Inspiration
Many of our parents and family members are also immigrants and struggled with the immigration process. Thus, we wanted to create a product that would help the many other people who are struggling with this process as a way of promoting social and economic good.
What it does
Our product uses an AI chatbot (featherless and elevenlabs) that is trained to help with immigration questions to help immigrants understand their next steps in the immigration process and also provides resources to help them
How we built it
Our project is comprised of a python backend connected to a javascript and html front end. The three major apis we used were featherless, elevenlabs, and whisper to make the llm and speech recognition features. To steamline collaboration we worked as a team with liveshare which allowed for real time file editing on the same project as well as github for version control.
Challenges we ran into
We used elevenlabs to do text-to-speech and speech-to-text and we initially thought we would receive free credits to use elevenlabs. However, due to the problems with receiving the credits we ran into problems with exceeding the free plan credit limit. In addition, we often had to change the UI a lot which slowed the process of modifying the back-end
Accomplishments that we're proud of
We managed to implement text to speech and speech to text into a program which many of us had not done before
What we learned
We learned how to use flask, elevenlabs API and Retrieval augmented Generation
What's next for BridgeMate
We want to add an accounts feature to keep track of users’ background to better tailor it to their circumstances
Built With
- css3
- elevenlabs
- featherless
- flask
- html5
- javascript
- opennote
- python
Log in or sign up for Devpost to join the conversation.