The Future of Indexing Video is Goog-411??

Carrie Hill | 30 December 2007 |

For those of you familiar with the many “pools” Google dips their toes into - you may have heard of GOOG-411. Those of you who dont know, might like to try it out. Bascially it’s a free telephone information service that will give you a phone number, an address or even send a map to a location to your cell phone. I’ve used it once or twice and it’s pretty handy.

The most interesting aspect of Goog-411 is what Google is actually USING it to accomplish. I hadn’t thought about there being an alterior motive to a platform that didnt seem to have any money making potential for Google, until now.

I read today over at Garret Roger’s blog “Googling Google” what the real “motive” is behind Goog-411 - they’re using to to train a voice to text software. Now - in and of itself that’s not earth shattering news. But if you know anything about online VIDEO - then you’ll know the search engines arent able to index the CONTENT within a video - and they dont even attempt to index video on an individual’s website. Right now the only way to get your video included in Google’s index is to have it submitted to a search engine that FEEDS video to the big “G.” So - with spoken word to text technology - any video can be turned into a text transcript and therefore indexed. This would be a huge breakthrough for Google and they are probably a lot closer to having to work than anyone else.

I went over to Bill Slawski’s SEO By The Sea blog to see if I could find any Google Patent Filing information on this technology. For those of you unfamiliar with this blog but interested in how Google develops their technology - you must give it a read. Bill has a great way of breaking down patent filings and giving us practical and real-life answers to questions we have about what the search engines are up to. I didn’t seem to get very far looking for video indexing patent information. That doesn’t mean it’s not there but I couldnt find it in 5 minutes or less - which means I’ll keep looking and let you know what I find. I imagine if there ISN’T a patent filing already, we will see one very soon.

Rogers quotes an interview with Google’s VP of Search Products and User Experience, Marissa Mayer, over at InfoWorld wherein she reveals the real “meat” of what Google hopes to accomplish with GOOG-411:

“Whether or not free-411 is a profitable business unto itself is yet to be seen. I myself am somewhat skeptical. The reason we really did it is because we need to build a great speech-to-text model … that we can use for all kinds of different things, including video search.”

She then goes on to elaborate on the process of building a robust video search engine:

“The speech recognition experts that we have say: If you want us to build a really robust speech model, we need a lot of phonemes, which is a syllable as spoken by a particular voice with a particular intonation. So we need a lot of people talking, saying things so that we can ultimately train off of that. … So 1-800-GOOG-411 is about that: Getting a bunch of different speech samples so that when you call up or we’re trying to get the voice out of video, we can do it with high accuracy.”

So what does this mean NOW? Well for you and I it means that SOMEDAY Google (and eventually the other search engines) will be able to index the spoken word in our videos. Today the best way to get a video indexed is to have it added to one of the Video search engines - YouTube, Yahoo! Video, MetaCrawler and the like. As far as ranking your site for what is IN the video - embed the video on your site and add a FULL TEXT transcript wrapped around it. Use good metadata in your video is key to having it ranked in the video engines. Learn more about how to build an optimized video in our whitepaper on Optimization for Images and Video