Midday, A leading daily tabloid in Bangalore, Writes about Speech recognition.

No more keys!. Mid Day, Bangalore. Author: Priyadarshini Nandy Date: 27 May 2008 Looking to save time on typing out letters? Ever considered a speech recognition software?Things need to get typed and not everyone is an ace at the keyboard. Fortunately, one doesn’t always have to be a keyboard champ. Thanks to voice recognition software that converts speech to text, all one has to do now is talk and leave the typing to the software. There are a handful of good speech-to-text software that are available in the market, but a user should always test the product out before making the purchase.Speak awayMicrosoft Windows’ speech recognition enables the operating system to convert spoken words to written text. An internal driver, called an speech recognition engine, recognises words and converts them to text.The speech recognition engine may be installed with the operating system or at a later time with other software. The Microsoft Speech Recognition Training Wizard (Voice Training Wizard) guides you through the process, recommends the best position to place the microphone, and allows you to test it for optimal results.For the impairedThere is an increasing use of speech-to-text technologies to provide communication access for those who cannot hear, or otherwise cannot directly perceive spoken language. Caption Mic is a do-it-yourself captioning system that produces captions with the help of speech recognition technology. A voice captioner echoes what is spoken at an event, and the resulting captions are displayed on a TV monitor or other display device. C-Print is a speech-to-text system developed at the National Technical Institute for the Deaf (NTID), a college of Rochester Institute of Technology (RIT), as a communication access service option for deaf and hard-of-hearing students in educational environments. TypeWell is a system for transcribing speech to text. It is an effective learning tool for deaf and hard of hearing people, as well as those with visual, physical or learning difficulties.e-Speaking, another software that can only be downloaded from the web comes with over 100 commands built-in and ability to add more commands. It runs on Windows2000 and WindowsXP, utilises latest technologies from Microsoft and seamlesly integrate with MS Office and is easy to use.Latest additionNuance, a 10-year old company has also launched their speech to text software called Dragon Naturally Speaking. Of all the software available, this is the only one that comes with and Indian edition. “It takes the software seven minutes to recognise speech and customise it and delivers up to 95 per cent accuracy. Dragon works on pretty much all sorts of text-related software,” says Manish Rajkumar Goenka, head speech and imaging department.The software has seen a clientele that ranges from the legal offices, corporate houses to even journalists. “We also have visually-impaired people using the product. The good thing about the product is that it has variations, one of which (the preferred edition) allows you to read back the text in speech so in case there are any errors, you can make the necessary changes,” he adds. “What sets us apart from the rest is that the software actually identifies Indian English. You don’t have to speak with a unaccented voice or an American accent at all. Plus you can customise the software for typical words,’ concludes Manish. http://www.mid-day.com/web/guest/lifeatwork/itadda/article?_EXT_5_articleId=1150767&_EXT_5_groupId=14 www.icons.net.in