Generates beautiful slides with content in real time based on the presenters' speech and context
Lets you dynamically add images, automatically chooses the best layout for your content
Eliminates the work required in formatting slides
How we built it
The presenters' speech is recognized using Google Cloud Speech API, then the Natural Language is parsed using API.ai to understand the context and generate data.
The layout changes to match the content put in each slide and images can be sourced from the web by the user by just mentioning the object name using the Microsoft Bing Search API.
Front-end is written in Facebook's React.js while the backend is made using node.js with webpack for bundling.
Challenges we ran into
Continuous background recognition of speech
Training of Natural Language Parser for the specific use case
Asynchronous calls and client/server interaction
Accomplishments that we're proud of
Being able to complete a fully functional prototype written with design patterns in mind
Having a beautiful customizable user interface that can serve both educational and professional purposes
What we learned
Enhancing our skillset with the state of the art tech
Having a logical project plan before jumping into coding really goes a long way
Talk is cheap, show people the code
What's next for Slides In Time
Ironing out bugs and getting it ready for deployment
Log in or sign up for Devpost to join the conversation.