Voice processing tiny AI chip delivering voice extraction

POLYN is the first to implement the combination of voice detection (VD), and voice extraction (VE) within the low-power analogue neuromorphic silicon. Integrating NeuroVoice allows customers to create miniature products that improve their end users’ hearing experience.

NeuroVoice operates at 10 µW power and performs inference in 20µsec. POLYN’s solution offers especially clear voice and immediate adaptation to any background conditions along with voice extraction/transparent mode.

“Earbuds, smartphones, hearing assistance, gamer headphones, and intercom systems need new technology to bring voice processing to a new level,” said Eugene Zetserov, VP Marketing and Business Development of POLYN. “Current methods of voice signal processing are power hungry and, in some cases, fall short. Immediate voice recognition is important for hearing assistance devices. AI-based extraction of the voice signal in a noisy environment, including irregular noises, provides a better hearing experience than standard noise cancellation filters. A neural network is the perfect tool for voice processing and POLYN offers it on a tiny neuromorphic analogue chip.”

POLYN develops the framework and tools for a neural network conversion into an analogue neuromorphic chip, which speeds up chip generation. It makes it simple to customise the existing chip to provide additional voice features such as wake word detection (WWD) and keyword spotting (KWS).

“A combination of NASP technology for voice extraction for clear communication and other voice processing features gives NeuroVoice additional advantage,” Zetserov said.