Back in July, Google announced the availability of a video search tool that would search within a subset of Youtube political videos that had been analyzed using speech-to-text recognition technology. Speech-to-text recognition can be a powerful way to index and better understand the actual video content, especially with videos considered to be infotainment.
, a company that we have covered many times in the past, not only utilizes this technology as a means to better index video content, but also as a means to optimize video content for search. Video SEO is benefited through exposing the full text transcript as additional, relevant on-page text. So, we know how powerful the technology can be.
Today, on the Official Google Blog today, Google announced that they have released Google Audio Indexing (aka GAudi) as its own dedicated site within Google Labs. Although the current index still is limited to the political videos that were utilized earlier, they have added some additional features such as “search within video” and “sharing,” and a more robust user interface.
“Google Audio Indexing is a new technology from Google that allows users to better search and watch videos from various YouTube channels. It uses speech technology to find spoken words inside videos and lets the user jump to the right portion of the video where these words are spoken.”
Of course, you might be wondering why we titled this “Google’s Sexy Audio Indexing…” Well, they have already received some interesting feedback on the Google Labs group discussion:
As Big Oliver points out, on the feedback forum for Google Labs:
“I think from a much higher view, this technology is interesting and will be quite useful… but at the moment, it has the potential for some very embarrassing moments.”
Bob points out that in this video -http://labs.google.com/gaudi?q=%22penis%20of%20my%20boss%22&start=0&n… – the speaker says
“What I have learned from the pain of my loss and from my journey…”
What did GAudio hear?
“What I have learned from the penis of my boss and from my journey…”
Here is an additional example. In this video – http://labs.google.com/gaudi?q=%22penis%20anderson%22&start=0&num=10&… – the speaker states;
“So here I was; a young student; a young boy in Miss Anderson’s class and a seed was planted by that dear woman…”
What did GAudio hear?
“So here I was; a young student; a young boy in Penis Anderson class and as he was planted by that dear woman…”
And then there is this… a more understandable example. In this video -http://labs.google.com/gaudi?q=a%20picture%20of%20penises&start=0&num… it is very difficult to determine what is said, but as Bob points out, it likely was NOT:
“…uh my knees I’ve got a picture of penises there a part of plans…”
All in good fun. Of course, as anyone will tell you that knows anything about speech-recognition, it is incredbily complex, and the above examples are bound to occur. When it does, we can all laugh.
I guess my only question would be; What possessed this person to locate specific instances where the “P” word was used? Hmmmm…… ;-)