Supply Chain Management

Startup Offers Compact Speech Recognition Core

12 February 2014

3iLogic-Designs Pvt. Ltd. (Gurgaon, India) has developed SimSim, a configurable speech recognition circuit that does not require user training. The company, founded in 2011, claims that the SimSim core, available as licensable intellectual property, can provide a voice-activated user interface for a range of consumer devices.

The core is synthesizable and provides language and speaker independent recognition along with configurable vocabulary and grammar support. And the core is compact occupying 135K gates allowing it to be integrated with microcontrollers, SoCs and FPGAs.

The idea is that consumer electronics developers can drop in a hands-free embedded speech user interface into their products, courtesy of 3iLogic. The range of equipment 3iLogic is targeting includes cell phones, tablet computers, digital cameras, wearable devices, smart TVs and remote controls, automobiles, and home appliances including microwaves, washing machines and so on.

Traditionally speech recognition is a computationally intensive technology, requiring either local processing power or leaning on cloud-based support and even then with marginal results

3iLogic-Designs claims SimSim delivers accurate recognition from a low energy and silicon footprint. In a 28nm process, voice activity detection requires 100kHz clock frequency for processing and consumes 5-microwatts of power. Speech recognition for a 100-word vocabulary requires 50MHz clock frequency for processing and consumes 2.5mW. SimSim can be configured to support 16kHz or 8kHz audio streams and is host processor agnostic, working with embedded or external system memory. Development tools and APIs provided.

"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," said Dhaval Ajmera, CEO of 3iLogic-Designs.

The company has uploaded numerous Youtube videos that discuss its technology and potential applications.

Related links and articles:

News articles:

Giving the World a Listen

Voice Recognition Installed in More than Half of New Cars by 2019

Powered by CR4, the Engineering Community

Discussion – 0 comments

By posting a comment you confirm that you have read and accept our Posting Rules and Terms of Use.
Engineering Newsletter Signup
Get the GlobalSpec
Stay up to date on:
Features the top stories, latest news, charts, insights and more on the end-to-end electronics value chain.
Weekly Newsletter
Get news, research, and analysis
on the Electronics industry in your
inbox every week - for FREE
Sign up for our FREE eNewsletter