Inspiration
In the world of diverse cultures, communication is often hindered by the use of different languages. Our vision is to enhance communication and content sharing by bridging the language barrier to make sharing ideas even more seamless and accessible.
What it does
BabelJam is a web app that aims to solve this problem. In just a few clicks, anyone can make their videos reach an even wider audience.
The creator can upload their video to the BabelJam to generate a transcription. The creator can then double-check the transcription and make amendments when necessary.
The creator can then translate the transcription into a wide selection of languages. They can also customise the translated text and make changes when necessary.
BabelJam can then regenerates the audio to speak in the desired language. The newly generated audio will sound like the original voice!
How we built it
Using Next JS and TailwindCSS in the frontend, we can deliver a beautiful and intuitive user interface to make the technologies accessible to everyone.
We used Python with FastAPI on the backend along with OpenAI Whisper, Google Translate API, ElevenLabs Voice Cloning, ffmpeg, and MoviePy to generate the transcription, translation, and final video with translated audio.
Challenges we ran into
Selecting the correct libraries and resolving dependency issues when building the project.
Accomplishments that we're proud of
Despite facing many challenges in model integration, we still managed to deliver the end product that worked well!
What we learned
Developing BabelJam in a team is a precious opportunity for us to learn about collaboration in a software development team.
What's next for BabelJam
We will continue to improve the application by enhancing the user experience, performance, and capabilities, making it useful and accessible for everyone.
Log in or sign up for Devpost to join the conversation.