Syntiant develops Neural Decision Processors™ that enable the deployment of deep learning models on power-constrained edge devices, significantly enhancing efficiency and throughput compared to traditional microcontrollers. Their technology addresses the limitations of cloud dependency by providing ultra-low-power, high-performance processing for applications in battery-powered products like hearing aids and smart speakers.
Funding
$311.4M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
+3Founders
Product
Problem
Many edge devices are limited by the power consumption and processing capabilities of traditional microcontrollers, making it difficult to deploy complex deep learning models directly on these devices. This often necessitates reliance on cloud servers, increasing latency and power usage while raising privacy concerns.
Solution
Syntiant provides Neural Decision Processors (NDPs) that enable efficient deployment of deep learning models on power-constrained edge devices. Their at-memory compute architecture significantly reduces power consumption and latency compared to traditional microcontrollers by minimizing unnecessary data movement. These processors are specifically designed to directly process neural network layers, achieving high levels of efficiency, and offer scalability for various edge workloads through multi-generational product offerings. Syntiant also offers hardware-agnostic, production-ready deep learning models optimized for compute-constrained environments, as well as high-performance MEMS microphones and vibration sensors.
Target Audience
The primary target audience includes manufacturers of battery-powered devices such as hearing aids, smart speakers, automotive systems, and smart home devices, as well as developers seeking to implement edge AI solutions.
Features
- Neural Decision Processors (NDPs) deliver 100x the efficiency and 10-30x higher throughput compared to low-power MCUs.
- At-memory compute reduces power consumption and latency.
- Deep learning models optimized for audio event, speech, sensor, and computer vision applications.
- Hardware-agnostic models deployable on a wide variety of hardware, from GPUs to MCUs.
- High-performance MEMS microphones and vibration sensors for voice interfaces and clear audio capture.
- Scalable product offerings to accommodate diverse edge computing needs.