Augmentative Alternative Communication for Speech Delay Children
Nor Kamaliah Bt Mohd Kamaruddin1\*, Nora Yanti Bt Che Jan2
1 Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA, 02600 Arau, Perlis, Malaysia
Corresponding author: \*noray084@uitm.edu.my
ABSTRACT
*Augmentative Alternative Communication (AAC) is a method of communication that can ‘add-on’ to speech and are used to get around problems with ordinary speech*. *The purpose of this research is to design and develop an AAC for speech delay children. The technique that is applied is speech recognition to improve their speech developmental among them. In methodology, the model that used in the project is Waterfall model. There are two types of testing used to evaluate the application which are heuristic evaluation and usability testing. A heuristic evaluation has been conducted to three expert users where feedbacks and suggestions from them were taken into refinement. After refinement , Usability Testing was conducted with thirty speech delay children at age four to nine years old. Usability Testing was used to evaluate the application based on User Interface Satisfaction, Interaction and Navigation, Functionality, Content and General Feedback. Result from Usability Testing, reported that speech delay children agreed the functionality of the application very well where the icon and image used are suitable. Furthermore, there were also recommendation to improve the AAC application in the future such as to include verbal instructions along with the instructions of the word for the users to understand where the users can play the application themselves without being instructed by the therapist or teacher. As a conclusion, AAC application for speech delay children has been successfully designed and developed to assist the children in improving their communication skill.*
***Keywords***: *speech delay, Augmentative Alternative Communication, application*
# INTRODUCTION
This paper focused on AAC’s web application. AAC is a method of communication that can ‘add-on’ to speech and are used to get around problems with ordinary speech. It covers an area of structures from low technology systems such as communication boards to a change of high technology speech output conversation aids. AAC application typically provide language support in the form of words and phrases organized into categories Nowadays, speech and language delay are getting worst among the children in Malaysia (Sundus, 2018). Some children still could not say all of the words perfectly even in age of seven years old. Parents who have kids of disability, they will get their children to a speech therapy center. Normally, this class session will last for one hour but the children will only focus on the first 20 minutes. This may be due to lack of interest in the learning session. As children are known to be interested in playing handheld devices, there is an innovative technology to improve speech development among children. In the study, AAC application is developed by applying multimedia principle for speech and language children. AAC application aims to facilitate the functional communication to preschooler children between four to nine years old that have speech delay problem. The study will involve Heuristic Evaluation and Usability Testing. The Heuristic Evaluation will involve three experts who have experiences in the Human Computer Interaction (HCI) field. The Usability Testing will be conducted to preschoolers and primary students between four to nine years old.
## RELATED WORKS
Baldassarri, Rubio, Azpiroz & Cerezo (2014) introduced research about encourage useful correspondence to individuals with complex correspondence needs. The tool is divided in two applications. The first application is AraBoard Constructor which allows developers to create and edit communication boards adapted to the particular by therapist. AraBoard Constructor is characterized by its ease of use in all aspects, since it has a graphical interface designed to make possible anyone to create and edit boards intuitively in a few minutes. The second application is AraBoard Player, that shows the board in the full screen, and the user interacts with in by pressing the different cells that composed it. Once a cell has been marked, it plays the audio associated with it. In order to be accessible to users with physical disabilities, the audio playback mechanism is designed so that although not stating exactly the desired cell, always run the audio associated with the closest cell to the area that has been pressed.
Figure 1: Infrastructure of the E-learning-based speech therapy
system.
Based on Figure 1 given above, the client has access to their individual speech training programs composed by their therapist. For each client, the training procedure consists of the following steps. First, the kids have to select a training item from the prescribed course. Second, the kids have to listen on a target speech sample that have been downloaded from the central server, using a headset. Third, the kids have to record the imitation attempt via the headset’s microphone connected to the laptop or computer. Fourth, the system will compare the target and the recorded samples by selecting ‘‘compare.’’ This comparison is based on auditory feedback only. Fifth, deciding whether the speech attempt will be uploaded to the central server by selecting the ‘‘save’’ button. Then once the recorded speech has been uploaded, a client’s auditory discrimination is supported by automatic visual feedback for intensity and overall pitch. Lastly, the system will determine whether a new speech attempt is required to approach the target speech. (Rietveld et al., 2010)
> **METHODOLOGY**
>
> Processing mapping has been applied in this project as guideline to illustrate the flow of the application. Process mapping is an exercise to perceive all the step in diagrammatic (Okrent, M. & Vokurka, R., 2004). The purpose in designing the flow system is to guide the researcher in understanding the process of AAC application. The technique used in this developed application are speech recognition, dragging alphabet, dragging image and hearing alphabet. In speech session, each category of item must be answered by recording their speech whether the phrases or word is right or not. Figure 2 given below shows processing mapping for spell session. The children will be given three levels which are easy, medium and hard. In easy level, the children have to hear each of alphabets from A to Z. In medium level, the user has to hear the sound before drag the alphabet into each of the boxes. In hard level, the user has to drag the image into suitable boxes. Score will be given based on correct drag.
