Inspiration
Student learning features a substantial amount of time sitting and reading content.
What it does
Turn pdfs into an engaging podcast between a guest and host. These simulate insightful conversations.
How we built it
Backend utilises Google Cloud: Gemini for scripts, Google Cloud: Text To Speech to generate a podcast from a pdf upload and Google Cloud Storage (GCS) to store files. PostgreSQL used to store metadata about what files exist in GCS.
Frontend uses a simple UX similar to Spotify for a familiar experience, with obvious sections for main functionalities.
Challenges we ran into
Connecting frontend and backend.
Accomplishments that we're proud of
Achieving programmatically alternating voices for a podcast script to simulate an engaging podcast. Successfully connecting frontend and backend.
What we learned
Docker is important.
What's next for Team SIT102: Luyu
Luyu can be improved by making the podcast voices more natural, can be done by modifying the existing implementation or waiting for Google's two-person voices to release. We plan to continue participating in hackathons.
Log in or sign up for Devpost to join the conversation.