Assessing the Development and Viability of an Android App for Auditory Training in Older Adults with Hearing Impairments

Purpose: This study aims to develop an android application that is more compatible and user-friendly for the older adult population with hearing impairment and validate the developed app. Method: The app’s framework involved a training mode platform consisting of four levels with four sub-levels each. Every level includes stimuli of word-in-noise containing high and low-frequency words with three different noise types – traffic noise, competing for speech, and speech babble. The levels in the app increased in complexity to provide an efficient auditory training feature. The developed application was validated on older adults and professionals based on a questionnaire with both closed and open-ended questions. Results: Based on the validated responses of the participants, the app is a reliable tool for auditory training in older adults with hearing impairment. The app contains ease-of-use features for older adults and has been considered a platform for improvement in aural rehabilitation. Conclusion: The study marks as a tool in auditory training for older adults to provide the utmost benefit. Increase in the cases of the hearing-impaired population in a few years, a demand to set up a user-friendly management option is beneficiDigitizationtion of aural rehabilitation, especially for older adults, has thus been noted in the present study.


Introduction
Hearing impairment is known to be one of the highest occurring sensory deficits in the world, accounting for 30% of the population over the age of 55 years. In the Indian context, a similar prevalence is found and comprised of about 56-62% [1]. Hearing loss in older adults affects their communication skills, which results in poor comprehension and expression of language skills. This can further lead to reduced functional ability, cognitive decline, and depression, which are strongly associated with decreased quality of life [2]. The typical clinical symptoms observed secondary to hearing loss are difficulty understanding speech in noise, ringing in the ears, make perceptual distinctions between the trained sounds in a systematic order [7].
The training-related improvements support the need to consider auditory training strategies in managing individuals who express concerns about hearing in difficult listening situations [8]. The type of stimulus plays an essential vital stimulating the auditory system. Using speech stimuli, it has been observed improvement in overall speech perception abilities [9]. Speech stimulus with high familiarity has known to provide a been shorter time for response and acts as a primary cue to understand the stimulus [10].triggerition of noise in the training program along with highly familiar words provides a natural setting for the scope of improving communication [11] and enhancing function [12].
There has been a huge demand to create a dynamic and effective mode of digitized aural rehabilitation and other sensory training [13]. Although desktop-based apps are enormous, mobile apps are limited. A potentially beneficial program Angel Sound has been considered a reliable tool for auditory training. Likewise, mobile apps such as Hear Coach and AB Clix have used the features of providing stimulus in the presence of noise. Developing a flexible mobile app for older adults in Indian languages has not been explored much. Kannada is a Dravidian language spoken in the state of Karnataka, and there has been limited evidence of the development of mobile apps which focus on auditory training. A comprehensive management model for older adults with hearing loss for native speakers of Kannada is essential. This shall contribute as a tool in clinical rehabilitation for older adults with hearing impairment to initiate auditory training along with hearing aids or amplification devices.
The present study is primarily addressed the development of a mobile app and validating it. This was to ensure that the app provided portability and flexibility of words in noise implemented in the Kannada language. To address this need, "Shrutham" was developed as free mobile health (mHealth) app to provide clinically relevant auditory training.

Methodology
The study was approved by the Institute of Research and Ethical Committee (IEC -153/2019) and Clinical Trial Registry-India (CTRI) (2020/08/027449). It has been studied under two objectives -the development of the android app and the validation of the developed app.

Development of the Android Application
To record the stimulus of word-in-noise, 600 words, and 300 high-and low-frequency words were selected from a study, Lexical Database in Kannada [14]. These words pertain to the frequency of occurrence, i.e., the more and less frequently occurring words. The three types of noise chosen for the stimulus were competing speech and speech babble [15] along with traffic noise which was measured through a calibrated sound level meter and processed according to standard parameters. The selected word list was recorded with a sampling rate of 44,000 Hz and a time window of 60 min in the Computerized Speech Lab (CSL). The recording was done in a soundproof room with a microphone placed 6 cm away from the speaker. A male and female native Kannada speaker aged 25 to 35 with good intelligibility was selected for recording. The recorded words and noise were merged with the Audacity software version 2.4.2 installed in Windows 10 Asus A555L. Initially, the noise sample and words were imported through the .wav file to the software window for which they change SNR, mainly 0 dB SNR and − 10 dB SNR.

