This repository was archived by the owner on Nov 20, 2023. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathScript
More file actions
32 lines (17 loc) · 2.68 KB
/
Script
File metadata and controls
32 lines (17 loc) · 2.68 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
#####this will be our script for the video - feel free to make edits
*Slowly start speeding up the below paragraph*
Hey everyone! This is Metrophon, and we’re so excited to present this video to you today. But before we begin, we want to thank our sponsors, McDonald’s and Burger King. As you watch the video…
*Zoom out of that frame, switch to normal speed*
Imagine listening to the fluff of a YouTube video…at full speed. Now, students are getting smart, increasing the speed of videos to get through them faster. But then, just as you speed up the video, the content starts getting complex, forcing you to slow the speed down to understand.
This is all due to something called the information density of educational content, aka the complexity of information presented in a given segment of a video.
Our team at Metrophon has realized that information density varies throughout a video, giving you that urge to speed up the simple stuff and slow down the hard content.
But what if we told you that you’ll never have to sit through a video again, figuring out the right speed to watch it at? Metrophon’s algorithms calculate the information density of each segment of a video–we then standardize it throughout the video by increasing or decreasing the segment speed based on whether the information density is low or high.
Let’s walk through it together.
*Demo*
How did it feel to learn at the right speed, for once? Hopefully, it was blissful.
Metrophon uses AI to bring this software to your computer. First, we use Youtube’s latest feature - heatmaps - which shows the most replayed sections of a video, which presumably coincide with the video segments that are the most difficult to understand.
We then use these heatmaps as labels for training data. Along with that, we use symbl.ai to create a summary of content of the video. Using this, we analyze the complexity of each one-minute segment of the video by measuring parameters such as the length of phrases, the sum of the complexity level of the segment’s words, and more.
Finally, we use this complexity in order to output the segment at a speed that makes sense.
On average, people spend 16 hours every week watching video content. Furthermore, the majority of Americans watch Youtube to learn new things, taking initiative to explore their horizons and fulfill their potential. With Metrophon, every learner can reach their educational goals faster, by viewing content at a pace that engages them, without burning them out.
With this same technology, Metrophon can filter through everything from podcast episodes to audiobooks to read to you at the right pace.
So what do you say? Are you ready to broaden your horizons in your own time?