I’m writing this from the first annual Voice Search Conference in San Diego. I’m also banging this stuff out real quick-like so I apologize in advance for any slovenliness of form and grammar. Additionally, the chambermaid at the Chateau de Marriott is battering the door because she insists my bed needs to be made and mints strewn lovingly across my pillow. All very distracting.

In the meantime, some scattered thoughts:

- First, somebody really needs to come up with names distinguishing voice-powered search versus audio mining (there’s ambiguity because both are often labeled “voice search”) This convention, based on the opening keynote (which I covered in today’s Speech Tech news) and the titles of upcoming panels, focuses pretty much on voice-powered search. There is a panel tomorrow called “Searching Audio/Video Sources On The Web And In Enterprises” that I will attend. Though so far, it looks like the only panel dedicated to audio mining.

- Given the great potential for audio mining (i.e., more accurate multimedia searches, targeted advertising, etc), it would be nice to see greater representation. Of course, the convention just got started, so if I’m wrong, I’ll be sure to post a picture of myself devouring a crow [We hope so -Ed.].

More on Ryan’s [mis]adventures, after the jump.

Just got through this wacky panel called “Delivering High-Volume Voice Search Applications.”

Spoken Communciations gave a presentation. They do agent-assisted IVRs (check out Lauren Shopp’s feature in this month’s issue of Speech Tech. Which, incidentally, was given out for FREE along with the Official Voice Search Conference Tote. That’s good reading!).

So why was Spoken here? I don’t know. Haven’t a clue. The company uses human agents (or “guides” as they call them) to assist the IVR during complex customer interactions.

All well and good, but I’m hard-pressed to think of a human as a “voice search application” per se.

Valantine Matula from Avaya presented next. When he found out there were very few contact center technicians in the audience (two of maybe around 70) he began a slideshow depicting the physical setup of speech servers.

Matula definitely knew his stuff, but this presentation felt off for a voice search convention. Again, I wanted to learn about voice search applications, not just voice applications.

["Artwork" provided by Lauren.]