Inspiration
We've all been there: desperately trying to follow a brilliant tutorial, lecture, or educational video on YouTube, only to be held back by a heavy accent, poor audio quality, or fast speech that makes the content incomprehensible. This is especially true for subjects like advanced mathematics or niche programming guides, where every single word matters. Our team was inspired by the simple frustration of not being able to fully access and benefit from the vast, free educational content online, just because of an audio barrier. We decided to create a tool that democratizes learning by eliminating this "mumble factor.
What it does
The Unaccentuator takes a YouTube URL as input and performs a three-step audio transformation:
Extraction and Conversion: It extracts the audio track from the video and uses a Speech-to-Text (STT) model to convert the spoken word into a clean, accurate text transcript.
Accent Neutralization: The generated text is then fed into a Text-to-Speech (TTS) engine.
Clear Playback: The TTS engine reads the entire transcript back to the user in a clear, standard, and highly intelligible voice (e.g., standard American, British, or Canadian English), effectively "unaccentuating" the original speaker's voice. The user gets a brand new, clear audio track that delivers the same information without the struggle.
How we built it
We used Google Gemini to generate the HTML, CSS, and JavaScript code to implement our idea in less than an hour.
Challenges we ran into
We encountered several privacy issues regarding accessing copyrighted YouTube videos. We also spent the majority of our Hackathon time trying to debug faulty code generated by Google Gemini.
Accomplishments that we're proud of
We are proud of utilizing Google Gemini to implement our vision for The Unaccentuator in such a short time frame.
What we learned'
We learned that although Google Gemini is a powerful tool, it must be coupled with human capabilities to develop a fully functional project. Solely relying on Google Gemini to generate and implement code often leads to many bugs and errors.
What's next for The Unaccentuator
To move this project from a simple prototype of concept to a robust, scalable service, our immediate next steps involve implementing dedicated API calls and a proper back-end infrastructure.
Log in or sign up for Devpost to join the conversation.