Custom Voice Assistants That Offer Total User Privacy Are Here

Custom Voice Assistants That Offer Total User Privacy Are Here

Sensory powers new voice-enabled Farberware microwaves, the first of many home devices to feature a private, custom voice assistant


SANTA CLARA, Calif., May 6, 2021 – Sensory today announced that custom voice assistants that deliver total user privacy are officially here and are already revolutionizing how people interact with things they own.


Sensory’s technologies are helping custom voice assistants go mainstream, powering everyday appliances like the Farberware FM11VABK voice-controlled microwave oven. Immediately available in the US market, this new microwave oven utilizes Sensory’s TrulyHandsfree and TrulyNatural technologies to provide all the benefits of a custom voice assistant without the privacy compromises of cloud-based, general-purpose assistant platforms.


A recent survey by called out custom voice assistants as the biggest voice AI trend for 2021. Conceptually, custom voice assistants are voice assistants trained for specific domains, making them capable of outperforming general-purpose assistants like Google, Alexa, and Siri in quality of experience.


Amazon has already jumped on this trend with its announcement of custom voice assistants. Sensory has led the market in this same direction, but with a few substantial differences – Sensory does it 100% on device, keeping voice data completely private, with no data collection from the product and no recordings sent to the cloud. This means no added costs for cloud connections.


Due to the edge architecture of Sensory’s domain-specific assistant technology, customers can enjoy hassle-free setup – no Wi-Fi, no apps, no third-party hardware, and no cloud-connected assistant always listening in. Just plug the microwave into a power outlet and it is ready to understand numerous voice commands, such as, “Open door,” “cook popcorn,” “set timer to one minute 36 seconds,” “defrost,” “reheat for 2 minutes,” and many more commands. Ease of setup and use is just one of the many benefits of Sensory’s AI on the edge. Since Sensory’s speech recognition is all done on-device, concerns about audio recordings being sent and stored in the cloud are put to rest — privacy comes standard!


“We’re at a turning point with voice UI technology. People love the convenience of mainstream voice assistants, but privacy, accuracy, complicated setup, and connectivity issues continue to be a growing concern among users. These concerns have intensified the need for custom private voice assistants,” said Todd Mozer, CEO at Sensory. “Sensory’s customer base used to be driven by companies like Amazon, Google, Microsoft, and Samsung that licensed Sensory technology to improve their general assistant offerings. Today we’re seeing a major shift, with most of our customers being brands that want to own and manage their own private custom voice assistant platforms.”


The Farberware microwave is an example of what can be done at the edge. The large-vocabulary recognizer with a custom statistical language model and NLU can support complex actions like, “Cook four baked potatoes.” Because the Sensory NLU engine looks for “intents” within a limited domain, the chance of being misunderstood is substantially less than any off-the-shelf assistants. Sensory’s product testing division, evaluated the device for speech recognition accuracy and task completion rate, and when benchmarked against the Amazon Basics Microwave with Alexa, the Farberware microwave accurately recognized and executed 68% more spoken commands than Alexa. The complete evaluation report is available from Sensory.


“Smart appliance adoption is growing among US broadband households. Thirteen percent of US broadband households now own a smart appliance and smart microwaves are a leading category. Voice control is a highly-valued feature for smart appliance owners, with more than one-fifth of those who own or intend to purchase various smart appliances rating voice control as one of the most appealing features of the appliance,” said Patrice Samuels, senior analyst, Parks Associates.


A demo video of Sensory’s technology powering a microwave can be found here:


This news expands on Sensory’s recent announcement of its VoiceHub platform, a free-to-use set of web tools that give brands the power to quickly create custom wake words and powerful, private voice assistants supporting many languages in just a few minutes.


