Ceva and Edge Impulse Collaborate on AI Technology

Ceva, a leading provider of semiconductor intellectual property (IP), has announced a strategic collaboration with Edge Impulse to develop small AI frameworks for the Ceva NeuPro-Nano neural processor unit (NPU). This partnership aims to empower AI developers by enabling them to train, optimize, and deploy embedded machine learning models on the Ceva NeuPro-Nano NPU IP, even before silicon production, using the Edge Impulse Platform. The primary goal of this collaboration is to accelerate the deployment of edge AI implementations on low-power system-on-chip devices.

According to Ceva, this collaboration marks a significant milestone in the industry as it is the first of its kind to focus on an NPU IP. By providing AI developers with the tools to accurately benchmark and refine the performance of edge AI models, the collaboration is expected to greatly reduce the time-to-market for NeuPro-Nano based silicon targeting AIoT devices. The NeuPro-Nano NPU IP for Embedded AI is part of the highly efficient Edge NPU family, designed to offer a balance of ultra-low power consumption and high performance in a compact form factor, making it ideal for various embedded ML applications.

Chad Lucien, Vice President and General Manager of the Sensors and Audio Business Unit at Ceva, emphasized the increasing demand for on-device AI across industries and market segments. He stated, “To address the acute demand for low-power, small form factor, and powerful AIoT devices, our customers require a solution to efficiently and cost-effectively incorporate embedded ML applications into their NeuPro-Nano based products. Edge Impulse Platform, in conjunction with NeuPro-Nano, provides AI developers with a seamless, user-friendly experience to rapidly design, develop, and deploy silicon-optimized AI applications with market-leading performance.”

The Edge Impulse Platform offers 'no code' AI tools tailored for the NeuPro-Nano hardware, streamlining the deployment and benchmarking of embedded ML applications. The NeuPro-Nano architecture supports AI inference, feature extraction, and DSP workloads commonly used in audio, voice, and vision applications. By combining these capabilities with the Edge Impulse Platform, developers can benefit from accurate cycle count emulation for the full end-to-end AI application, enhancing the overall development process.

Zach Shelby, Co-founder and CEO of Edge Impulse, expressed enthusiasm about the collaboration, stating, “Ceva-NeuPro-Nano is an exciting new NPU IP core that will enable a wide range of embedded ML applications to efficiently run in small, low-power, and high-performance edge devices. Our collaboration with Ceva on this NPU IP empowers NeuPro-Nano customers to leverage the Edge Impulse Platform, unlocking the full potential of embedded ML in their AIoT products even before silicon production. This facilitates the rapid and scalable deployment of edge AI solutions across industries and markets with unprecedented performance.”