Aizip Announces TinyML Models and Service

Saratoga, CA, March 30, 2021 – Aizip, Inc., a Silicon Valley startup company providing powerful tiny artificial intelligence (AI) models for the Internet of Things (IoT), the so-called AIoT, announced today the availability of its broad portfolio of tiny machine learning (TinyML) models.

 

Leveraging its automated design tools and its leading technical expertise, Aizip has developed models for three series of applications:

 

  • Aizip Intelligent Vision (AIV) offers comprehensive deep neural networks (DNN) for image/video applications, such as classification, face identification, and object detection.
  • Aizip Intelligent Audio (AIA) includes a wide range of AI models for audio-related applications, such as keyword spotting and speaker identification/verification.
  • Aizip Intelligent Time-Series (AIT) provides TinyML for numerous applications with time-series-based datasets, such as ECG, EEG, and inertial measurement.

 

Building small-size AI models for low-power MCUs while maintaining high accuracy and low latency has been a major challenge for the AI community. Aizip has made significant progress in this direction through its extreme quantization technology and its toolsets. Mixed precision has been widely used in the AIV, AIA, and AIT models, and binary neural networks (BNN) are used in some situations. The resulting improvement is significant across the board, and leads to 100x or more improvement in model size and in inference speed for multiple cases.

 

In parallel with its technical breakthroughs, Aizip has established tremendous traction with its TinyML services. The value in model performance, delivery speed, and ROI is well-recognized by Aizip customers. Aizip received its first domestic revenue within three months after the company set up its operation in Silicon Valley, and generated its first international revenue one month later. So far, Aizip has successfully delivered models to its customers in all three categories of the AIV, AIA, and AIT series, according to Yan Sun, Co-Founder and CEO of Aizip. Sun, who received his Ph.D. from Stanford University, is a serial entrepreneur and has also held leadership positions in public companies. Aizip has won customers in both the startup and public company space.

 

The general availability of the Aizip TinyML models should allow its customers to quickly adopt Aizip’s Design as a Service (DaaS) and offer differentiated products on the market. Putting intelligence into IoT devices helps to make the world a much better place. Aizip is currently expanding its customer base in the U.S. as well as overseas. For more information, please contact info@aizip.ai.

 

* * *

About Aizip, Inc.

Aizip, Inc., develops AI models for edge applications. Based in Silicon Valley, Aizip provides models with superior performance, quick turnaround time, and excellent ROI. These models can be used in a wide range of applications for an intelligent, automated, and connected world. For more information, visit www.aizip.ai.