About Sensory


Sensory Inc. creates a safer and superior UX through vision and voice technologies. Sensory technologies are widely deployed in consumer electronics applications including mobile phones, automotive, wearables, toys, IoT, and various home electronics. Sensory’s product line includes TrulyHandsfree voice control, TrulySecure biometric authentication, and TrulyNatural large vocabulary natural language embedded speech recognition. Sensory’s technologies have shipped in several billions of units from leading brands all over the world.

Finally, a Biometric Solution That Recognizes Users Wearing Face Masks and Doesn’t Require Touch

Finally, a Biometric Solution That Recognizes Users Wearing Face Masks and Doesn’t Require Touch

Sensory’s TrulySecure face and voice biometrics platform provides greater convenience and benefit for the new normal; capable of recognizing users wearing face masks, and can now recognize coughs and sneezes

Sensory, a Silicon Valley company pioneering AI at the edge, today announced that its TrulySecure face and voice biometric fusion platform has been adapted to help device and app developers create products for the new normal. With the ability to recognize users while wearing masks and accurately detect coughs and sneezes, TrulySecure provides greater convenience and safety for today’s environment, without compromising on user security.

As state and local governments around the world have started requiring people to wear face masks, many facial recognition systems, like those found on smartphones, are not able to function. Further, most facial recognition solutions cannot enroll the user’s face when partially covered by a mask. Data sensitive tasks that are reliant on facial recognition, like unlocking a phone, logging in to an app and processing a mobile payment fail to work. This leaves consumers frustrated and not able to access key functionality on their smartphone.

Sensory’s TrulySecure utilizes biometric fusion and multiple enrollments to avoid the limitations of typical facial recognition solutions.

Biometric Fusion of Face and Voice

By combining both face and voice into one SDK, developers can leverage both biometrics to address real-world challenges such as face masks or noisy environments. When faces are partially obstructed, the voice biometric provides the extra confidence needed to enable secure access. Alternatively, when used in a noisy environment, such as a restaurant or sporting event, the voice biometric is enhanced with a facial recognition score. Developers can leverage both face and voice biometrics in real-time to provide a seamless, contact free experience to the end-user.

Multiple Enrollments

TrulySecure natively supports alternate appearances by enabling multiple enrollments. For example, users can enroll for facial recognition as they would normally and also create a second enrollment with a face mask on. Many competing software solutions block enrollment when a face mask is worn, but Sensory’s technology can train and use face even with a mask on. For ideal performance and user experience, enrollments for both face and voice can be saved with mask on and mask off. This way the software can always apply the appropriate template to ensure accurate recognition in all conditions.

“In unprecedented times like these, device manufacturers and app developers are being challenged to find new ways to keep people safe and secure,” said Todd Mozer, CEO at Sensory. “TrulySecure does just that, allowing apps and devices to recognize and authenticate users in all conditions, even when wearing masks, thus eliminating the need for people to touch their face or remove their masks when in public. But we’re taking things a step further by updating the Sound ID component of TrulySecure to recognize certain noises that can be indicative of users being sick or around people who are sick.”

Added Features for Post Pandemic Life

Since early 2019, Sound ID has been a state-of-the-art feature within the TrulySecure suite of AI solutions. Sensory has now applied their expertise with deep learning model training to include cough and sneeze detection. At the public level, a technology capable of listening for, and recognizing the occurrence of coughs and sneezes can help business owners track the health of their employees, or cities generate health data statistics for the general population. At the personal level, smartphones and wearables can be used to track the health of users, and even alert them when their cough and sneeze count crosses a certain threshold.

For more information about Sensory and its technology, please visit

About Sensory

Sensory Inc. creates a safer and superior UX through vision and voice technologies. Sensory’s technologies are widely deployed in consumer electronics applications including mobile phones, automotive, wearables, toys, IoT and various home electronics. Sensory’s product line includes TrulyHandsfree voice control, TrulySecure biometric authentication, and TrulyNatural large vocabulary natural language embedded speech recognition. Sensory’s technologies have shipped in over a billion units of leading consumer products.