Designing a Talking System for Sign Language Recognition of people with Autism Spectrum Disorder

Project Title

Designing a Talking System for Sign Language Recognition of people with Autism Spectrum Disorder

Project Area of Specialization Artificial IntelligenceProject Summary

This project introduces sign language recognition of people with an autism spectrum disorder. In the whole world’s population, there is 1 in 59 persons who are autism spectrum disordered. As the ASD people cannot properly communicate with the normal persons properly through speaking so as a consequent they communicate in the sign language which is not understood by the normal person. Hence there is the social gap between the normal and the ASD people. To overcome these problems, we developed a talking system that is beneficial to the ASD and as well as the ASD caretaker.

We implemented this system through the following steps:

1) Data Acquisition of ASD people by a wearable sensor (accelerometers and gyroscope).

2) Extraction of features from the acquired data by using Matlab.

3) Classification of the extracted features by using different algorithms by using Matlab.

4) Raspberrypi '3' having BLE blue-tooth and Python is used for the real-time implementation.

So, this system will be able to overcome the social communication gap between the normal and the ASD people so that ASD people can have a good quality of life.

Project Objectives

•    Data acquisition of ASD people.
•    Feature extraction of gesture data. 
•    Use of classifiers to recognize user activities and provide performance comparison for different classifiers.
•    Developing voice/vision hardware for ASD caretaker.
•    Provide a better and save lives to ASD people.

Project Implementation Method

Benefits of the Project

• It helps the ASD people to easily convey their message to a normal person.
• It helps the normal person or caretaker to understand ASD people speech immediately and pay attention to them.
• So it will directly benefit the ASD people by making understandable communication  with a normal person
• Indirectly it will benefit a normal person by understanding ASD people  sign language

Technical Details of Final Deliverable

• Sign language database using Wearable sensors
• Software ASD people gesture recognition.
• The hardware platform for ASD children gesture acquisition and platform

Final Deliverable of the Project HW/SW integrated systemType of Industry Medical , Manufacturing , Health Technologies Artificial Intelligence(AI), Internet of Things (IoT), Wearables and Implantables, Big DataSustainable Development Goals Good Health and Well-Being for People, Sustainable Cities and CommunitiesRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	74000
Tri-Axial Accelerometer Sensor	Equipment	2	8000	16000
Tri- Axial Gyroscope Sensor	Equipment	2	8000	16000
Raspberry Pi	Equipment	2	9000	18000
LCD	Equipment	1	6000	6000
Speaker	Equipment	1	3000	3000
Bluetooth Module	Equipment	2	2500	5000
Documentation, Posters, Stationery etc	Miscellaneous	10	1000	10000