Sunday, 8th October 2006
Multimedia Search Tech Provider Nexidia Unveils Public Demo on Atlanta TV Station Site
A multimedia search product we've mentioned many times on ResourceShelf is Nexidia.
Nexidia is a multimedia search company out of Georgia that creates a searchable corpus (words and phrases) but does it unlike other products that provide transcript (every word spoken) search.
Nexidia takes words and phrases and breaks them down into phonetic sounds (phonemes) and then indexes them. This allows for faster indexing (65x real time), reduced computer power requirements, and, what's most important, more precise indexing and search retrieval.
Three other ways to search video/audio are speech-to-text, closed captioning, and metadata. Of course not every show, conference, etc., offers closed captioning and those of us with tools that can transcribe our words to text know it often takes some time and training to get it correct.
About 40 phonemes exist in every language with about 400 in all spoken languages. Trust us, the Nexidia documentation and explanation makes for some interesting reading including this page. We think it's easy to see how easy and quick an installation could be ready to go. The only negative is that no text transcript is created.
To this point, it's been difficult to demo Nexidia technology since there haven't been any public demo sites. Most of their business is with private companies (recording call center chats, for example) and the government.
However, as of today, we have a publicly accessible demo to take a look at with Nexidia. It comes from Channel 11 (WXIA) in Atlanta and allows you to keyword search all of their news programming (no sports) plus some exclusive web footage. Look for the search box in the middle of the page. Of course, don't forget that this is a beta release.
There are several ways to limit by date range, and note the advanced search page where you can also tell the database within how many seconds you want your search terms to be spoken within each other. For example, within 10 seconds x number of words need to be spoken. Note: You will need Active X installed.
When you find your result(s), click on the link and go precisely to where those words are spoken. Amazing and from what I've seen with some very limited testing, very accurate. Results are rated as to how accurate they are versus your search terms.
With the speed and accuracy of indexing and the need not to have to train a speech-to-text tool, we think the day is rapidly approaching when a student can come home from a lecture, plug in their recording device, and within minutes be able to keyword search the material right from their desktop or laptop. Think of all of the other possibilities.
More than worth a look not only for the technology but for the concepts and how this type of indexing might be an example of the future, today.