ListenTTC

Inspiration

We always hear lots of news about harassment on the TTC and wanted to address it!

What it does

The project is designed to record conversations between TTC (Toronto Transit Commission) workers and passengers and perform real-time sentiment analysis on these conversations. The sentiment analysis provides a numerical value ranging from -5 to 5, indicating the emotional tone of the conversation. When the sentiment analysis indicates a very negative sentiment close to -5, the system triggers an alert to notify security personnel through the form of a text message, potentially indicating a dangerous or problematic situation. The key goal of the project is to enhance security and passenger safety within the TTC system by monitoring and identifying potentially hazardous situations in real time.

How we built it

For the front-end, we used HTML/CSS/JS with the help of Bootstrap. We also used Flask for the API and web sockets. After, we used Hugging Face for the AI Sentiment analysis and OpenAI Whisper to translate speech to text.

What we learned

Since our first hackathon for most of us, we learned how to apply many new technologies, such as OpenAI. We also improved our teamwork and communication skills.

Built With

• ai
• bootstrap
• css
• flask
• html
• javascript
• openai