VoiceRecognition
Latest
Cyberpyrot's Xbox 360 voice recognition controller mod
Usually we spot console modders adding a few LEDs to their boxen, squeezing in some extra memory or perhaps even changing a form factor altogether, but it's rare to see these hackers actually hack in some original functionality. Cyberpyrot over at Acid Mods is taking this whole modding business to a whole new level with his voice recognition controller for the Xbox 360, which involves some heavy hacking and a bit of hardware kit wizardry to let you control teh Halos with merely the silky smooth sound of your voice. The mod is ostensibly designed to help out the mobility impaired, but we all know it's going to be primarily used by slothful nerds (you know, those "purists" that didn't pick up a Wii) who have decided that even repetitive finger motions are too much of an effort. The current mod can handle 10 different voice commands, check out a video of it in action after the break.[Thanks, GrandpaSmurf]
Pioneer's AVIC-F500BT GPS nagivator offers voice control of iPods and Bluetooth cellphones
If you were with us during the Pioneer press conference, then their new AVIC LINC system is old news to you tough guy. However, the details are new. Designed to integrate with any factory car audio installation, Pioneer's new AVIC-F500BT system offers the ability to add in-car navigation and extended entertainment without modifying your existing car stereo. The primary feature is the VoiceBox Conversational Voice Search Platform -- an advanced voice technology for control over Apple's ubiquitous iPod and Bluetooth enabled devices. It's sophisticated enough to analyze and then execute on commands as varied as "I want to hear the artist U2" or "play U2" or even a Foghorned, "Why I say, why I say boy, um, play U2." It also features a 5.8-inch, WVGA touch screen, text to speech, and the ability to accept media via a USB bunged device or SD cards for MP3 or WMA playback. New POI's can be added via SD cards and a rechargeable battery lets you take it on the go. Look for it this Spring for an undisclosed price.
Siemens developing call center voice-recognition software that determines age and gender
We're not certain why call center employees need even more reasons to treat customers condescendingly, but Siemens has announced that it's developing tech that will "intelligently" assign reps to callers based on determinations of age and gender. No word on how the system makes the determinations, but we're not going to be happy if we have to start doing Marilyn Monroe impressions just to get someone on the phone.[Via The Raw Feed]
Pudding Media eavesdrops on internet calls, displays relevant ads
Pudding Media, a California-based startup, is opening up a beta test of its internet calling service, and while calls are said to be completely free, there's most definitely a catch. Users making the call will be presented with advertisements that actually relate to the conversation taking place, as the company's speech recognition software picks out key words and beams in ads based on what you're yappin' about. Notably, the company actually seems quite interested in licensing its technology out rather than becoming "an independent provider of ad-financed internet phone calls," and while we're sure hordes of privacy advocates will be none too pleased with the setup, those without issue can head on over and give it a try today.[Via Yahoo]
VoiceSignal ports voice recognition software to iPhone
Those not preoccupied with unlocking their iPhone may be interested in what VoiceSignal's talking about, as it has apparently ported several of its applications to Apple's handset. Currently deemed "proof-of-concept applications," both VSearch (speak for search keywords) and VTunes (speak a band you'd like to hear) enable users to simply talk to their mobile and allow the software to handle the rest. Of course, speech recognition apps can be explained much better with, you know, sound, so be sure and check out the video of VTunes in action after the break.
Azentek stocking DiMora's $2 million ride with in-car electronics
If you thought the MyGIG or even Bose's Media System was high-class, Azentek is out to change your perception. Recently, the firm inked a deal with DiMora Motorcar "to provide high-performance computers for the Natalia SLS 2 sport luxury sedan." The ride, if you weren't aware, will be a hand-crafted $2 million, 16-cylinder production automobile, and will sport "at least two Atlas units" which boast voice recognition, FM / XM Radio / Sirius / HD Radio, Bluetooth, CD / MP3 / WMA / DVD support, GPS navigation, VoIP / cellphone integration, "over a terabyte" of storage, and Windows Vista for good measure. Additionally, users will be able to interface with the machines via touchscreen, keyboard / mouse, a handset, USB, auxiliary inputs, etc. Sure makes Blue&Me look like child's play, no?
