Syntiant and Sensory Demonstrate First Ultra-Low-Power Multi-Language Voice Solution At The Edge
January 9 2020, 01:10
At CES 2020, Syntiant announced a partnership with Sensory, combining the expertise and technologies in AI and speech recognition at the Edge, to deliver a deep learning, multi-lingual voice interface for battery-powered devices. The combined solution merges Sensory’s larger, higher-performance TrulyHandsfree wake word engine and voice control with Syntiant’s microwatt-power Neural Decision Processors (NDPs), bringing low-latency, real-time inference to edge devices.
At CES 2020, Syntiant announced a partnership with Sensory, combining the expertise and technologies in AI and speech recognition at the Edge, to deliver a deep learning, multi-lingual voice interface for battery-powered devices. The combined solution merges Sensory’s larger, higher-performance TrulyHandsfree wake word engine and voice control with Syntiant’s microwatt-power Neural Decision Processors (NDPs), bringing low-latency, real-time inference to edge devices.
The Sensory and Syntiant cooperation enables manufacturers to implement seamless “voice” commands in dozens of languages. Working together, the two technologies could also support additional features, such as voice-based user identification.
“A busy mom in Korea setting a house alarm or teenager in Barcelona raising the volume on his smart speaker, voice commands are becoming more ubiquitous driven by worldwide consumer demand,” states Kurt Busch, CEO of Syntiant. “Collaborating with Sensory allows us to combine their AI with our silicon technology, providing customers a large multi-language library of local commands for just about any application.”
Syntiant and Sensory are accelerating the delivery of a fast, efficient, cloud-free multi-language interface in devices, such as earbuds, smart speakers and smartphones, at a power level orders of magnitude lower than typical MCU offerings.
Custom built to run neural workloads, the Syntiant NDP100 and NDP101 NDPs can support dozens of local voice commands and consume less than 140 microwatts while performing local processing of audio events, increasing privacy, reliability and responsiveness. In addition to voice triggers, other device capabilities include audio event and environment classification, as well as sensor analytics.
Syntiant demoed the combined technologies at the Bosch Global booth (LVCC, Central Hall) during CES 2020 in Las Vegas.
“Sensory’s embedded AI voice software combined with Syntiant’s neural network technology opens up a new market for both companies, bringing ultra-low-power, multi-language, local command support to the edge,” said Todd Mozer, Sensory’s CEO. “Our joint solution provides added security because commands are performed at the edge, free from a cloud connection, while devices and applications come alive without pushing a button.”
Sensory is focused on improving user experiences through embedded machine learning technologies such as voice, vision and natural language processing. The company pioneered the use of neural network approaches for embedded speech recognition for consumer electronics with a well-engineered and patented codebase that has shipped in over 2 billion consumer products. www.syntiant.com | www.sensory.com