3 ms·

It'd be great if we could extract subtitles from video and put into a nice text format that's easy to read so it's skimmable. Then when you get to an area of in

It'd be great if we could extract subtitles from video and put into a nice text format that's easy to read so it's skimmable. Then when you get to an area of interest, you should be able to click the place where you're interested and a video will pop up and start from there. Then when you've heard what you wanted to hear you can minimize the video and skim the text for the next area of interest and so on.

enobrev 10y ago
I imagine there could be an interesting means of moving the playhead along with your skimming, so when you highlight a block of text (as I do, habitually), the video would play through the selection and pause at the end. I'd love to use text as a video controller.
- trevyn 10y ago
  This is a great idea -- text is much faster to read, but watching a speaker is much more information-dense due to body language and intonation. This would be a fun hack. Looks like the YouTube caption data is available via API: https://developers.google.com/youtube/v3/docs/captions/download https://developers.google.com/youtube/v3/docs/captions/downl...
  - hanniabu 10y ago
    I can't try this out at the moment, but from looking at the documentation I don't see it say that it returns the timestamp of the caption anywhere.
- hanniabu 10y ago
  Yes, this is pretty much what I was picturing except without having to select the end, this way it doesn't just cut off when you were still interested in what was being said even though you might not have planned on watching past that particular point.
y14 10y ago
This is a problem we (accessible.ai) have been working on in the past few months. We're still tuning and improving but with a multi-modal approach of signal processing, speech recognition and text analysis we were able to come up with a pretty good MVP to solve this issue. You can take a look at our small demo here: http://accessible.ai/nav http://accessible.ai/nav
gliese1337 10y ago
This is in fact exactly the approach that my lab has been working on for the last several years in developing a media player for language-learning applications (which has actually also been used for teaching linguistics and introductory chemistry, so clearly the pedagogical benefits generalize beyond just language learning). Being able to switch between text-centric and video-centric views freely has been a goal of my (now-former, because he just retired) boss for a while (and by "a while" I mean he did his dissertation on interactive, text-augmented video technology with applications to education in the 70s), and we're not quite there yet, but we do have the ability to show a transcript that can be synchronized with the playing video (automatically scrolling and with highlighted text to show where the video is), or scrolled freely and searched through to select a specific part of the video to skip to. You can see an implementation of the technology (including playing around with some of our editing tools) at http://ayamel.byu.edu http://ayamel.byu.edu

enobrev 10y ago

I imagine there could be an interesting means of moving the playhead along with your skimming, so when you highlight a block of text (as I do, habitually), the video would play through the selection and pause at the end. I'd love to use text as a video controller.

trevyn 10y ago
This is a great idea -- text is much faster to read, but watching a speaker is much more information-dense due to body language and intonation. This would be a fun hack. Looks like the YouTube caption data is available via API: https://developers.google.com/youtube/v3/docs/captions/download https://developers.google.com/youtube/v3/docs/captions/downl...
- hanniabu 10y ago
  I can't try this out at the moment, but from looking at the documentation I don't see it say that it returns the timestamp of the caption anywhere.
hanniabu 10y ago
Yes, this is pretty much what I was picturing except without having to select the end, this way it doesn't just cut off when you were still interested in what was being said even though you might not have planned on watching past that particular point.

y14 10y ago

This is a problem we (accessible.ai) have been working on in the past few months. We're still tuning and improving but with a multi-modal approach of signal processing, speech recognition and text analysis we were able to come up with a pretty good MVP to solve this issue. You can take a look at our small demo here: http://accessible.ai/nav http://accessible.ai/nav

gliese1337 10y ago

This is in fact exactly the approach that my lab has been working on for the last several years in developing a media player for language-learning applications (which has actually also been used for teaching linguistics and introductory chemistry, so clearly the pedagogical benefits generalize beyond just language learning). Being able to switch between text-centric and video-centric views freely has been a goal of my (now-former, because he just retired) boss for a while (and by "a while" I mean he did his dissertation on interactive, text-augmented video technology with applications to education in the 70s), and we're not quite there yet, but we do have the ability to show a transcript that can be synchronized with the playing video (automatically scrolling and with highlighted text to show where the video is), or scrolled freely and searched through to select a specific part of the video to skip to. You can see an implementation of the technology (including playing around with some of our editing tools) at http://ayamel.byu.edu http://ayamel.byu.edu