Is Mobile Voice Technology Going to be the Next Technological Resurgence?

The idea that people will be able to talk to computer and the computer will understand the command, is surfacing since the days of 70s. But, it has never been able to get into mainstream. In the recent months, voice activation technology got enough momentum and it is going to be the next big thing worldwide.

Now, you can send message to your friend by talking instead of typing by using Mobile voice-recognition technology or go through transcriptions of voice mail instead of  listening  them all.  The technology also allows you to tell your phone what to look on the Web and down the road, you will be able to post at Twitter by talking while you drive.

“It’s now possible to pick up your phone and press a single button and say, ‘I want the Yelp.com review of the Capital Grille in Burlington, Massachusetts. Period,’ ” said Vlad Sejnoha, chief speech scientist at Nuance Communications, a major producer of voice-to-text software.

Now, there are quite a few mobile apps like ShoutOut to Dragon and Vlingo, that can translate your voice to text message and emails. The technology listens to a voice, translates it into digital data and figures out what kind of sounds or words will come next. This is different than previous voice recognition technology that tried to recognize every sound and used lot of computer resources.Bill Meisel, an independent consultant said that the accuracy from voice to text message is almost 80% to 90%.

However, there are still some obstacles. One of them is to determine noise recognition. When you talk in a crowded place or noisy area, the technology can not differentiate between noise and speech. So the transcription comes very poor in this circumstances. But, some of the smartphones address this sort of problem. Google Nexus comes with two microphones, one for voice and another one for noise recognition. Eventually, the software deducts noise part from the voice file to produce final voice that has to be transcribed. The other problem is if you are asked to pronounce same sentences twice, you may sound different and that can create trouble for voice recognition software.

Additionally, computers are getting smarter in terms of recognizing voices when the accent is very common and this made a big difference today over the voice recognition technology in 70s. Since, currently the biggest constraints is computer power, the user needs very sophisticated smartphone to use this technology.