The Framework of the App and User Interface
The mobile application is designed to provide auditory training and promote efficient learning through user responses. The interface involves recording user details along with their hospital ID (Fig. 2). Once the details have been submitted, authorization of the user's phone number is done using Google authentication service by inputting the received One-Time Password (OTP). The interface presents a gamelike home screen, having 4 levels with increased complexity and each level having four stages (Figs. 2 and 4). Each level increases in complexity from high to low-frequency words and decreases SNR as shown in Fig. 1. The complexity of each level in the training is based on the amount of varying noise levels (0dB and 10dB SNR) based on the integration of speech in the presence of noise. With increasing noise levels (background noise), the auditory tests tap higher centres of the brain. On the press of the respective stage, a new page is shown, where the stimuli are played to the user, and the response is recorded.
With the help of Google's speech recognizer Application Programming Interface (API), the application records the user's response in the native language (Kannada) and matches the response to the voice in the audio (stimuli). On the occasion of the user, not being able to record, or distinguish the noise accurately, he/she is also provided with 4 choices of different words which on selection would indicate whether the option selected is correct/incorrect based upon whether the word in the stimuli is the same or not. Upon the completion of the stage, the user's tally or score is saved in a database with the help of FIREBASE. The android application has been coded mainly in JAVA while using XML for designing along with FIREBASE as the database in the backend for storage and retrieval. The data stored is further versatile in its functionalities, allowing the user to replay the audio or skip to the next audio.

Validation of the Android Application
The process of validation was done with two groups of the population, namely, older adults and professional participants.

Validation on Older Adults
For the validation of the first set of participants, 20 older adults of age 55 to 75 years of age were recruited. This group consisted of older adults with near-normal hearing and native Kannada speakers. The participants who chose to participate in the study and agreed through consent received a message with instructions to download and install the app on their phone. The other set of participants who chose to participate, met the investigator to directly explore the app and give feedback. While the participants were going through the app, the internal logging-in function would record the app usage data based on the profile and its functions created. Soon after exploring the various features of the app, the participants were sent a feedback form in Kannada which consisted of questions about the usability of the app, ease of use, and an open-ended question for additional comments.

Validation of Professional Participants
The professionals recruited for the study have at least 5 years of experience in Aural Rehabilitation and Computer Applications. The participants who chose to participate in the study and agreed through consent received a message with instructions to download and install the app on their phone. Similar to the previous validation procedure, the internal logging-in function would record the app usage data based on the profile created and the functions used. Soon after exploring the various features of the app, the participants and professionals were sent an online feedback form. A standard format of System Usability Scale (SUS) developed at the Digital Equipment Corporation and an open-ended question to receive feedback on acoustical and technical aspects.
used to carry out analysis on the user's response and validate the technique fruitfully.
The application is designed to run only in android as it is coded in JAVA and is limited to the scalability of FIRE-BASE. The authentication service provided by Google has limits to the number of users and duration of OTP generation (Fig. 3), hence the application makes use of shared preferences to store sessions of the user. The application is

Validation on Older Adults
Descriptive statistics were used to summarize the data from the checklists, both closed and open-ended questions. The results of the usability questionnaire are presented in Table 1.
Overall, all participants strongly agreed that the app was easy to use. The majority of the participants have rated the app to be useful in functions and have found the app less complex. Almost 60% of the participant's rate that they would be willing to use the application and found that most people would learn to use the mobile app very quickly. Another 65% of the participants strongly agree to recommend the app for older adults with hearing impairment. For the feedback on assistance required, there were mixed responses (40% disagree and 60% agree).
The response to the open-ended question highlighted certain points such as the app being useful and well developed for older adults. With regards to improvement in the app, certain feedbacks also included that the "app stops responding abruptly and captions are needed for symbol understanding", "More levels can be included and with more varieties" and "More interesting and useful features can be added."

Validation of Professional Participants
With the System Usability Scale (SUS) results on professionals, Table 2 delineates the ratings.
All three professionals have given equal outputs that they would be willing to recommend the app to older adults with hearing impairment and also other professionals. Additionally, a professional gave valuable feedback regarding the app in detailed terms, highlighting that there is ease of handling and a good voice recognition feature. The suggestions were to improve on the theme of the app, fonts, and the overall response patterns and reinforcement to the participants. This provided feedback on the better flow of the content and easier navigation of the app.

