Zoom Room

Zoom Rooms Integrate Sensory’s On-Device Processing for Voice Commands

Zoom has partnered with Sensory to shift voice command processing for its Zoom Room platform from the cloud to the device. Sensory’s TrulyNatural speech recognition software will handle the voice commands locally, keeping the interactions private, without any concern about transmitting or storing the audio.

Local Control

The Zoom Room platform is designed for conference rooms and more high-end home offices. They offer more features than the standard personal Zoom meetings, including voice commands to awaken the room, join or start meetings, and adjust volume and other settings. Until now, those voice commands were sent into the cloud to be processed before the software responded. Now, all of that happens on the device running the Zoom Room, even if it isn’t connected to the internet yet. Sensory’s TrulyNatural technology handles the speech recognition and command process using its own neural network-based natural language understanding engine, augmented with Zoom-specific commands and controls, including enabling users to spell out the codes for the meetings and their personal IDs and passwords. Keeping the voice commands local offers the benefit of privacy by completely eliminating any chance of the audio being stored or intercepted in the cloud, an unlikely but still possible concern, especially for Zoom’s enterprise clients.

“Zoom understands the important role privacy plays to enterprise customers and selected Sensory to bring the convenience of voice control to Zoom Rooms,” Sensory CEO Todd Mozer said. “Utilizing our advanced embedded voice AI stack, we’ve bolstered the voice experience while maintaining total privacy and brand recognition.”

Zooming Along

Zoom has been rapidly building out its new features, especially after the COVID-19 pandemic and subsequent spike in usage. Most recently, Zoom bought AI-powered translation startup Kites to bring real-time translation to its services. Though Zoom can certainly afford many outright acquisitions after selling around $1.75 billion worth of shares earlier this year, the company has made an effort to bolster its partnerships as well. Australian voice AI developer Dubber launched its Unified Call Recording in April, capable of recording and analyzing conversations directly from Zoom. A month later, Otter released a virtual assistant that runs quietly in the background of meetings, records and transcribes the audio, and takes notes in real-time. In terms of Zoom Rooms and their voice commands, the platform integrated Alexa for Business controls back in April. The Sensory arrangement extends that strategy even more.

“Zoom is passionate about making collaboration easier, but we always put our customer’s privacy and security front and center,” Zoom lead product manager Cynthia Lee said in a statement. “Sensory’s technology checked all the boxes for us: accurate, fast and private. We look forward to adding more voice features as our public beta progresses towards full launch later this year.”

Sensory’s VoiceHub Expands to Offer Easier Way to Build Voice User Interfaces for Edge Devices and Apps

Zoom Integrates Alexa for Business into Conference Room Platform

Zoom is Coming to Amazon, Google, and Facebook Smart Displays, But Won’t Include Voice Controls