For some time I have been investigating the possibilities of embedding metadata in video’s. The sad conclusion is that it can’t be done when you use Final Cut Pro and deliver in Flash video: “No, we can’t”.
Why would I want metadata? Because search engines don’t have eyes and thus can’t determine the content of videos. If they could, they would index it and link to the content of your video in the same way they link to text webpages. Metadata adds a text description to video’s that makes it understandable for search engines.
There are more uses to metadata but this is the most important one.
Search engines bypass this problem by looking at the text information close to the video. Filenames, headers, page titles, tags, descriptions. This way the main subjects of the video can be determined. Real video metadata would be way more accurate and offer the opportunity to link to certain parts in a video. Say you have a 40-minute documentary with multiple interviews, then a search engine could link to each interview separately.
This ability would be nothing short of a revolution to internet video. It enables a plethora of new formats, free from duration, free from source, even free from editing. Video would become part of the semantic web, interweaved with text, pictures, audio end video from multiple sources.
And Final Cut Pro can’t do it. But Adobe Premiere Pro can! It has this really cool voice recognition feature that transcribes spoken word to text. This text can be used as the basis for subtitles, transcripts, video metadata, closed captioning or whatever (navigational) use you can think of. Or add your own video metadata. See how it works.
I haven’t used it yet but I’m exited. This could be the breakthrough I have been looking for for years. Video, compared to text, is so static. Accurate video metadata gives us the handles to play with it. Explore new formats. Innovate beyond what we can imagine.