Inspiration

Staying at home cooped up with your family is difficult enough. It becomes even harder when those you live with pose a threat to your safety and well-being. However, that is the reality for many. In fact, domestic abuse calls have gone up by 22% since COVID-19. Our app aims to help people who fear domestic abuse feel secure and allow them to contact a trusted friend or the authorities with a simple voice command.

What it does

Violet is an app that's designed to be secretive, and installed by people who fear domestic abuse. The app records conversations above a certain volume and uses Google speech-to-text API to listen to what is being said. Recordings are only saved if negative words are being used, meaning that a fight is happening. These recordings can be later used as evidence, viewed, or deleted. Additionally, the app listens for a secret trigger word that the user sets themself. If it hears this word, it sends a discreet text to a trusted friend warning that the user may be in danger. You can customize the icon of the app so the appearance can be whatever you want (such as the Discord logo). The app is password protected and can be disguised as a food delivery app for more security.

How we built it

We used Adobe XD to prototype the UI and HTML, CSS, and Javascript for the website. The app was created using Android Studio.

Challenges we ran into

Some challenges we ran into were how to design the app in a way that we can still spread awareness about the app without worrying that abusers might find out about it.

Accomplishments that we're proud of

We're proud of being able to create something that can help and bring awareness to an issue we feel strongly about. Additionally, we're very pleased about creating our first website and how visually appealing it looks.

What we learned

We learned how to create a website, use Android Studio, and more about material design.

What's next for Violet

The next steps for Violet would be to do more research about abuse patterns and find a way to track cues from speech. This could potentially allow us to filter conversations better.

Share this project:

Updates