Ambient Scientific has introduced its innovative AI module, the Sparsh board, which operates on a coin cell battery, enabling a wide range of on-device AI applications. The module is designed to power tasks such as human activity recognition, voice control, and acoustic event detection, all while offering the remarkable capability to run for months without the need for frequent battery replacements.

The Sparsh board is equipped with a variety of components, including motion sensors, a digital microphone, a Bluetooth Low-Energy (BLE) module, and more. This makes it an ideal tool for developers looking to quickly prototype battery-operated AI applications.

According to GP Singh, the Founder and CEO of Ambient Scientific, traditional microcontroller units (MCUs) typically require a trade-off between AI performance and power efficiency. However, with the introduction of their ultra-low-power GPX10 AI processor, Ambient Scientific is pioneering a new direction. This processor utilizes advanced analog in-memory computing technology, allowing developers to create precise AI applications without compromising on battery life or performance. As a result, the Sparsh board offers greater versatility for various applications.

The company also provides a comprehensive software stack that simplifies the development process for AI applications. This stack supports popular frameworks like TensorFlow and Keras, as well as a proprietary compiler capable of accommodating most neural networks. Developers can also access sample AI applications and algorithms, enabling them to start building projects right after downloading the AmbiSense SDK.

The Sparsh board’s applications extend across numerous sectors, such as predictive maintenance, AI-enabled medical devices, wearables, and voice-controlled toys. As demand increases from product developers, students, and researchers, Ambient Scientific plans to release a variety of reference designs for battery-driven AI applications, helping make AI technology more accessible and affordable.

A key feature of the Sparsh board is its GPX10 processor, designed for power-sensitive edge devices requiring continuous AI processing. The GPX10 processor includes ten DigAn AI cores, an integrated system-on-chip (SoC) with an ultra-low-power analog-to-digital converter (ADC), custom SRAM, and support for various peripherals. This design is optimized to execute neural network functions while balancing power consumption and programmability, making it highly adaptable for edge AI applications.

To assist developers, Ambient Scientific offers its Development Kit (DVK), which provides a comprehensive platform for testing and fine-tuning embedded AI applications. The accompanying software development kit (SDK) includes tools for creating AI applications using popular frameworks, as well as a compiler and training toolkit to streamline the development process.

Ambient Scientific plans to showcase the Sparsh AI module at CES 2025, offering live demonstrations of AI applications powered by coin cell batteries. These applications will include use cases such as fall detection and voice recognition. The launch of the Sparsh board marks a significant advancement in bringing efficient, battery-operated AI solutions to the market, reinforcing Ambient Scientific’s commitment to advancing AI technology. Both hobbyists and professionals can now look forward to new opportunities to develop powerful, practical AI solutions.

By Impact Lab