model that predicted higher pre-VAS (p = 0.013). Bilateral injections (p < 0.05) and higher VHI-10 (p < 0.05) contributed to a model that predicted higher total SF-MPQ (p = 0.001) and affective SF-MPQ (p = 0.001) scores. Not being a professional voice user (PVU) significantly (p < 0.05) contributed to a model that predicted higher post-VAS (p = 0.008) scores. BTX injections were well tolerated with low pain scores. Factors associated with higher relative predicted or experienced pain included bilateral versus unilateral injection, PVU status, and higher VHI-10. 4 Laryngoscope, 134:240-246, 2024.

SkillSim: voice apps similarity detection

Virtual personal assistants (VPAs), such as Amazon Alexa and Google Assistant, are software agents designed to perform tasks or provide services to individuals in response to user commands. VPAs extend their functions through third-party voice apps, thereby attracting more users to use VPA-equipped products. Previous studies demonstrate vulnerabilities in the certification, installation, and usage of these third-party voice apps. However, these studies focus on individual apps. To the best of our knowledge, there is no prior research that explores the correlations among voice apps.Voice apps represent a new type of applications that interact with users mainly through a voice user interface instead of a graphical user interface, requiring a distinct approach to analysis. In this study, we present a novel voice app similarity analysis approach to analyze voice apps in the market from a new perspective. Our approach, called SkillSim, detects similarities among voice apps (i.e. skills) based on two dimensions: text similarity and structure similarity. SkillSim measures 30,000 voice apps in the Amazon skill market and reveals that more than 25.9% have at least one other skill with a text similarity greater than 70%. Our analysis identifies several factors that contribute to a high number of similar skills, including the assistant development platforms and their limited templates. Additionally, we observe interesting phenomena, such as developers or platforms creating multiple similar skills with different accounts for purposes such as advertising. Furthermore, we also find that some assistant development platforms develop multiple similar but non-compliant skills, such as requesting user privacy in a non-compliance way, which poses a security risk. Based on the similarity analysis results, we have a deeper understanding of voice apps in the mainstream market.