Sensory Announces Full Release of On-Device Large Vocabulary Speech Recognition Solution

“Perhaps most importantly, the system is designed to operate entirely on a given device, with no need for data to be sent to an external server for processing, helping to protect end users’ privacy.”

Sensory Inc. is looking to bring sophisticated speech recognition technology to a range of mobile and Internet of Things devices with its TrulyNatural SDK platform.

While TrulyNatural was first announced back in 2015, Sensory says this is now the platform’s “first full feature release” of the platform, reflecting years of R&D and beta testing. TrulyNatural can be implemented in a package as small as 10MB, and is highly accurate, allowing for intelligent voice-based User Interfaces to be implemented in all kinds of smart devices.

Perhaps most importantly, the system is designed to operate entirely on a given device, with no need for data to be sent to an external server for processing, helping to protect end users’ privacy.

As Sensory CEO Todd Mozer explained in a statement announcing TrulyNatural’s launch, companies taking advantage of other speech platforms like Amazon’s Alexa “have been forced to risk customer privacy by allowing always listening devices to share voice data with the recognition service providers,” whereas “TrulyNatural does not require any data to leave the device and eliminates the privacy risks associated with sending voice data to the cloud, and as an added benefit it allows product manufacturers to own the customer relationship and experience.”

On that note, the solution can also be combined with Sensory’s TrulyHandsfree solution, allowing companies to configure their own custom wake words for devices using TrulyNatural. Additionally, it can be combined with Sensory’s TrulySecure solution to enable speaker verification with biometric voice recognition or even facial recognition.

The TrulyNatural SDKs are available for Android, iOS, Linux, and Windows, and with the platform currently supporting voice interaction in English, support for French, German, Italian, Japanese, Korean, Mandarin Chinese, Portuguese, Russian, and Spanish are in the pipeline for later this year and 2020.