3iLogic-Designs, a provider of speech recognition technology, has unveiled SimSim, an embedded speech recognition IP core that enables voice-activated user interfaces in a wide range of consumer devices.
Semiconductor manufacturers can easily integrate SimSim, the flagship product from 3iLogic-Designs, in their application processors, microcontrollers, ASICs and FPGAs.
Speech recognition is widely recognized as a successful technology mainly because of the ease with which it enables communication between a human being and machine. Many devices including cell phones, digital cameras, wearable devices, smart TVs, automobile and home appliances are fitted with the hands-free, voice-activated user interface powered by these technologies.
However, most speech recognition solutions available in the market are software based and available only on high-end devices with adequate resources. They also demand a large memory footprint and significant host-processing power, which lead to higher power, increased latency and rise in cost.
SimSim IP core addresses these constraints by delivering high performance and ultra-low power, in an extremely small footprint without sacrificing features or accuracy, the company said. Localized processing removes dependency on cloud connectivity and provides very fast and deterministic response -- always, everywhere.
"Voice-activated user interface for everyday consumer devices is clearly causing a disruption, just as touch did to the keyboard,” said Dhaval Ajmera, CEO of 3iLogic-Designs, in a statement. “SimSim is breaking new ground in the field of embedded speech recognition by enabling voice-activated user interfaces on consumer devices that current solutions simply do not allow without adding significant resources to the device.”
3iLogic-Designs features a high-performance, ultra low power IP core for localized vocal user interface. The synthesizable solution easily integrates into microcontrollers, application processors, ASICs and FPGAs.
With its small silicon footprint of 135K gates, the solution fits into any small device. The Always On capability includes Voice Activity Detection (VAD) and Keyword spotting. The highly accurate, speaker independent recognition engine requires no user training.
SimSim is available, as a technology license, to semiconductor companies as well as system companies designing their own semiconductor. In addition, SimSim ships with a full set of deliverables, support and training, the company said.
Edited by Rory J. Thompson