Sparse Artificial Intelligence MCU from Femtosense and ABOV Semiconductor Delivers AI-Based Voice Processing On-Device

Femtosense, in partnership with ABOV Semiconductor in South Korea, has launched the AI-ADAM-100, an artificial intelligence microcontroller unit (AI MCU) built on sparse AI technology to enable on-device AI features such as voice-based control in home appliances and other products. According to the Silicon Valley startup, the new AI-ADAM-100 AI MCU is a low-power, low-cost option for on-device language processing and voice-cleanup capabilities.

Microchip's Silicon Storage Technology (SST) subsidiary announced that its SuperFlash memBrain neuromorphic memory solution has helped WITINMEM solve speech processing challenges at the edge. The company's new neural processing, computing-in-memory SoC is the first in volume production that enables sub-mA systems to reduce speech noise and recognize hundreds of command words, in real time and immediately after power-up.

CEVA and Fluent.ai, a leading provider of on-device, multilingual speech understanding solutions, announced a partnership to offer ultra-low power speech-to-intent solutions for intelligent edge devices. Fluent.ai’s suite of speech-to-intent technologies has been ported and optimized for CEVA’s low power audio and sensor hub DSPs, providing a high performance, robust solution for OEMs and ODMs looking to integrate intelligent voice activation and control into their wearables, consumer devices and IoT products.

Amazon announced that for the first time it is making the high-performance far-field microphone array and voice processing technology found in Amazon Echo available to hardware makers who want to build the Alexa experience into their products. With the Amazon Alexa 7-Mic Far-Field Development Kit, Amazon is making it easier for commercial developers to build and launch far-field voice-initiated products with Alexa by reducing development time and cost.