SpeakKind

Inspiration

Many women and girls face harassment or solicitation in public spaces, often with limited options for immediate help. SpeakKind aims to address these scenarios by providing a means for real-time intervention.

What it does

SpeakKind listens to conversations in public spaces (like children’s homes, schools, clubs, and restaurants) and analyzes the context to detect signs of bullying or solicitation. When concerning interactions are detected, administrators or managers receive instant alerts, enabling them to intervene quickly or contact authorities if needed.

How we built it

SpeakKind is built using a NodeMCU-32S microcontroller paired with an INMP441 microphone, which captures audio for analysis.

Challenges we ran into

One of the main challenges was the limited time available to configure the microphone with the microcontroller effectively.

Accomplishments that we’re proud of

We successfully developed the app and dashboard, enabling administrators to monitor and respond to alerts in real time.

What we learned

We learned the importance of beginning app development early to allow more time for refining and troubleshooting individual components.

What’s next for SpeakKind

Next, we plan to enhance the system’s capabilities with advanced machine learning models to improve conversation analysis accuracy. Expanding the system's reach to additional public spaces and adding features like sensitivity adjustments, location tracking for incident mapping, or integration with law enforcement systems (where feasible) are also key goals. Each of these could make SpeakKind an even more impactful tool for promoting safety in public environments.

Built With

• c/c++
• google-web-speech-api
• python
• ts