Ambiq Supercharging Edge AI

Let's get started

Community

Using AI in Nursing Homes for a Better Retirement

Community

Pardon My Woof: Will AI Help Humans Talk to Animals?

Community

How AI-Powered Smart Glasses Are Redefining Wearable Experiences

Community

Hearing 2.0: How OTC Hearing Aids are Reshaping Hearing Loss

Practical Ultra-Low Power Edge AI

Let's get started

Practical Edge AI is here - Ambiq's Apollo4 Family of AI-capable devices enable high-performance AI capabilities while sipping power. Explore how to bring Speech UI, intelligent activity recognition, and real-time health analytics to your edge products.

Introducing AI at Ambiq

Learn the basics of how to use neuralSPOT to add AI features to your Apollo4 Plus applications, including how to integrate Tensorflow Lite for Microcontrollers, how feed sensor data to your AI models, and how to optimize for maximum performance at the lowest power possible..

Welcome to neuralSPOT

neuralSPOT is an AI-focused ADK (AI Development Kit) and toolkit which makes using Apollo4's hardware capabilities easier than ever. neuralSPOT takes care of embedded platform details, letting your data scientists focus on AI.

Ambiq ModelZoo

Ambiq's AI Model Zoo is a library of reference AI models optimized for Apollo4 Plus, including Speech UI, sound classification and processing, activity recognition, and more.

Full Performance, Ludicrously Low Power

Apollo4 is incredibly power-efficient, and our benchmarks show it. Explore our AI benchmark results and learn how replicate them using our open-source benchmark repository.

ModelZoo

Models

View all 4 models

Heart Arrhythmia Classification

Framework

The Heart Arrhythmia Classification example model performs real-time heart arrhythmia classification using 1-lead ECG and optionally PPG. Classification can be performed on either rhythm (e.g. normal, AFIB, AFL) or beat (e.g. PAC, PVC).

Health

Time Series

Neural Network Speech

Framework

NN Speech (NNSP) integrates 3 neural networks, voice activitiy detection (VAD), keyword spotting (KWS) and Speech to intent (S2I).

Audio

Voice

Speech Enhancement and De-noising

Framework

End-to-end development and deployment reference implementation of an efficient speech enhancement and de-noising model.

Audio

Voice

View all 4 models

Welcome to Ambiq AI

Visit on ambiq

Ambiq's VP of Artificial Intelligence, Carlos Morales, introduces neuralSPOT and Ambiq's ModelZoo

EXHIBITS

TRAININGS

Accelerating the Development of Optimized AI Features with Ambiq’s neuralSPOT

Ambiq’s neuralSPOT® is an open-source AI developer-focused SDK designed for our latest Apollo4 Plus system-on-chip (SoC) family. neuralSPOT provides an on-ramp to the rapid development of AI features for our customers’ AI applications and products. Included with neuralSPOT are Ambiq-optimized libraries, tools, and examples to help jumpstart AI-focused applications.

Using neuralSPOT to Ease Embedded AI Development

Ambiq's VP of Artificial Intelligence, Carlos Morales shows how to use neuralSPOT to connect the evaluation board (EVB) to your computer, and use it as an integrated component using remote procedure calls.

Downloading, Compiling, and Deploying AI on Apollo4 Using neuralSPOT

Carlos Morales shows how to use neuralSPOT to get a sample AI model running on Apollo4. This process will require an Apollo4 Plus evaluation board (EVB), and a microphone.

Ambiq AI GitHub Repository Introduction

Introduction to Ambiq AI repository (repo) in GitHub including source codes, models, SDK, and more. The main repos include neuralSPOT and ModelZoo. Learn more.