Neuromorphic Chip Enables Voice Detection & Extraction In A Single IC

The neuromorphic analog chip can make your headphones smarter, it can enable AI to extract and process the voice without any external components.

Image showing process of voice extraction (source: Polyn)

Neuromorphic analog ultra-low power chip is industry’s first solution to combine voice detection and voice extraction in a single compact package. The neuromorphic chip is developed by Polyn which is a fabless semiconductor company based out of Israel. The company develops technology and products that enable a wide range of edge AI applications such as wearables, Industry 4.0, Connected Health 4.0, Smart Home, and more.

The tiny AI chip is capable of extracting voice from any background noise, thus, enabling hearables such as wireless earphones, hearing assistance, gamer headphones, and intercom systems to provide better sound quality even in noisy environments. AI-based extraction of the voice signal in a noisy environment, including irregular noises, provides a better hearing experience than standard noise cancellation filters. POLYN’s solution offers an 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 analog chip.”

To simplify the customization of the existing chip and speed up the chip generation process, the company has develops the framework and tools for a neural network conversion into an analog neuromorphic chip. The framework also enables adding additional voice features in the existing chip. Some of the additional features include wake word detection (WWD) and keyword spotting (KWS). The advanced neuromorphic chip can also enable, security personnel and firefighters to communicate better with their peers and others in noisy environments.



Source link

This site uses cookies to offer you a better browsing experience. By browsing this website, you agree to our use of cookies.