Google Rolls Out Voice Search for Google Voice
Google bridges the gap of miscommunication as the company incorporates a new innovation that will personalize speech recognition on Google Voice.
Google first launched its patented cloud-based voice recognition system in 2007. The system was made to power the GOOG-411 information service. It means that more than screening hundred of phone calls, it will be now used to power automated voicemail for Google Voice. And now, Google makes a further development for Android by adding “personalized recognition” to its algorithm.
Google’s current system is designed to work “regardless of gender, age, and accents, or variations in pitch, pace, and other factors.” But those who have been using this service knows well that training it to recognize a person’s voice and speech pattern makes the accuracy more significant. But looking at the big picture, Google Voice Recognition isn’t 100% accurate after all.
Google hopes to change that aspect by adding a personalized recognition to its Android Voice Search app. By putting it in their system, Google will store recordings of voice search history and voice commands ina uers’s Google Account. These will be used to build a recognition model for the user’s voice and other speech idiosyncrasies.
“Although subtle,” wrote Googlers Amir Mané and Glen Shires, “accuracy improvements begin fairly quickly and will build over time.”
The personalized recognition feature is now available but for Android 2.2 and higher only. There’s still no news on improvements on other platforms. The Voice Recognition as of now is limited to American English, but support for other languages will be out soon. To add up this new platform, the new version of Voice Search includes speed and performance upgrades for name recognition.View Article Source »