It's always boring to look through youtube videos and not know when the content you want to watch is going to come. What if you could search for the exact spot you want in a video?
What it does
After entering a search word, a set of video times near when the word is said is displayed as links. These can then be clicked to skip straight to when the person talks about a specific topic.
How We built it
We used HTML and CSS to do the front end. For the backend, we used Python, specifically the Flask web framework. For the implementation of the algorithm, we had to use the google-cloud api as well as the google web-speech api to tell where in the video the word is said.
Challenges I ran into
It was very hard to figure out how to convert video to audio, specifically as it pertains to youtube. It was also difficult to manipulate files that weren't necessarily text, in this case wav files.
Accomplishments that I'm proud of
We were able to completely finish the application, which is normally a very rare occurrence with hackathons. We were able to implement something in the back-end and used google api's.
What I learned
Gained a deeper understanding of Flask and google api's
What's next for C⌘me and + Find
There are still a lot of improvements. For example, we could do straight video to audio conversion for ourselves. Also, the algorithm could be more efficient, and the website could be a little more iteractive.