Grundig's DigtaSonicMic, a high-end USB dictaphone
We're not certain who still uses dictation machines -- isn't illegal to not know how to type? -- but Grundig's ready to drag you into the future with its DigtaSonicMic, which features USB sound in / out as well as trackpoint mouse control so you can manipulate files on your machine as you walk around the room pompously dictating away. The unit also interfaces with Grundig's full line of dictation and transcription software, allowing it to plug into your network and offload your spoken-word brilliance to a remote transcriptionist. No word on price or availability, but we're guessing most of you can think of dozen alternate solutions that'll work just as well.
Drug kingpin IDed using advanced voice identification
Advanced voice recognition technology busted one Juan Carlos Ramirez Abadia recently, a drug kingpin who had drastic plastic surgery in a failed attempt to mask his true identity. Because of his plastic surgery, Brazilian Police were unable to gain positive identification required for an arrest warrant. Instead, Colombian officials provided the US Drug Enforcement Agency with taped conversations, which the US body was able to match to Ramirez Abadia. It's not clear how advanced the technology actually is, since the Government keeps it under wraps: judging by the importance of this arrest, and the impending trial, we'd hope that it's beyond the capabilities of regular off-the-shelf desktop software.[Via PlasticSurgery101]
David Pogue blogs mini-review of Parallels 3
David Pogue, the NY Times' answer to Walt Mossberg (as distinct from "This WSJ writer, nicknamed Uncle Walt, hosted two technology legends at a 2007 conference" -- that's the Jeopardy answer to Walt Mossberg) hasn't reviewed Parallels 3.0 yet, at least not in print. On his blog, however, Pogue has given the new version an enthusiastic thumbs up.Pogue uses Dragon NaturallySpeaking on Windows to do a large chunk of his writing (Mac voice-rec tools like TrueVoice or iListen, though available, did not measure up to his needs) which has meant either lugging two laptops around or, more recently, Boot Camp. Up until now the USB audio support in Parallels hasn't worked well enough for speech recognition to go smoothly. With the 3.0 release and an XP virtual machine (Vista was still too resource-intensive), Pogue is now running the voice-rec app completely flawlessly and happily.Pogue's post ends with a reminder of how the platform equation is changing; he writes, "Whether you're a Mac person or a Windows person, the point is that you can now run 100 percent of the world's computer software on a single machine, faster and more easily than ever." It's no longer a matter of justifying answers to "Why would you buy a Mac?" but simply responding "Why wouldn't you buy a computer that can run all your applications, whichever platform you need?"
Hanwang pen tablet offers voice correction
Just under a week ago, Hanwang was out boasting about its handwriting recognition mouse, which sported a built-in writing area that solved the need for one of those huge platters to be lugged around; now, however, the company is indeed showing off one of those aforementioned tablets at CHITEC 2007. While details are obviously scant, the folks at AVing were able to catch some hands-on time with the writing device at the expo, and it looks to do a fair job of converting your handwriting to text if you're more comfortable wielding a pen than searching for the home row. Of course, if it does happen to get off track, it purportedly touts a "voice correction system to collate the inputs." The WACOM-styled unit also touts a number of hotkeys to go along with the stylus and decorative design scheme, but unfortunately pricing and availability still remain a mystery. Click on through for a few more sneak peeks.