Discussion
World Health Organisation estimates around millions of people with hearing loss within few years and the lack of auditory training is a substantial public health need. With advancements in technology, it will be beneficial to provide auditory training apps to more people with hearing loss. Specifically, in older adults, where communication is hampered, understanding in the presence of noise is substantially preserved and considered to get better with regular training.  Another important aspect of the mobile app is, presenting stimulus (words) in presence of noise. This has proven to be effective in the areas of assessment and management [17,18]. Researchers have shown various methods of enhancing speech perception in the background noise or reverberant conditions, particularly among hearing-impaired individuals [19,20] and older adults (Bergman, 1980). These studies indicate that perception of speech can be improved with intensive training of words or sentences in various difficultto-listen conditions [17,21]. A customized training program, cLEAR™ [22] included word-in-noise training for adults with or without hearing loss. This program enhanced the training benefits for individuals with hearing loss and a need to implement it clinically for aural rehabilitation. Likewise, a long-term training program was studied with words (high and low frequency) and sentences in the presence of speech-shaped noise. The training was provided to older adults of aided and unaided groups. The training resulted in significant betterment in speech perception in noise for an extended period [23]. With respect to an Indian study, similar to Listening and Communication Enhancement (LACE) [24], software was developed in Kannada language to find the efficacy of auditory training. It included various levels of degraded speech and cognitive tasks. The efficacy of training was found to be significantly enhanced in older adults with hearing impairment indicating an improved Signal to Noise ratio [15]. Thus, training in presence of noise will aid the individual to cope up in a realistic environment and enhance cognitive and other sensory abilities.
With advancements in technology with difficult times such as a pandemic, an app on a personal mobile phone will be a model for tele-rehabilitation.
The present study aimed to develop an android app for auditory training in older adults with hearing loss and validate the app through a user feedback questionnaire. Based on the validation process done by older adults and professionals, both groups have given positive reviews on the developed app. Based on the specifications of the app, they have noted that the app has considerable set-up favoring older adults with hearing impairment. Certain features in the mobile app such as response modes like multiple-choice answers, voice recognition, increased complexity of stimulus in training depicted in various levels, and instruction page have been considered as a valid and reliable method of the training application. The response mode of voice recognition and multiple-choice answers serve as an easier option to choose for older adults with hearing impairment. Voice-to-text technology has been supportive for individuals with communication difficulties in the elderly, especially hearing impairment. This feature has been used in a mobile app for teaching hearing-impaired children which proves to be effective for responding to a stimulus [16]. For older adults with sensory issues like vision and dexterity issues, responding through a voice recognition feature can encompass an easier route for using the app. This enables a voice-based interaction in an auditory training program where older adults can reply to new opportunities. Furthermore, with feedback-specific updates in the app, an older adult with hearing impairment will receive a better tool in the area of aural rehabilitation. This platform thus digitizes a   Additionally, to find the efficacy of the app, a group of target population i.e., older adults with Hearing loss and hearing aid users should take part in the study. This enables a higher effect size of the study and the overall clinical relevance of the app can be accounted for in updated versions. The target population can be a part of the training program as a part of a treatment plan and provide sufficient feedback which can be a reliable response.
With advancements in technology in difficult times such as a pandemic, an app on a personal mobile phone for older adults with hearing impairment will be a model for tele-rehabilitation and telehealth programs. This mode of training can also be improvised in different clinical population such as children with hearing loss, generally adults with central auditory processing disorder to train them with speech in noise stimuli.
Funding Open access funding provided by Manipal Academy of Higher Education, Manipal

Conflict of Interest No conflicts of interest were reported.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.
Adding to the speech-in-noise stimulus, the type of stimulus plays an important role -linguistic and non-linguistic content. Auditory training programs have used different types of stimuli -phonemes, monosyllables, phonetically balanced words, high and low-frequency words, and sentences with target words. Similar to the current study, the use of high and low probability words has been used in other auditory training tasks. With better retrieval capacity, high probability words have obtained better results. It also provides better contextual information and is also stated in the literature that listeners tend to focus on the familiarity of the words during identification of speech in presence of speech babble as a reliable noise type. [25] also witnessed trainingrelated benefits in adults with hearing impairment by using high probability words and further provided evidence that significant changes can be observed with these paradigms. A combination of words of high and low probability along with noise was used in a standardized software named iAngel Sound [26]. This software contains different stimuli for training, among which, high probability sentences in the presence of noise are considered as a stimulus advantageous for auditory training. With these findings, a mobile app using words with high and low familiarity will be beneficial with the hearing-impaired population. Thus, speech perception in noise along with contextual words containing high frequency words provide a better model of auditory training.

Conclusion
The developed app in the present study provides an insight into rehabilitative audiology, mainly in auditory training for older adults with hearing loss. This marks as a beneficial tool by including various features of Kannada high and low-frequency words along with various noise types, response modes, instruction manual. These create a unique program on an android platform. It has been marked as a reliable, flexible, and useful platform for auditory training. Lastly, with the increased use of technology in audiology and mainly in the areas of telehealth (assessment and management), the use of a digitized network will be a dynamic modification.

Future Directions
As an effective mode of management in auditory training, the developed mobile app is found to provide sufficient benefit according to older adults and professionals. By adding additional features to the app, it plays a reliable and sensitive tool in the rehabilitation of older adults.