Speech Technology Magazine SpeechTEK Conference
 

News Freak Show: A Series of Sad, Hilarious, and Awesome Announcements.

Lauren @ 10:30 am

Hope you had a nice weekend! It’s raining here, and pretty miserable. We’ve been out for awhile due to an issue close, but should be back in action this week. Based on some news from RSS feeds, it looks like the day is getting off to a nice start, news-wise. And yes, that is a picture of William Shatner in his Star Trek days. And yes, this post mentions William Shatner as he relates to voice biometrics.

* BlackBerry announced its new BlackBerry Bold 9000, a new smartphone equipped with 3G capabilities. BlackBerry is one of a line of companies that have beat the new iPhone to the 3G finish line. Apple’s new version of the iPhone is expected to see a release this summer. The Bold phone is, “banking on the … sleek appearance to win over executives and field workers who’ve grown tired of sacrificing style for functionality.” I think that’s PR-speak for, “Please buy this instead of the iPhone. We made it look cooler.” [CRN]

* My favorite British pharmacy, Boots, announced it has chosen Intervoice to develop its natural language self-service application. [TMC]

* Another woman has joined the high ranks in the speech technology field — Susan Almeida will serve as vice president of global services for speech analytics provider CallMiner. You go, girl. [EarthTimes]

* Speech technology for Danish people. Yay! [PR.com]

* Sensory’s BlueGenie Lite = speech synthesis for Bluetooth devices. Todd Mozer, Sensory’s CEO, says in the article: “We’re able to replace all these awkward presses and beeping with speech.” Awkward presses and beeps - GONE. [Wireless Week]

* Every employee’s worst nightmare: Companies’ HR departments using voice biometrics to detect lying about sick days. Don’t ask me what ’skivers’ are; but it’s another funny British word to say. [Birmingham Post]

* Mother’s Day was yesterday, but I just found this article today. It’s about cool presents to buy your mom. My favorite is the ‘SmartShopper’, which uses voice recognition to store grocery lists. Great - groceries! Just what every mom wants to be reminded of on her special day. I’m also imagining the totally awesome possibilities regarding how this device could seriously mess up utterances. [The Sudbury Star]

* When I was growing up, my mom’s idea of enhancing my cognitive abilities was reading to me, and letting me watch PBS. Today? “Smart toys.” Experts say the toys, “contain technological enhancements that enable a child to form dynamic, emotional relationships. Smart toys incorporate microchips, voice recognition and wireless capability so that toy and child can spend quality time together.” This is so incredibly sad. Another ‘Baby Einstein’? I hope not, because it has been proven that ‘Baby Einstein’ does absolutely nothing that would make your child more intelligent. [The Telegraph]

* Umm…apparently William Shatner has an award called ‘William Shatner’s Heartbeat of America’ that he gives away. This year it went to a voice biometrics company. Voice biometrics: the heartbeat of America. [PR Web]

Round Up & Release (Colossal!!! Squid Edition)

Lauren @ 8:31 am

Good morning! This post is coming to you super-early, because I have way too much to do later today. So, before my boss creeps in, here’s all the news for the week so far. Our faithful, intrepid managing editor Len hath returned from the Genesys G-Force conference (read his posts about G-Force here), and he’ll have more to report to you in coming days. Such as what he did with the Genesys-emblazoned belt buckle … and how the rest of the office kind of wants it to add to our wall of “free stuff vendors send us.” So, unless you’re totally intrigued by how Xerox (haha, remember photocopies?) plans to compete with Google and Salesforce, or that giant squid is still freaking you out (hello, it has the world’s largest eyes), follow the jump for less-disgusting news stories! [Speech Tech Blog, Information Week, Associated Press]

(more…)

Meet Jenn

Ryan @ 2:41 pm

The New York Times had a story about Jenn, Alaska Airlines‘ new web-based virtual assistant. I’d compare her to Julie, though Jenn responds vocally only to typed questions. I had a girlfriend like that.

She was developed using Next IT’s ActiveAgent technology - the same technology used for the Army’s Sgt. Star (a name like that makes me think he should be off fighting Cobra Commander, not virtually attending).

