Inspiration

Many people struggle with depression, without many people even noticing. Often they have no place to turn to, don't know who to talk to or suffer from additional anxiety and don't even want to talk on the phone because they would feel overwhelmed. Opening up is the hardest step of recovery - our App tries to make it easier for people to start a conversation.

What it does

TALK TO ME offers an anonymous, discrete way to chat with therapists using AI to find the therapist who is most fitted to your needs. Analyzing your language and tracking your mood and overall feeling to find specialized help and give you the help you need. Whether you just need somebody to talk to or a looking for acute help - TALK TO ME will figure out what's best for you.

How we built it

We build it using azure AI for analyzing chat messages sent by the patient and used a specialized algorithm to find a therapist with fitting specialties. Using a Client-Server Connection to communicate between the three parties (the patient, azure AI, and the therapist) where the Server establishing the connection between therapist and patient. We used XML and android-studio to create an Android App interface and JavaSwing to build a Desktop Application that allows the therapist to chat with the patient and analyze its data in parallel.

Challenges we ran into

Integrating azure API was more difficult than expected, no one knew XML and the azure bot service needed to be trained to our specific needs.

Accomplishments that we're proud of

Creating a prototype that at least shows the theory and having an idea that could improve the lives of people dealing with depression and dynasty.

What we learned

XML, integrating azure, working in a group and with git, working on different types of codes and including them in the same project

What's next for TALK TO ME

Integrating a better structure to determine the best suitable therapist. Implementing encryption to make sure only as much data as you want to share is read by who you want it to share with. Training the bot and the AI to analyze and evaluate the best possible output for the patient. Connecting history, statistics and the chat function even better and maybe work on the overall performance ...

Share this project:
×

Updates