AI reference design for edge applications such as face identification and keyword spotting

Maxim MAXREFDES178 Artificial Intelligence Reference Design, available now from Mouser, comprises a connectivity board based on the MAX32666 and an AI board based on two MAX78000 chips. A short 33-position flex cable connects the connectivity board and AI board. The MAX78000 is an Arm Cortex M4F microcontroller with a Convolutional Neural Network accelerator unit.

The unique energy-efficient architecture of the MAX78000 allows battery-powered AI for at the edge applications such as face identification and keyword spotting. The design is a cube camera reference design based on the MAX78000 and MAX32666 microcontrollers to assist AI at the edge device designers to expedite their proof-of-concept to the market phase.

The design includes two MAX78000 Arm Cortex M4F microcontrollers with CNN accelerator, including BLE5 wireless connectivity, and is Li-ion battery powered. It includes a colour image sensor and a digital microphone. The design provides multiple audio codecs with stereo audio input and output, colour TFT capacitive touch LCD, Micro SD connector, and onboard QSPI FLASH and QSPI SRAM.