TMCnews Featured Article
August 01, 2011
Google's Voice and Speech Recognition Software is Paving the Road to Unlimited Mobility
By Susan J. Campbell, TMCnet Contributing Editor
Speech recognition software has made significant progress since its inception nearly 50 years ago. Even in its primordial phase, voice technology had vision and expectations to be mobile and become a way of life. But it hasn’t escaped the difficulties.
Google has led the way for innovative voice and speech recognition technology. For the past three years, there has been a development of products that turn voice into text and action, according to this Mashable report. Google (News - Alert) Search by Voice for the Android hit the markets running in 2009. Now, the demand for more has called for this software solution to be implemented on desktops.
In order to arrive at the next level, Google would need to bump up the 192 MB of RAM (News - Alert) the Android provided. While this technology was a break through, consumers expected more. Vincent Vanhoucke, a Google research scientist, said Google Voice Actions transformed everything for speech recognition. This new concept gives users the ability to send texts, make calls, send e-mails, get directions and play songs without using a keyboard.
Since then, Vanhoucke said voice and speech recognition has advanced to voice-activated commands. There has been a six fold increase for user traffic since Google’s Android (News - Alert) has simplified the microphone button and provided endless options for voice commands.
Making the move to a desktop has been another matter entirely. The voice and speech recognition technology does not work as easily on that system. Google’s team, along with Vanhoucke, has experienced extreme difficulties in creating the desktop voice search and speech recognition.
Like Google Search by Voice for Android, the Google Voice Search for the desktop listens to a user’s word through a microphone and calculates them internally and circulates the words back to the user. This entire process of voice and speech recognition takes seconds.
Developing this cutting edge speech recognition technology was no easy feat. Vanhoucke explained that cell phones are designed for voice input where desktop systems are less sensitive to pick up a users voice. Desktop microphones didn’t have the ability to decipher between a user and a television set or any other noise in a room.
New algorithms were written to improve voice clarity and decrease ambient noises in the speech recognition software. Developers had to include the approximately 230 billion words Google Search by Voice comes across, as well as the thousands of languages used by potential customers.
Conquering the quest to implement voice technology into daily tasks is a significant advancement in technology. Vanhoucke said that speech is another part of the accessibility and ubiquity story of being able to input information on a device. Speech recognition technology will not likely replace keyboards but accessing information on- the-go will be in high demand.
As if it wasn’t enough to have voice and speech recognition in our pockets Vanhoucke and the Google team are currently working to provide the same technology in the comfort of your home – literally. Google has a goal for [email protected] which would turn light switches, alarm clocks and other appliance on without the use of your hands—the ultimate smart home.
Susan J. Campbell is a contributing editor for TMCnet and has also written for eastbiz.com. To read more of Susan’s articles, please visit her columnist page.
Edited by Jamie Epstein