Speech Heads, after many a voicemail message and perilously rigorous scientific testing, we’re finally ready to give you STBlog’s assessment of Nuance’s VM2T (voicemail-to-text) client.

How the review breaks down

A couple weeks ago, I had a briefing with Nuance Communications about setting the service up where they explained the lay of the land. They explained that version I would be testing is a little bit different from the one you’ll find out in the wilds of market. As we’ve mentioned before, Nuance’s marketing strategy with VM2T is to distribute through its partners–in this case carriers. Nuance provides the underlying technology to its partners, but each iteration is likely to look a little different according to those partners’ needs. The version I was using was hosted directly by Nuance, so interface specifics probably wouldn’t bear any relation to what most end-users will see.

For one, I had to set up a forwarding service to use it which an end-user would never have to do. For two, all of the messages were emailed to me rather than sent as text messages. In real deployments, Sean Brown, product manager for mobile applications at Nuance, assured me the messages will be sent as SMS texts under most carriers. Also varying from provider to provider are settings for live agent intervention. Depending on what a provider wants to pay for/provide they may bring in real people to clean up the texts if a message scores low-confidence.

All that said, the recognition engine (Dragon 10) is identical to the one that carriers will be using, so we focussed on that for the purposes of this review.

The process began when I set up my account, dialing a number that would, from that brave moment on, forward all my voicemails past my provider’s system to Nuance VM2T HQ. There, they’d be subjected to pinch-and-pull of Nuance’s automated recognition, possible human oversight depending on the strength or weakness of  confidence scores, and spat back out to my email as a text with a .wav of the message attached for review. If the system was unable recognize what was said, it would be indicated this with [...]. Likewise, if it didn’t have high confidence and guessed a word it would write [?] after it.

The results

(more…)

The world of Creepy Talking Robots just got whole lot creepier thanks to Japan’s Little Island.

The company—which recently released this Creepy Talking Wedding Robot—took a giant step forward in bringing about The Rise of the Machines: The availability of the Look Alike Doll.

What is the Look Alike Doll?  Well, Speech-Heads, the Look Alike Doll is a miniature, talking, robotic YOU.

Just send Little Island a photo of yourself, and they will—for a couple thousand dollars—make you a robot counterpart, complete with VoIP calling capability and the ability to read RSS feeds.

A few months ago, I called the company and asked them to build me a Creepy Talking Robot version of my Speech Brother Eric B. Since its arrival, Mini Eric B. has started doing my laundry, gone out for coffee, and brutally assaulted Normal Size Eric B. with a stapler and ballpoint pen.

To make matters worse, the Customized Creepy Talking Wedding Doll that Normal Size Eric B. ordered has since left him for Mini Eric B.

I’m sure I speak for Eric B., when I say: These are Sad Times at Speech HQ.

[editor's note: the look alike doll may be  more of a creepy talking doll than a creepy talking robot.  nevertheless, it is creepy and it talks.]

Election Day 2008 is finally here, and with it comes, what I hope, is the end of the political robocalls. With so many hotly contested races around the country, I would guess that there aren’t too many people who haven’t received one of those annoying recorded phone messages for one candidate or the other.

As the managing editor of Speech Technology magazine, I suppose I should be praising these outbound IVR messages, after all, they are as much a part of what we cover as the speech engine that the doctor uses to dictate the results of Mrs. Stein’s colonoscopy into his electronic medical records system. However, I can’t, in good conscience, do that. Like so many others, I really don’t like these calls all that much.

They might not be so bad if it wasn’t for the fact that they always seem to come at the most inopportune time of the day—like when my prized beagle has just run off with the expensive universal TV/VCR/satellite/DVD player/stereo/microwave oven/coffee pot remote control, ready to treat it like his favorite chew-toy.

I live in New York, and my particular Queens neighborhood this election year was home to a hotly contested State Senate race between a long-seated Republican incumbent and a young Democratic upstart. All the top politicos—including Bill and Hillary Clinton, Gov. David Patterson, and Mayor Mike Bloomberg—recorded robocalls urging voters to back one candidate or the other. Those robocalls blanketed the district yesterday and today. And though I didn’t receive one of those calls—or if I did, I wasn’t home to answer it—I know a good number of people who did. So how did robocall do?

While the candidates, their campaign managers, and political strategists alike would love for us to believe that these last-minute robocalls can sway undecided voters, or at least be helpful in getting a candidate’s name out there, none of the people I’ve spoken to said robocalls like these made a difference to them. Not one claimed to have been influenced by a phone call or to have been so impressed with the automated phone call from Mayor Mike that he decided to throw his much sought-after vote to Bloomberg’s candidate too.

So here’s a lesson to all you would-be politicos: If you really want my vote that badly next year, call me yourself. Don’t lock a political ally in a studio and ask him to record a 10-second message on your behalf, only to have it sent en masse to me and thousands of other voters simultaneously at the 11^th hour. I just might vote for you if you can tell me—in person and for longer than five seconds—why I should vote for you.

Yesterday, we at Speech Tech Blog got TWO emails from readers.

TWO!

Granted both emails were from the same person.  But still.  That increases our reader mail tally by 200 percent!

I hate to admit it, but there are times–picture a man sitting in a cold, dark office, listening to the sounds of homeless people drinking Night Train Wine and sleeping in the hallway–when I begin to have doubts; I begin to wonder: does anyone read the Speech Technology Magazine Blog?

Well, the answer is evidently YES.

Our aforementioned faithful reader emailed me with some feedback about Conversations 2008 and suggested that I make it easier for readers to contact me.  So, to that end, if anyone feels a burning desire to send me an email as opposed to just leaving a comment, please send communiques to aboretz@infotoday.com.

That said, blog comments are also appreciated.  As are kind thoughts.  As are even relatively kind thoughts.