Sensory Inc., announced Thursday that its collaborating with the Alexa Voice Service (AVS) team to give developers the ability to use the wake up word “Alexa” with its TrulyHandsfree technology. As part of the AVS for Raspberry Pi project on GitHub, Sensory will enable device manufacturers to develop prototypes that feature Alexa voice control functionality and accelerate device time to market.
Sensory has developed a suite of voice models for Alexa that are part of Sensory’s TrulyHandsfree speech recognition engine. These models can range in size and performance features to handle low-power battery operated usage to highly noise robust far-field applications with extremely high accuracy. The AVS for Raspberry Pi project allows a GitHub download of three different Alexa triggers from Sensory, varying in size and performance, while fitting on a Raspberry Pi platform.
The Alexa Voice Service (AVS) is Amazon’s intelligent voice recognition and natural language understanding service that allows users to voice-enable any connected device that has a microphone and speaker. As an AVS developer, products with access to Alexa’s growing list of capabilities can be build through its regular API updates and feature launches, as well as over 3,000 custom skills built by the active developer community.
AVS provides the flexibility to choose the best experience for the Alexa-enabled product. Whether building a handheld device like a TV remote or smart watch or developing a hands-free experience for a connected speaker or home intercom system, AVS offers Automatic Speech Recognition (ASR) Profiles tailored for specific types of user interaction.
Third-party developers are starting to come to market with Alexa-enabled devices that utilize Sensory’s wake up word models. One such company is Nucleus, that recently released their Alexa-enabled intercom.
“When we were developing Nucleus, we tried other speech recognition technologies; however, after experimenting with Sensory’s voice models, we quickly realized why everyone says their technology is the best. The performance and accuracy benefits of Sensory’s wake up voice models for Alexa have been fantastic,” said Isaac Levy, founder and CTO of Nucleus Life.
Sensory has over 20 patents in the area of on-device voice wakeup, and its TrulyHandsfree voice activation has shipped in over 1 billion devices, making it the most widely deployed hands-free speech recognition technology in the world.
“Sensory’s technology makes it very easy for manufacturers to integrate a TrulyHandsfree voice-controlled user experience,” said Todd Mozer, CEO of Sensory. “The key to offering an effective voice wake up word technology is ensuring that the system is not only activated when the trigger phrase is spoken, but also ensuring the system is not accidentally activated when the wakeup word is not spoken. Sensory and the AVS team have worked hard to bring manufacturers innovative voice capabilities, so it is exciting for us to be working together.”
“Alexa was designed to make it easy for companies like Sensory to build on the service and offer a variety of choices to manufacturers when creating voice-experiences,” said Aaron Brown, director of Alexa. “We are excited to work with Sensory on Alexa voice trigger models that enable hands-free Alexa experiences for a wide variety of products.”