Why Robotic OEMs are Searching for Better Voice Technology
Over the last few years, ArkX Labs has seen first-hand the explosive growth of voice technology across many verticals, including healthcare, education, home and industrial security, and entertainment. But one of the most active verticals in search of better voice experiences is robotics and autonomous systems.
That should come as no surprise to anyone familiar with the role that robots and co-bots will increasingly play in everyday life and the need for high-performance voice integration.
Using voice technology for human-to-machine speech recognition enhances the ability of robots to interact with humans in a more natural and intuitive way to assist with household tasks, provide support for elderly care or create interactive and engaging entertainment experiences. Voice will enhance learning experiences in educational settings and improve healthcare outcomes by assisting with surgeries, delivering medications, providing physical therapy, and helping with patient monitoring. And the proper voice technology will increase safety and efficiency in manufacturing, logistics, and agricultural spaces.
The possibilities are endless.
But there is a common challenge causing headwinds in adopting voice technology by Robotics OEMs: The current voice solutions in the marketplace lack clarity and accuracy in noisy environments and at a distance. The reality is that most voice recognition technology is highly sensitive to competing background noise, making it difficult for robots to understand and interpret human speech accurately. And giving voice commands from 2-4 meters away or in direct line of sight is not always practical. That is a problem. Until now.
EveryWord™ Ultra Far-Field Voice Capture is Disrupting the Status Quo
ArkX’s far-field voice capture solutions are based on 3D reverberation technology and can uniquely identify and suppress other single-point noise sources so voice commands can be captured from three times the standard distance (up to 9 meters), around corners, in noisy and reflective environments. That’s a big deal.
No more competing with a TV, engine noise, or other talkers or noise sources in the space. No more having to scream at a device from a few meters away or repeat yourself multiple times.
EveryWord™ Audio Front End (AFE) technology does not require source-ducking for reliable interaction, provides linear, circular, square, triangular, or 3-D mic array geometries, and requires fewer microphones. The technology features ultra-low power battery operation for wake-on-word, and the flexibility for placement of microphones allows for in-wall, ceiling, or dashboard solutions. The 3D mic array (unlike others’ linear beam-forming approach) enables fewer blind spots and increased performance while incorporating fewer redundant microphone arrays for coverage.
And all ArkX solutions are Alexa-compatible and meet or exceed all requirements for the Amazon Voice Services (AVS) qualifications. In addition to Alexa, EveryWord is compatible with other platforms such as Google, Siri, Cortana, AliGenie, Baidu/Kitt.ai, or Tencent. We also have a partnership with Sensory to create branded wake words and command sets.
In almost every application, our ultra far-field technology outperforms existing OEM options giving robotics a real leap in their ability to deliver on the promise of a hand-free world