Ambiq Micro has introduced compressionKIT™, a beta AI-based codec designed to significantly reduce the power and memory burden of handling continuous sensor data in wearable and edge devices.
In always-on devices such as medical wearables, smart home products, and industrial sensors, constant data collection can quickly drain memory, battery life, and system resources.
compressionKIT tackles that challenge at the source by compressing sensor data while preserving the key information needed for AI processing, helping devices do more with less.
The new platform strengthens Ambiq’s edge AI portfolio by addressing a critical bottleneck: how to efficiently represent sensor data before it is stored, transmitted, or analyzed.
Its benefits include up to 20x data compression, up to 16x lower on-device memory usage, reduced transmission power, and more flexible deployment options. Developers can run inference on-device, in the cloud, or through hybrid edge-cloud setups using either compressed or reconstructed data.
compressionKIT also gives developers configurable compression targets from 2x to 20x, along with a visual tuning interface to help balance data rate, signal quality, and system constraints.
It supports both hybrid DSP and machine learning approaches, as well as AI-first neural compression for maximum data reduction.
Leadership Comment
“For always-on devices, managing sensor data efficiently is just as important as running inference efficiently,” said Dr. Adam Page, Head of AI at Ambiq. “compressionKIT gives developers a practical way to reduce storage and transmission demands while preserving the signal information needed for meaningful AI insights.”
compressionKIT is currently in beta testing, with rolling improvements to be released in the coming quarters.
Learn more about compressionKIT here.
