Inspiration - Owner of the Big Baller Brand Lavar Ball's wife Tina who suffered a stroke and lost the ability to speak coherently and speak clearly as well as losing the ability to communicate to loved ones and say what is on her mind, she would speak 1 word at a time, and used techniques similar to what this project/prototype aims to do with people that want to learn to speak fluently again and those who need to relearn.
What it does - Allows users to have a platform for speech therapy using a duolingo like score, streak and exp alongside increasing difficulty making the user say the word that relates to the image shown which is then analysed by Open AI to see whether this was correct or not which in result allows the user to efficiently train their mind to speak the words that come to mind.
How we built it - We used Fast API to handle he backend and any queries alongside SQLite and a React js frontend to create a UI useful to the user and easy to use.
Challenges we ran into - Cost of usage of Open API as a token is required to use the transcription service, Time needed to add many more images and build onto the difficulty aspect.
Accomplishments that we're proud of - Usage Speech to text to help people communicate and lead to the greater good of society. GitHub scammed me so bad asw RIP
What we learned - Fast API and how to use it thoroughly, alongside React, it was a first time using both for me and my team mate Nik Nik Muhammad Imran Azhar
What's next for ReLingo - Building on the system adding range to the different techniques that can be implemented for speech therapy along with difficulties and user profiles.
Log in or sign up for Devpost to join the conversation.