Arm has introduced two new computer chip designs (ArmCortex-M55 processor and Arm Ethos-U55 NPU) for artificial intelligence (AI) applications. The new IP and supporting unified toolchain will allow AI hardware and software developers to innovate as a result of unprecedented levels of on-device ML processing for billions of small, power-constrained IoT and embedded devices.
Arm Cortex-M55 is the latest model in ARM’s Cortex-M line of processors that offers up to a 15x improvement in machine learning performance and a 5x improvement in digital signal processing performance as compared to previous Cortex-M generations. The other chip named Arm Ethos-U55 NPU (Neural Processing Unit) offers 32x improvement in the machine learning process. It is designed to run in conjunction with the M55 to accelerate the applications that require neural networks. By combining the two newly introduced chips, Arm aims at achieving a 480-times increase in the speed of running machine learning tasks compared to its previous Cortex M chips. The additional benefit of combining two chips is that it will bring 25x improvement in energy efficiency compared to the existing ones.
Cortex-M55 and Ethos-U55 are fully supported by Arm’s industry-leading Cortex-M software toolchain. This improvement in data throughput is sure to bring about a big difference in the performance of edge AI platforms. The current Cortex-M platforms can handle basic tasks like a keyword or vibration detection while the new edge AI chip, Cortex-M55 brings add-ons like object recognition. Besides, the full power of a Cortex-M chip combined with the Ethos-U55 promises enhanced functionality, with the potential for local gesture and speech recognition.
The latest technology is receiving widespread industry support from ecosystem partners, including Amazon, Alif Semiconductor, Bestechnic, Cypress, Dolby, Google, NXP, Samsung, STMicroelectronics and more.