Voice operated Fall Detection System

Project Title

Voice operated Fall Detection System

Project Area of Specialization Artificial IntelligenceProject Summary

Caretakers are not available to old patients all the time in hospitals or in houses etc. Our main aim is to develop such an app which will detect sound of a person who fell on the ground, our application will detect and analyze it through his voice and send a message to the respective caretaker. Once caretaker is informed on time, first aid will be provided and person can be brought to hospital in case of major casualties which is beneficial for them.  

Project Objectives

• Ensuring the safety of the adult, children and patients.
• If patient or children is alone and comes across any mishap they must be protected.
• The caretakers should be informed in time so that they can provide the first aid if required.
• All the Fall events should be detected, system should not report any false event.
• System should be available all the time.

Project Implementation Method

We will collect the dataset of maximum audios that contain any fall event. The voice can be yelling or serious crying or may be a louder and sharp voice indicating the extreme pain of the people. Then will use this dataset and will train a Model based on  Machine Learning algorithm. This will be done by Extracting the features of the audio, the technique we are using to extract the feature is Local Ternary Pattern (LTP).  This feature will be used for training purpose. Different ML algorithms will be used, it may convential model or neural network model. Once the model is trained we will convert it to application and will test it on un seen data like extremely painful voices.

In real scenario the, the input voice will be taken through a smart watch that will be attached to the patient. Those wrist watch will constantly analyze the voice and classify it accordingly. If the voice is classified as a Fall event then alert will be generated and the caretakers of the patient will be informed through it.

Benefits of the Project

This project will be useful at different areas: 

• Hospitals
• Old age hoomes
• Day Care centers

This project will be very useful at daycare centers where babies needs attention all the time. This would reduce the headache of caretakers over there.  Aged persons do need attention, specially disabled persons when they are performing any activity on their own, this product will act like an assistant for him. It would be beneficial if we saved a person in case of serious injury at critical time. First aid at that time can be very helpful and lifesaving. Generating alert or sending message to the relatives will let them know that an unwanted situation has occurred.

Technical Details of Final Deliverable

Mobile Application: We will build a mobile application that will use the trained model we would have developed. This model will be operational and listening to the voice in background. The activity will not go to sleep or pause state. Then it will detect the painful voice and classify it using the model. If it is classified as fall event, the care takers who are added in the application will be notified. This app will run in a smart watch and it will be used to input the audio because smart watch will be attached with the person even, he is anywhere, app will be running in background.

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther IndustriesCore Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Good Health and Well-Being for People, Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
Samsung Galaxy S4 watch	Equipment	1	35000	35000
Kingston SSD 512GB	Equipment	1	20000	20000
GPU component Graphic Card 2GB	Equipment	1	15000	15000
Thesis binding	Miscellaneous	1	5000	5000
App Publication on Play store	Miscellaneous	1	5000	5000