Ambient Scientific Announces Beta Release of Software Compilers for its Programmable AI Processor GPX-10 – EEJournal

Ambient Scientific is excited to announce the release of Beta versions of its compilers for GPX-10 AI processor using general purpose programmable MX8 cores, designed for enabling OnDevice AI in endpoint and edge devices.

Ambient is releasing two types of compilers(i) Vertical Compilers that enable fast development of AI applications and (ii) A Generic compiler that provides programming flexibility to application developers for AI models of their choice.

Using Vertical compilers, developers can create applications for range of applications for voice, computer vision and sensor fusion using pre-determined neural networks. Vertical Compiler for Voice enables voice-based AI applications such as creation of custom wake words, Menu of command phrases and Voice-IDs. End users will be able to create or change their wake-words, phrases and voice-IDs as many times as they choose over the life of the product. This is particularly important in situations where the device use-conditions are changed, or the user decides to repurpose an already ship device. Vertical Compiler for Vision enables creation of various computer vision applications with up to twenty-one classes of vision-objects as defined by OEM. The Vertical Compiler for Sensor fusion enables developers to implement multi-parameter-based sensor fusion AI inferencing – a first of its kind in endpoint devices.

Developers will be able to download these compliers along with the development environments for training, validation, and deployment to the GPX-10 AI Processor. Sample voice, vision and sensor-fusion applications are also provided for quick proof of concept tests.

Ambient Scientific’s Generic Compiler enables developers to implement computer vision applications for recognition of up to 100 object classes. It supports most of the pretrained and prebuilt computer vision models (based on Mobilenet v1/v2, Resnet, VGG etc.) that are commonly used in industry for commercial or benchmarking purposes. Developers can also use Ambient Scientific Generic Compiler to develop and implement CNN based custom applications for computer vision such as object recognition and classification, presence and occupancy detection. These applications can be built with transfer learning from pretrained models or developed from scratch. The Generic Compiler is framework agnostic. It allows developers to create neural network models using their own choice of AI frameworks such as TensorFlow, PyTorch, caffe, paddle-paddle, CNTK, C++ etc. and deploy their models on the GPX-10 AI Processor.

Ambient Scientific’s Vertical and Generic compilers are built to scale user applications across all ten programmable AI cores of GPX-10 to maximize the performance while optimizing power consumption. The compilers are available to developers under NDA.

About Ambient Scientific

Ambient Scientific is a leading developer of industry’s lowest power programmable AI processors designed to address the explosive demand for Inference and Training in endpoint, edge and Battery-operated AI devices giving both connected and unconnected devices and appliances their own personalities. Ambient’s AI processor and microcontroller products and software libraries are designed to bring a plethora of exciting new innovative products to the market quickly.

By enabling AI at the edge, Ambient’s technology plays a key role in sustainable computing by alleviating data traffic across networks and reducing the demand on the cloud datacenters which today power most of the AI ​​inference and training resources.

Learn more at https://ambientscientific.ai/

Leave a Comment