The low-power co-processor enables wearable AI and sensor-integrated applications across consumer, healthcare, IoT, defense, and wake-up technologies.
BrainChip Holdings Ltd has introduced Akida Pico, a low-power acceleration co-processor. Akida Pico enables compact devices for wearable AI and sensor-integrated applications in consumer, healthcare, IoT, defense, and wake-up technologies.
The co-processor accelerates specific neural network models through a digital architecture focused on energy efficiency. It supports secure personalization for use cases such as voice wake detection, keyword spotting, noise reduction, audio enhancement, presence detection, personal voice assistants, automatic doorbells, wearable AI, and appliance voice interfaces.
Built on the Akida2 event-based computing platform, Akida Pico operates on less than a milliwatt of power, making it suitable for battery-powered devices. It can activate microcontrollers or system processors only when necessary, using neural networks to filter false alarms and conserve energy. This makes it effective for sensor hubs and systems that require monitoring but need processing only when triggered by events.
Key features of Akida Pico include:
- Low-power standalone NPU core consuming less than 1mW
- Support for power islands to reduce standby power
- Industry-standard development environment
- Small logic die area with configurable buffer and model parameter memory
Akida, BrainChip’s event-based computing platform, is built for early detection and low-latency solutions across robotics, drones, automotive, and sense-detect-classify-track applications. BrainChip offers software, hardware, and IP products for current and future designs, with plans for multi-modal AI deployment at the edge.
“Like all of our Edge AI enablement platforms, Akida Pico was developed to further push the limits of AI on-chip compute with low latency and low power required of neural applications,” said Sean Hehir, CEO at BrainChip. “Whether you have limited AI expertise or are an expert at developing AI models and applications, Akida Pico and the Akida Development Platform provides users with the ability to create, train and test the most power and memory efficient temporal-event based neural networks quicker and more reliably.”
For more information, click here.
