Inspiration
The inspiration for SafePath stems from the everyday safety challenges that women and girls face globally. With incidents of harassment, violence, and unsafe environments disproportionately affecting girls and women, I wanted to create a tool that could help them feel secure and protected. SafePath empowers girls to support each other through advanced AI technology that provides real-time safety alerts and preventative guidance.
What it does
SafePath is a mobile app designed as a real-time safety companion for girls, providing features like:
- Real-Time Threat Detection: Powered by Google’s Gemini AI, it analyzes environmental sounds and nearby conversations using NLP to detect potential threats.
- Voice-Activated Distress Signals: Users can trigger an alert with a specific voice command.
- Location Sharing: Instantly shares the user's location with trusted contacts.
- Safe Route Guidance: Offers safe route suggestions to help users navigate through safe zones.
How I built it
SafePath is developed with a focus on user-centric design and robust, AI-driven safety features. Key elements included:
- Google Gemini AI for dynamic threat detection.
- NLP to analyze sounds and identify potentially unsafe situations.
- Firebase for real-time data storage, location-sharing, and user authentication.
- Google Maps API for safe route guidance.
I used Python and Flask for backend API development, with Kotlin for mobile app development on Android.
Challenges I ran into
One of the major challenges was accurately detecting threats based on real-time audio cues and ensuring that SafePath wouldn’t generate false alarms. I also encountered challenges with ensuring reliable location services in areas with poor GPS reception. Balancing ease of use with security was essential, requiring me to implement robust protections for user data while maintaining accessibility.
Accomplishments that I am proud of
- Integrating real-time NLP-based threat detection.
- Creating a seamless and intuitive user experience.
- Ensuring data privacy and security for users, particularly given the sensitivity of location data.
What I learned
Working on SafePath has shown me the transformative potential of AI for social impact, specifically in the realm of gender equality and empowerment. I explored various AI capabilities, from NLP to location intelligence, and gained a deeper understanding of how safety features can enhance real-world support.
What's next for SafePath: Girls' Real-Time Safety Companion
I aim to enhance SafePath by:
- Adding multi-language support for wider accessibility.
- Expanding AI capabilities to detect a broader range of environmental cues.
- Exploring partnerships with local safety organizations to increase impact.
Built With
- android
- android-studio
- api
- firebase
- flask
- google-cloud
- google-gemini
- google-maps
- java
- kotlin
- machine-learning
- python
- tensorflow
- twilio
Log in or sign up for Devpost to join the conversation.