Voice and Speech Recognition Blog - covering all aspects of speech and its use in technology everywhere
2008-02-01
CBay Trasnscription Systems Using Advance Speech Recognition
CBay Systems a leading provider of medical transcription services has announced it is using M*Modal speech recongtion and language undeerstadning tools. Thsi includes the "most advanced speech understanding service" according to M*Modal. That's another transcription provider using Speech Recognition to improve efficiency. The technology has reached tipping point and those not on board will be left behind
Increasing Useage in Medical Transcription by Transcend (TRCR)
Transcend Services inc (NASDAQ: TRCR), a leader medical transcription services in the U.S. healthcare market announced results year end December 31, 2007 with a significant increase which included an increase from 20% to 24% of editing done using speech recognition tools
Approximately 24% of the Company's total production volume was edited using speech recognition technology in the fourth quarter, compared to 20% in the fourth quarter of 2006. The Company's goal is to grow this percentage to 40% over the next two years, assuming the mix of work on BeyondTXT versus other platforms stays relatively constant. In addition, the Company processed approximately 15% of total volume offshore during the fourth quarter, compared to 7% in the fourth quarter of 2006. Offshore volume as a percentage of total volume is expected to grow gradually over the next several years. We do not expect the growth in offshore volume to impact our domestic workforce.
More physicians moving towards speech recognition - and in a challenging speech recognition environment
Christopher Obetz, M.D., an emergency medicine physician at Abbott Northwestern Hospital in Minneapolis, has been using speech recognition software for two years and likes the faster turnaround time and cost savings. "I'm able to complete my charts and consult other physicians about patients in real time," he says. "In the past, you might not see dictated notes for six to 12 hours, but now it's instantly accessible."
Cutting Time for Physicians with Speech Recognition
Cutting Time for Physicians with Speech Recognition
In a busy Orthopedic surgeons practice in Detroit MI an orthopedic surgeon is using Speech recognition to help him document and communicate all the information he gathers on his patients - 60 in a busy day in his office.
Paperwork for this took hours to complete and included all the forms required for billing/insurance as well as prescriptions and referral letters and his own notes. Like many he used to use dictation and Medical Transcription that not only cost him money but cost him and his patient's time - in many cases days.
But now he is using gloStream gloEMR electronic medical record solution featuring clinical decision tools, document management optimized for clinical workflow and built in speech recognition technology.
In this model gloStream is aiming to replace the costly transcription that small doctors offices pay for each month with a monthly fee for access to their system which will allow them to document and provide all the tools necessary for managing patients, documents, laboratory tests etc and loose the transcription requirement by integrating Speech Recognition into the application all for $1,000
Doctors like Nallamothu gave it high marks, citing in particular its voice-recognition tool.
No details on what speech recongtion solution is powering this applciation but it uses "a Microsoft Vista platform"
Not Everyone Wants Speech Recognition - Gethuman.com
David Hull @ "Field Notes on the Web" is not fully taken with Speech Recognition and he does have a point
...I'm not sure speech is going to take over as completely as one might think. Why do people send text with cell phones? One would be hard-pressed to imagine a more tortuous way of producing text than to thumb it in on a tiny numeric pad, especially before word recognition, but people did, and do, even when they could call, or leave a voice mail.
He's right - Texting was the new way to communicate seasonal greetings this year:
About 65 million texts were sent through O2 on Christmas Day... The three largest mobile phone service providers in the country said more than 62 million messages were sent on December 24 and 25 - five million more than in 2006. Swisscom counted more than 25 million electronic messages on its network, with a little over half sent on Christmas Eve..... One billion text messages are sent every week in the UK.... The Mobile Data Association (MDA) today announced that 4,825 billion messages were sent during September 2007, an average of over 1.2 billion messages every week, staggeringly, the same number of messages sent during the whole of 1999
Anyway - I digress. Speech Recognition is not going to replace all other forms of data capture with technology but it is going to become more pervasive. There are lots of reasons why you might not want to use speech to interact with technology:
Personally I would prefer not to talk to my computer much of the time, either because the environment is noisy (playing havoc with accuracy), or because it's quiet (and I don't want to disturb anyone), or because there's someone else in the room I might like to talk to without confusing the UI.
In short, speech recognition is useful now (particularly if typing is difficult or impossible for a person) and will continue to become more useful as the technology continues to improve, but I don't see it taking over the world.