NTT DoCoMo handset "slows down" cellphone voices for the elderly
We love those movies where the kidnappers use voice-altering equipment to talk to the authorities and such -- lower voice octaves with seemingly lower speed rule, yes? The elderly of Japan may like that concept as well, as Japanese carrier NTT DoCoMo is about to roll out the "Raku-Raku" handset that personifies that functional but useful line for customers that just want to make calls. For the elderly, that's usually the only goal with a cellphone. An extra monthly revenue stream for NTT DoCoMo is not bad either from a business perspective. What is killer about the Raku-Raku is the "Slow Voice" feature. It actually slows down spoken words coming into the handset itself for those that prefer a more laid-back conversation instead of a caffeine-fueled one. Other goodies include noise compensation that raises or lowers mic sensitivity and incoming audio based on ambient noise and voice reading of SMS and email messages. Seems like NTT DoCoMo feeds the power user to the basic user in very decent fashion, yes?[Via Digital World Tokyo]
SmartShopper prints out your shopping list using voice recognition
Don't you just hate shopping for groceries? Being around other people, having to spend time thinking about what you want to buy, and the inevitable "shopping hangover" caused by the later realization that you bought a dozen things you don't need or want: yuck. A lot of people head online for their shopping (where else can you read Engadget while you order loo roll?), but for those who prefer to take their cash into the real world, there's now a solution that could prevent the likelihood of encountering one of those aforementioned shopping hangovers: SmartShopper is a device that can convert a spoken shopping list into a printout to carry around in your wallet or purse. Just mount the SmartShopper on your fridge using the magnet, start waxing lyrical about apples and oranges -- don't worry, your neighbors already think you're crazy -- and the little gray box will neatly print all your choices onto a little piece of paper. At $149, the SmartShopper is on the low end of smart fridge solutions, but at least you can feel safe knowing that its limited artificial intelligence prevents it from sticking "5 kg of Soylent Green" on your list.[Via The Raw Feed]
IWARD nursebot looks to clean hospitals, fend off intruders
We all know security gets a little heavy eyed when the wee hours of the morning roll around, and we highly doubt the janitorial staff is humming along at maximum efficiency when the residents are snoozing away, so implementing a robot to tackle both tasks seems quite practical. The IWARD project hopes to develop a "nursebot" that wears several hats, and can handle cleaning up spills, utilizing face and voice recognition technology to "communicate with patients and spot unauthorized visitors," and even working in "swarms" to distribute tasks between the robotic crew. Researchers are aiming to have a three bot prototype ready to rock by 2010, and want to integrate sensors and camera to avoid collisions whilst "traveling along high-speed lanes in the hospital corridors." Better steer grandma's wheelchair clear of the robotic raceway, eh?[Via TechieDiva]
XM vehicle to showcase real-time "infotainment"
Although XM's supererogatory features that compliment the actual music transmissions have been available for awhile, the company is apparently going all out to demonstrate its latest inclusions at CES. Dubbed the "infotainment" concept vehicle, this mysterious whip will reportedly feature the "first personal weather tracking system for GPS navigation," which is intended to compliment XM's real-time traffic service (NavTraffic). The system is designed to focus on weather patterns for your specific route, and also provides short-term forecasts to help you decide whether the scenic route would suit you better. Partnering with Baron Services, XM plans to bring the WxWorx technology already used in "airplanes, boats, and emergency responders" to the comforts of your ride. Additionally, the company is set to showcase "in-car video systems" co-developed by On2 Technologies that can not only receive audio, but even video (sound familiar?) from the same satellites and ground repeaters that XM already delivers tunes by. Of course, we can expect to see a fleshed-out demonstration of the parking locator technology (dubbed ParkingLink) that has been in the works, and XM even plans to demonstrate a voice-recognition control scheme developed by VoiceBox Technologies. Notably, XM suggested that its weather-tracking system should be officially available "this summer," while the other introductions might just hold off until this whole "sorta maybe merger" thing gets ironed out.
