Inspiration
According to a report by the Bureau of Justice Statistics, about 50% of assault culprits get released from prison due to lack of evidence of their presence or lack of evidence of violence. This inspired the development of Watch Out, a game-changing app designed to provide crucial evidence that can help increase the chances of catching assault culprits, ensuring justice and safety. This app is a true blend of Traditional Software development tools, instincts and latest Artificial Intelligence integration.
This app is based on one of the 17 UN principles, JUSTICE; specifically.
What it does
1) Records audio of the surrounding environment for a predetermined time period after clicking the SOS button; providing contextual and concrete evidence for happenstances related to the incident.
2) Contains a one of a kind AI hazard detection mechanism that converts the audio file saved into text using Microsoft Azure's speech detection model and determines potential causes for emergencies and sends the information to the emergency contacts.
3) Uses Bluetooth signals emitted by nearby devices to create a unique list of all the devices present in the vicinity and sends it to the authorities and user-selected emergency contacts, identifying potential witnesses or suspects who were present at the scene of the crime.
4) Corroborates other evidence gathered from CCTV footage or eyewitness accounts.
5) Contains a one of a kind AI hazard detection mechanism that converts the audio file saved into text using Microsoft Azure's speech detection model and determines potential causes for emergencies and sends the information to the emergency contacts.
6) Has a child safety feature that initiates an SOS response when the device is sensed to be in a moving car, and records any conversation or audio inside the car; along with detecting the model of the car through bluetooth signals emitted from it; resulting in a one second detection of the car to be looked for in case of an unfortunate incident.
7) Sends a light warning message to emergency contacts when the battery is low, along with the last active location of the device.
8) Alerts emergency contacts if the smartwatch has been removed with a jerk, along with the location and a list of Bluetooth devices nearby, and starts a recording buffer.
Accomplishments that we're proud of
1) Creating a tool that has the potential to make a significant impact in increasing the chances of catching assault culprits and ensuring justice for victims.
2) Designing a user-friendly interface that makes it easy to access and activate features.
3) Implementing one of a kind Bluetooth technology, audio recording technology, AND AI hazard detection mechanism to provide concrete evidence in case of violent incidents and to put safety at the highest priority.
Mainly, this is the first project ever to deal with evidence gathering as it's priority and also; this is the first app to start a very important conversation about blending AI with more diverse range of use cases.
Main talking points for judging criteria:
First project in a hackathon to inculcate AI with something that is NOT education, communication or finance. First concept of idea project to work on a full fledged evidence gathering tool. Potential to make a huge range of diverse use cases in the coming world; with refining and reusing the idea of NLP AI to process emergencies and Bluetooth/Voice evidences.
Log in or sign up for Devpost to join the conversation.