That said it is already in use on many verbal exchanges taking place today. At least one government (with participation from at least another 3 or more) is listening in in the form of Echelon (allegedly) . And we are increasingly being connected to voice recognition systems on the phone. ON on that point - there are times when it works and it is better but many times the system is so poorly designed and treats the customer/individual poorly we all just want to get to a real human being....enter gethuman.com a site that was set up by Paul English. He's a successful entrepreneur - if I remember correctly he described himself as a serial entrepreneur (Boston Light Software and Kayak) and a philanthropist - a real gentleman.
If you get frustrated with the speech systems online check this site out - it provides you with the quickest way to ....... get a human being.
BlueAnt and Sensory Partner in Voice Interfaces for Bluetooth Devices
"Bringing intuitive Voice User Interface for the Bluetooth headset market where speech control and a complete hands-free control will dramatically enhance the user experience."
the first ever Bluetooth headset with a true voice user interface (VUI). The BlueGenie Voice Interface software suite enables manufacturers of Bluetooth products to integrate full voice control and synthetic speech output without the need for visual displays or complex user interfacing.
Users of the BlueAnt V1 will no longer deal with lengthy or confusing button pressing to access functionality. Instead, the VI can be controlled with simple phrases like "pair headset," "call home," "volume up" and "accept call." When it comes to checking headset status it will not be necessary to interpret confusing beep sequences or LEDs. BlueGenie's voice synthesis capability enables the V1 to speak back to the consumer, letting them know device settings including successful pairing of the headset, battery power level, etc.
I can't be the only one who can't remember the never ending list of button sequences to program my blue tooth headsets..... this has to be better.
Expect the Consumer Electronics Show (CES) coming up to feature many announcements that relate to speech recognition - Parrot who make several innovative products in the wireless world have come out with two new ones for the show
One for drivers of cars (Parrot RK8200) and the other for motorcyclists (Parrot SK4000)
See the Car unit video at CNET videos here and Engadgets review here Motorcycle unit can be seen here and
Apart from listening to music it integrates with Blue Tooth mobile phones providing hands-free function including automatic phone book synchronization and hands-free call functionality through voice recognition and voice synthesis (Text-to-Speech) that reads contact names directly to the user.
The software is multi-user eliminating the need to record names and train the system and dials numbers automatically reading the contact names from the user’s phone book through the earpiece and also identifies radio stations to help the rider/driver select a station without taking his or her hands of the handlebar/wheel.
As the video review said.... who needs wants in car stereo with CD's anymore
Voice Recognition Goes Main Stream in the Nintendo DS
I would describe the youth of today as technically savvy using most of the technology we have today with ease and adding new tools quickly to their everyday life.
So when watching a 9 year old pick up a Nintendo DS and load the Brain Age
I was as surprised as they were to discover it had speech recognition built in and the game was dependent on this technology. But what was interesting was the 9 year old was not familiar with this technology and did not just intuitively know what to do……
This was one of those moments where I realized that speech recognition is still not in the main stream quite yet. Hand a child a remote control and they just know how to use it.
In the case of the Nintendo DS and the Brain Age there was some simple instructions displayed on the screen and some verbal cues. Just before the game started the user is asked a question – “Are you in a place where you can talk”. It did not occur to me or the child that this was of significance. The game/exercise begins – it is a simple Left/Right brain exercise that challenges the mind to look at color and text and separate them
The Four used Blue, Red, Yellow and Black appear on either one or other of the screens (the game is designed to be played so these screens are on the left and right (not in the more typical up down configuration). Each color is displayed and could be simply
BLUE - that’s easy – all you have to do is say Blue. The game recognizes you verbalizing Blue and marks you correct.
But the game also displays that color:
BLACK
In which case the answer is still “Blue” but your brain will need to process this and not say “Black” There are multiple variants
YELLOW, RED, BLACK BLUE
Fort eh color RED and so on (that adds up to a total of 16 possible display choices with only 4 possible answers)
Back to the 9 year old – the first time a color was displayed they just looked at it with no idea what to do. Staring at the screen nothing happens but then as questions are verbalized the screen starts to react to what it is hearing.
See Nicole Kidman in Video using Brain Age
Once it is clear that the solution is to speak the color of the letters the game commences and the novelty of speech recognition that actually works proves to be quite the hit. I can imagine now some variation of the now old movie line from Star Trek IV: The Voyage Home where Scotty asks to use a computer and is shown to a computer. Having been immersed in the 21st century where computers all talk and understand speech he verbalizes his commands with the immortal words of “Hello Computer”. Nothing happens and he looks perplexed and when handed the mouse by a helpful observer he picks it up and says again “Hello Computer”. I can picture this same sequence playing out at least in the short term as the 9 year old and other kids now exposed to the natural interface of speech and the machine have a higher expectation of this technology being present in their everyday interactions with technology.
