Inspiration
We all know the pain of having some sort of issue, and then having to wait hours for a customer support representative for a problem that could have been fixed quite simply. It's hard on the customer, and hard on the representatives that likely don't want to be answering the same question over and over again. That's when we realized the power of automating the most frequently asked questions.
What it does
Revolutionizing the customer relationship management game, FAQSync simplifies and automates end-to-end process of gathering the most frequently asked questions, and updating them accordingly for the user. FAQSync uses a transcriber to convert the contents of each customer support phone call into text, and then an NLP model extracts the most commonly asked questions. Our platform seamlessly updates the FAQs of the company website in order to improve the user experience, and lessen the strain on customer service representatives.
How we built it
Harnessing the capability of T-Mobile's Your Number Anywhere (YNA) API, we allow for on-the-web calling to customer support. We also plan to fully implement the Google Cloud speech-to-text API as well as OpenAI's GPT-3.5 language model for gathering analytics. Our web application will be hosted through Heroku and will host our database on MongoDB.
Challenges we ran into
During our hackathon, we faced difficulties in setting up the T-Mobile API for calling, but with assistance from T-Mobile mentors, we were able to successfully implement the API. We also encountered challenges with implementing Google’s speech-to-text API for transcription and storing them in our MongoDB database.
Accomplishments that we're proud of
We are particularly proud of our intuitive user interface, we believe it really will streamline the customer service process. Additionally, we are proud that we were able to integrate the YNA API into our own web application.
What we learned
This hackathon gave us the opportunity to dive into new APIs and learn more about the backend. We overcame authentication issues and successfully integrated T-Mobile's YNA API into our own web-application. We also explored Google Speech-to-Text API and the OpenAI GPT-3.5 API and gained a deeper understanding of their implementation process.
What's next for FAQSync
We are quite excited to keep working on FAQSync. Our next steps involve fully integrating the speech-to-text API along with the GPT-3.5 language model, to have full capability. We also plan to apply our product in a specific industry, such as e-commerce.
Log in or sign up for Devpost to join the conversation.