>
> 
Figure 2: Processing Mapping for spell session
Figure 3: Dragging alphabet based on given image
> Figure 3 above shows the interface for dragging alphabet based on given image. The children have to drag the alphabet into each of the boxes. The alphabet will be randomly chosen by the children. The notification will be display if the children dragging alphabet into wrong boxes.
>
> **FINDINGS**
>
> After the children used the developed AAC application, the system was being evaluated by conducting usability test. Usability Testing was conducted to evaluate the application usability with the real users. According to Rubin. J., and Chisnell (2008), the term usability testing can be defined as the use of people as participants which represent the target users in the process to evaluate the product meets specific usability criteria. The usability testing was conducted to find out how does the application meet the users’ requirement. This testing was conducted with 30 speech delay children. The evaluation is conducted at five educational institutions; (1) Joyous Kiddy center, (2) Pasti Alhmabra1, (3) Pasti Alhambra2, (4) Tadika Yashabul Yamin and (5) Sekolah Rendah Integrasi Teras Islam (SRITI). The children explored and tested on AAC application with simple guidance from their teachers. The application was evaluated based on four criteria; (1) user interface satisfaction, (2) interaction and navigation, (3) content and (4) functionality. The teachers will rate the feature on behalf of the children based on their observation when the children were using the application. Table 4 below illustrates the scale of the usability testing. The mean was determined based on the rating by scale from one to five obtained from the feedback from the participants.
**Table 4:** Scale on Usability Testing
| Rank | Scale |
| ----------------- | ----- |
| Strongly disagree | 1 |
| Disagree | 2 |
| Moderate | 3 |
| Agree | 4 |
| Strongly agree | 5 |
> 
>
> Figure 5: Mean score for functionality category
>
> From Figure 5 above, it shows that most of the respondent agreed that dragging buttons work well. The bar chart shows that the mean score is 3.96 which indicate the children can drag images and drop them into suitable boxes with guidance. Next, only some of the respondents agreed that voice is clear to be heard. The bar chart shows that mean score is 2.96 which implies that voice tone in speech session does not consistent. Next, most of the respondents agreed that buttons were functioning well. This was proven with mean score as 3.83. This implies that children can clicked all the buttons provided in every session with guidance from the teachers. Next, only some of the respondents agreed that speech can be effectively detected during the test. This was proven with the mean of 3.13. This indicates that notification could not detect their speech and the children have to keep repeating until they voice out the correct words. Next, the result shows that most of the respondents agreed that the notification responded fast. This was proven with mean of 3.13. This implies the participants could voice out the words clearly and give fast respond. Lastly, the functionality testing asked if the application showed the score obtained by the respondents. The result shows all the respondents agreed that the application showed the score they get. The finding comes out the mean is 4.This indicates the application recorded the mark based on how many they are correct. Overall, the functionality of the application work very well as the children can work on their tasks in dragging alphabet and image with a guide from the therapist.
>
> **CONCLUSION**
>
> The research is carried out with the purpose of develop an AAC through web application and evaluate the usability of the developed AAC. This system has been developed based on Waterfall Model and the target users in using this application are speech delay children in between age four to nine years old. The Heuristic Evaluation was conducted to identify all the strengths and any weakness if present, associated with the design of the user interface of the developed application. Apart of that, Usability Testing also was conducted to evaluate what users feel by experiencing this system in order to enhance the user in using AAC application. In order to gain feedback from the user, there were four criteria that evaluator need to evaluate which are user interface satisfaction, interaction and navigation, content and functionality of the application. Result from Usability Testing, reported that speech delay children agreed the functionality of the application very well where the icon and image used are suitable. In conclusion, the application is believed to have a huge impact on speech delay children. This application is able to decrease the problem of speech developmental among children. From the conducted research, it is found that the application had achieved its objectives since the speech delay children can use it.
>
> **FUTURE WORKS**
>
> There are several recommendations and suggestions obtained from the testing phase for further enhancement in the future. First, it is recommended that instructions given must be clear so the children can enjoy playing the application. Based on the observation, the children need an explanation from the therapist and teachers as they do not know what the task is all about. Next, the application should include verbal instructions along with the instructions in the text form so that the users can understand and play the AAC application themselves without being instructed by the therapist or teachers.
# REFERENCES
Baldassarri, S., Marco, J., Cerezo, E., & Moreno, L. (2014, June). Accessibility evaluation of an alternative and augmentative communication (AAC) tool. In *International Conference on Universal Access in Human-Computer Interaction* (pp. 529-540). Springer, Cham.
Grossinho, A., Cavaco, S., & Magalhães, J. (2014, November). An interactive toolset for speech therapy. In *Proceedings of the 11th Conference on Advances in Computer Entertainment Technology* (p. 36). ACM.
Karberis, G., & Kouroupetroglou, G. (2002, April). Transforming spontaneous telegraphic language to Well-Formed greek sentences for alternative and augmentative communication. In *Hellenic Conference on Artificial Intelligence* (pp. 155-166). Springer, Berlin, Heidelberg.
Sundus, M. (2018). The Impact of Using Gadgets on Children, Journal of Depression and Anxiety, 7(1), 296.
Okrent, M. D., & Vokurka, R. J. (2004). Process mapping in successful ERP implementations. *Industrial Management & Data Systems*, *104*(8), 637-643.