BMW teaser unveils enigmatic do-it-all device
BMW's fairly well known for its nifty in-car implementation of technology, and although certain aspects have received their fair share of criticism, the firm has something brewing that could cause quite a stir -- of course, that would require us to actually know what it is. A snazzy flash website unveils a sleek, continually morphing device that touts the ability to play MP3s, DVDs and CDs, has USB and Bluetooth connectivity options, can recognize voice commands, and has a 65,000 color LCD to boot; but in the end, we're still left to wonder how it relates to the company's vehicles (if at all). Sure, we've seen BMW-branded watches and the like, but it's not exactly usual to see an automaker producing multifaceted consumer electronics, and while it may end up just another technological aspect of a car, it looks like something you just might tote around. Thankfully, we shouldn't have to wait too awfully long until the veil is removed, as it should get official in just a few weeks, but until then, be sure to click on through for a few more glamor shots.[Via Autoblog]
RoomRender system dictates living environment, summons Segway
While the choices are nearly limitless when it comes to automating your crib, the novelty factor of turning your microwave on from your bedroom will probably fade with time. Japan's SGI is aiming to making home automation a seamless, all encompassing affair with the RoomRender system. The self-proclaimed "intelligent room system" can not only control all the electronics, lighting systems, and appliances in a given room based on voice commands, but can even release perfumes and dim the lighting in response to emotions. Aimed at luxury hotels, board rooms, and A-class superstars, the system relies on AmiVoice technology to recognize spoken words, which can be used to automatically record specific programs, turn on your home stereo, or grind away those leftovers in the trash disposal. Additionally, the "FeelingWall" component relies on Sensibility Technology (co-developed with AGI) to interpret emotions "based on the intonation and rhythm of voices, adjusting the color and lighting accordingly." The pinnacle of all this, however, is the planned ability to call upon Segways if necessary; a user can demand that his / her Segway be cranked up and rolled over to them, so they don't even have to get up to fetch the reverse-minded machine. Although we're not sure when these will be rolled out in Japan, a working system has already been installed in the company's own office building, and it plans on fetching between ¥5,000,000 ($43,464) and ¥6,000,000 ($52,156) when they go mass market.[Via Pink Tentacle]
World's fastest texter gets pwned by voice recognition
A young lad by the name of Ben Cook might be able to school his pals with his lightning-quick text messaging skillz (he even prefers plain ol' multi-tapping to predictive text) but one company set out to prove that the world's fastest texter is still no match for voice recognition. Nuance Communications is gearing up to release its Mobile Speech Platform to carriers -- a system that'll ultimately allow users to dictate SMSes, among other tasks -- and they obviously needed some high-profile way to get the word out, like picking on Ben Cook, for example. For the contest, Nuance pitted its software against Mr. Cook and two Nuance employees packing QWERTY and T9 devices, throwing them each this little tidbit of everyday conversation: "The razor toothed piranhas of the genera Serrasalmus and Pygo centrus are the most ferocious freshwater fish in the world. In reality they seldom attack a human." Our world record holder finished in 48 seconds -- not bad, considering he was doing it without any predictive assistance -- but the software spanked everyone with just 16 seconds on the clock. Now we just need to take a hard look at why one would dictate an SMS (versus, say, placing a phone call) and we'll be jumping right on the bandwagon.
Pioneer AVIC-HD1BT brings plain speech voice-activated GPS units
Yeah, we love GPS around here as much as the next guy, but it really doesn't impress our lady friends too much when our mellifluous prose gets interrupted by Cro Magnon-like gestures while we have to ask for directions on our nav units. We'd much prefer to use a phrase like "Find me a hotel in Paris in the seizième arrondissement, play Serge Gainsbourg's 'Je t'aime' and adjust for mood lighting -- s'il te plaît" and have our ride respond instantly. Well, save for that last request, apparently plain English queries (or 13 other languges) to a GPS stystem are now possible thanks to software by IBM built onto Pioneer's new AVIC-HD1BT nav unit. It's loaded up with a 30GB hard drive (20GB for maps and the rest for music), a DVD player and an iPod connection of some sort. Although Pioneer claims that the AVIC-HD1BT is for the European-only non-"luxury car market" driver (sorry, us Lexus-driving Yankees will have to suffer), it still doesn't come cheap: €2,500 ($3,145). [Via IDG News Service]
Daewoo's voice-recognizing microwave
Apparently being able to zap your meal in favor of kindling up a fire using sticks and lighter fluid just isn't efficient enough. If you'd prefer to maximize your couch time even more by not even mashing buttons on the face of your microwave, Daewoo's new device hears you loud and clear. Sporting a countertop design (as opposed to the "in-a-drawer" variety), brushed aluminum finish, and ironically enough, a full-fledged keypad, the voice recognition microwave recalls up to 40 cooking commands to get things fired up with just a shout. Niceties include its ability to react to anyone's voice (not just the programmer) as well as automatically determining the time required to properly heat / cook your food. Daewoo claims a "95 percent accuracy" rate on its automated determinations, so be sure not to scold it too fiercely when it torches your dinners the other 5 percent of the time.[Via OhGizmo]
