3iLogic-Designs was founded in 2011 to provide speech recognition technology for everyday consumer devices. The company has raised capital from the co-founders and private investors. 3iLogic-Designs has offices in California and a development center in Gurgaon, India, and will soon open an office in Japan.

Voice is the natural and primary mode of communication amongst humans. The ability to communicate with machines through a natural interface, like voice, is an order of magnitude faster and easier than using a keyboard, mouse or even touch and gestures. Devices such as cell phones, tablets, digital cameras, wearable devices, smart TVs and remote controls, automobiles, and home appliances, enrich the user experience significantly by offering a hands-free, voice-activated user interface through speech recognition.

Speech recognition is a computationally intensive technology. Solutions available on the market today are software-based and available only on high-end devices that have adequate resources. Software solutions require a large memory footprint and significant host processing power (most often an additional DSP or assistance through the cloud). These requirements lead to higher power consumption, increased latency and increased cost, making embedded speech recognition expensive and prohibitive for resource constrained low to mid-range devices such as microwaves, alarm clocks, and printer.

To combat this problem, 3iLogic-Designs has introduced it flagship solution, SimSim, the first, full-featured, ultra-low-power embedded speech recognition IP core, enabling voice-activated user interfaces for consumer devices. The patent pending technology enables the convenience and utility of hands-free, voice-activated, command and control in any device.

“Always On” capability includes Voice Activity Detection (VAD) and Keyword spotting (with or without speaker verification). The highly accurate, speaker independent recognition engine has a language independent architecture and requires no user training. It features programmable keyword, dictionary, and grammar, with a scalable vocabulary up to 64K words. It supports simple to complex grammar design with Java Speech grammar support

The SimSim IP core delivers high performance and ultra-low power in an extremely small footprint without sacrificing features or accuracy. Localized processing removes dependency on cloud connectivity and provides very fast and deterministic response. The core takes spoken audio input in the form of 16-bit data (16 Khz or 8 khz), through Silence filtering, Feature vector generation, Senone scoring, Viterbi decoding and generates a result database. This database is analyzed by the host to interpret the spoken utterance.

The synthesizable solution easily integrates into microcontrollers, application processors, ASICs and FPGAs and has a small silicon footprint (135K gates). It is host processor agnostic and works with embedded or external system memories. In a 28nm LP process, voice activity detection requires only 100kHz for processing and consumes 5µW of power and 0.047mm2. Speech recognition for a 100 command vocabulary requires only 50MHz for processing and consumes 2.5mW of power.

SimSim is available now as a technology license to semiconductor companies and system companies. The company is currently engaged with roughly 8 semiconductor companies, who have been performing due diligence since November 2013. The company ­continues to gain depth and traction it its customer relationships.

Dhaval Ajmera, CEO (previously Dir. of WW Marketing and Business Development at Sun, VP of Sales North America at Ceva and VP of Sales at Mosys)

Pankaj Pailwar, Co-founder (previously VP Hardware at Innomedia)

Amit Joshi, Co-founder (previously a senior engineering manager at SiRF and staff manager at Marvell)