Ambiq introduces neuralSPOT, an open-source AI developer-focused SDK designed for Ambiq’s latest Apollo4 Plus system-on-chip (SoC) family.
While leveraging Ambiq’s proprietary SPOT platform, neuralSPOT unlocks a game-changing, multi-fold improvement in energy efficiency for our end customers’ electronic products. The SDK encompasses essential software and tools for developers to advance AI features for their applications. Designed with libraries for communicating with sensors, managing SoC peripherals, controlling power and memory configurations, tools for easily debugging AI models from a laptop or PC, and examples, neuralSPOT ties it all together.
Today, AI is used extensively in the multibillion-dollar intelligent Internet of Things (IoT) space for everything from industrial anomaly detection to speech-based consumer interfaces. The IoT endpoint installed base is expected to grow from 14.6 billion devices in 2021 to 30.0 billion devices by 2027, according to Ericsson, reinforced by growth in applications such as smart wearables, hearables, industrial IoT (IIoT), intelligent manufacturing, healthcare, smart home, smart buildings, and automotive.
With a turnkey approach, neuralSPOT helps unlock the Apollo4 Plus’s highly efficient AI capabilities by including Ambiq-specific optimized libraries for audio, I2C, power and memory management, accelerated inter-process communication, and other peripherals. Moreover, it offers customized tools to ease common cross-platform AI development challenges, such as testing a deployed model against public datasets by allowing PC-based AI models to communicate seamlessly with Ambiq’s embedded platforms. It also includes ready-to-use AI examples to help bootstrap AI feature development on Ambiq products.
The open-source and modular approach in neuralSPOT is designed specifically for embedded AI application developers, flexibly accommodating multiple workflows to drive development and collaboration.
An open-source technical preview of neuralSPOT is available now.