QuickLogic Corporation, a developer of ultra-low power multi-core voice-enabled SoCs, embedded FPGA IP, and endpoint AI solutions, has agreed to partner with Nuance Communications, a provider of conversational AI and ambient intelligence solutions, to deliver low power wake word and voice command technology for power-sensitive applications, including hearable and wearable devices.
Nuance's technology is considered a proven, reliable voice recognition solution, enabling QuickLogic's customers to get to market quickly with products they know will provide high accuracy even in noisy environments.
The partnership between the companies will enable QuickLogic to leverage its EOS S3 Voice and Sensor Processing Platform to give customers an end-to-end and highly reliable hardware and software voice recognition solution. The Nuance technology provides the performance and ultra-low power consumption required for always-on wake word detection, and specifically supports the Alexa wake word protocol. It also various technical enhancements to improve voice recognition accuracy in difficult or noisy environments.
The integrated system supports a wide range of features including always-on, always-listening fixed triggers, user defined triggers and phrases, and commands that can be accurately detected in silent to extremely noisy environments.
"Nuance has been delivering speech recognition solutions for mobile devices for over two decades," said Scott Haylock, director of product marketing at QuickLogic Corporation, in a statement. "In response to customer demand, and the growing hearables market, we've augmented the EOS S3 OPEN Software Platform to include Nuance's technology. This addition helps QuickLogic address the largest possible product mix of new and existing voice-controlled end-products."
The ultra-low power Nuance voice recognition technology has been integrated with QuickLogic's advanced EOS Voice and Sensor Processing SoC, whose architecture enables the industry's computational-intensive sensor processing capability, at a fraction of the power consumption of competing technologies.