The Jenn avatar isn’t annoying and is completely professional. If you curse at her, she says she doesn’t respond to such language and suggests a change in subject.

I’m just not sure if I see the point. I like that Jenn routes you to relevant areas on the company website depending on your question. She’s very helpful (and there I go, referring to her as if she’s a real person) if you want to know about baggage weight limits or about the history of Alaska Airlines. In other words, she’s almost like an auto attendant.

I’m not sure if I understand the point though. What’s wrong with simply optimizing the search feature? Why all the bells and whistles with the avatar and the voice, etc? The Times mentions that Jenn’s “voice has proper inflections.” In listening to her very carefully, her dialogue certainly sounds like synthesized speech. But how difficult is it to smooth the concatenation with such a limited phrase set?

VUI personas are no longer as faddish as they used to be. But I wonder if Jenn’s interface, which is like an IM client, would benefit from something more unique? People on the computer might have more time to futz around than people on a phone. Jenn’s persona is limited. If you ask her about her parents, for instance, she’ll say it’s the people that created her. “I hope they aren’t waiting for Father’s and Mother’s Day presents!” she’ll blurt in a moment of comedic inspiration. The Times writes that an earlier version of Jenn actually answered questions about Alaska Airlines plane crashes - a function that since has been replaced with a more generic “I’m not sure if I understand the question.”

Sure you don’t.

Attach leg to table using screw #jb4 in the socket furthest from the leg facing northwest.

Len @ 2:49 pm

This is Speech Tech senior editor Len Klie’s first post on our blog. Be nice to him. Welcome to the interwebs, Len.

New Jersey-based speech technology vendor iVoice only a few months ago received a U.S. patent for a methodology to make “Talking Consumer Products with Voice Instructions via Wireless Technology.” The company is moving quickly, as today it announced that it has already contracted with a DVD production company to assist with the marketing of the technology.

Jerry Mahoney, CEO of iVoice, says that the company’s invention lets users of new products to activate a speaking package that will take the place of a manual when putting together products. No more mangled hands and bruised ego (or is that bruised hands and mangled ego?) as I simultaneously struggle with trying to hold a half-assembled wooden “guillotine” together with one hand and thumbing through the mini-tome to find out what wrench I’ll need to more permanently connect the pieces with the other. Invariably, that one sentence will appear on page 39 of the 160-page instructions booklet buried deep at the bottom of the box.

The hands-free, eyes-free instructions will make product assembly safer and quicker, but still might not fix my damaged psyche. It will also be a welcome endeavor if it means I no longer have to download and print hundreds of pages of PDF files to learn how to sync my new digital camera to my laptop.

[What we would like to see: A speech recognition component that translates our yelling, anger, and frustration with assembling furniture, and says, "Next time, pay someone else to put this together." -- Ed.]

Lend me an ear

Lauren @ 10:42 am

I’m pretty sure Stelios Arcadiou is copying Van Gogh. The Australian performance artist, who goes by the name Stelarc, recently had a left ear surgically constructed and grafted to his arm. Why? According to an article on the UK Website GazetteLive, the artist wants his work to:

“Challenge perceptions about the human body by exploring its relationship with technology through means including prosthetics and robotics.”

OK, so Van Gogh didn’t have this type of technology when he infamously cut off his ear and mailed it to a lover. But in an age where exhibits like Bodies walk a fine line between educational and exploitative, Arcadiou’s work poses interesting questions about body modification and the possibility of changing our anatomical structure for advancement. Could it be that our original forms are not completely ideal?

In the case of his “Extra Ear Project,”Arcadiou says he plans to mix technology and anatomy by installing a Bluetooth transmitter into the ear.

So, how does this relate to speech technology? As part of an aptly-named “Prosthetic Head Project,” Arcadiou also made a head that can mimic human speech patterns using real-time lip synching and speech synthesis. We’re all for the advancement of speech tech, but a fake talking head? We’ll pass for now.

[Arcadiou story & image via Gazettelive.co.uk]

Previous Posts
Keyword Tags
Archives
© 2008 Speech Technology Media, a division of Information Today, Inc. About/Contacts | PRIVACY POLICY