Sensory Speech Recognition Delivers 80% More Accuracy
Sensory’s TrulyHandsfree™ embedded always on and always-listening speech recognition platform has received a major upgrade boost in accuracy. By adding advanced deep learning and filter-bank features, Sensory increased accuracy by up to 80%, even in noisy environments.
TrulyHandsfree 4.0 offers such features as new phrase spotting techniques and neural nets engine that supports deep learning acoustic models, dramatically improving speech recognition accuracy in real-world situations.
Designed to take advantage of deep learning and other cutting edge techniques, Sensory uses a unique form of a neural network with deep learning to achieve acoustic models an order of magnitude smaller than the present state-of-the-art. The user’s spoken request can be recognized in the middle of speech, or when surrounded by ambient noise.
The technology is able to fully operate with less than a 2mA draw on the battery, and as low as 1mA with Sensory’s low power sound detection intellectual property. Enhanced architectural scalability allows for low-power DSP implementations with secondary accuracy improvements at the operating system level.
“TrulyHandsfree was an industry-first that literally changed the way people interact with devices of all kinds,” says Todd Mozer, CEO of Sensory Inc. “Many thought it was impossible to create an always-listening voice user interface, that had both accuracy and low power, but Sensory did it, and the industry has followed but trailed in performance. TrulyHandsfree 4.0 takes performance to a whole new level with an accuracy, footprint and power consumption that others just can’t touch.”
To contact the author of this article, email firstname.lastname@example.org
Via: Google Alert for Deep Learning