As high school students, we often need to present slideshows with loads of information - this means speaking from memory and having a very scripted feel to our presentations. We tackled this by creating a software which will create the presentation as you speak, allowing you to have full control in the moment!
What it does
SmartPres uses Google Cloud services such as Speech To Text to recognize your voice, what you say, as well as specific keywords and points. Our software will then find the key words, add them to your presentation and find relevant images and visuals to match your words as well. You can also pre make some slides, and the software will automatically detect once you reach the end of the slide and switch it for you.
How we built it
We used Google Cloud to transcribe the live speech into text, the Pixabay API to find relevant images based on key words, and our Python script takes care of the rest with finding what to add, where, and when.
Challenges we ran into
Google Cloud was initially difficult to set up properly, as well as figuring out how to sync the program with slides.
Accomplishments that we're proud of
We are proud of the fact that we managed to build our first fully voice-based software with no issues or bugs and have it be impactful on our future lives.
What we learned
We learned a lot about how the Google Cloud Speech to Text and Voice API works, as well as a deeper understanding of JSON parsing with the Pixaby API.
What's next for SmartPres
We hope to polish the program and take it to the main stages all across the world!