Nuance is working on ‘always-on’voice recognition for mobile phones

0
371

Nuance, a developer of voice recognition technology, is in talks with a number of chip manufacturers about how to develop mobile chipsets that continuous speech recognition may make, even if they are stand-by.

Vlad Sejnoha, CTO at Nuance, says the company is in conversation with a number of manufacturers of chipsets for mobile phones and tablets, to continuous speech recognition possible, so read at Technology Review. There is particular attention to the limit of the power consumption of the voice recognition in the sleep mode. Also have the new chipsets are more capable of speech and background noise to distinguish in a continuous stream. Within two years there would be smartphones based on the new chipsets in the store.

With the new mobile chipsets must, according to Sejnoha be possible to use a voice command to a mobile phone from its sleep mode, after which the recognized command or the task to be performed. In current speech recognition software such as Siri for iOS and S Voice of Samsung, the user must use the speech recognition to first activate it. Probably is chosen to limit energy consumption.

They point Sejnoha on the risk of apps that take advantage of an ‘always listening’ smartphone. Although the privacy according to the Nuance’s executive chairman, in principle, to ensure, users will initially have some be reluctant to the ‘always on’-luistermogelijkheid to activate due to concerns about the storage of data and the susceptibility to hacking attempts.