New: Community Edition

Smarter Consumer Smart Devices

Product Differentiation From Truly Intelligent, Autonomous IoT Insight

Lifestyle Wearables

For successful implementations, wearable devices should be able to adapt and update with changing user needs, attributes, the desire for personalization, and evolving applications. SensiML addresses one of the key barriers in this process, the scalable development of application-specific algorithms based on a common set of physical sensors. By removing the data science and coding requirements, SensiML focuses the algorithm development effort on collecting well-labeled datasets across user types, activities, and an ever-expanding set of useful insights.

The CHALLENGE:

Wearables have been part of our life for many years and represent one of the fastest growing sectors of IoT applications. While wearable technology continues to improve, this segment has faced many challenges including limited battery life, a high degree of user-to-user variations resulting in poor models and the need for hand-coded algorithm development, and a less-than-ideal user experience.

Where SensiML Can Help

Application Summary

Companies are investing their efforts on sensors, IoT and machine-to-machine learning. Humans are making use of the applications and services that are enabled by these devices which has given rise to wearables getting smarter. SensiML can help with scalable algorithm development workflow, rich data collection and labeling features, and optimized machine learning models.

Enhanced AR/VR Experiences

There are many uses cases for AR and VR in consumer applications. The augmented reality platform allows new dimensions of gaming, travel, enhanced visitor experiences in museums and retail, and sports training. As a platform for immersive and digitally augmented experiences, the richness of that experience depends on real-time contextual insight. Sensor inputs beyond visual/image data play a key role in delivering AR/VR authenticity and usability including motion, microphone recognition tasks.

The CHALLENGE:

Smart glasses and immersive computing are major sectors for growth in wearable products. The biggest challenge for designers is achieving functionality and wearability at an affordable cost. Head-worn devices not only require real-time AI to identify contextual sensor and image data but must also feature reasonable battery life and meet demanding form factor expectations.

Where SensiML Can Help

Application Summary

SensiML Analytics Toolkit enables rapid and scalable coding of recognition algorithms for vital non-image related sensor data including motion and audio classification. The SensiML Analytics Toolkit can build sophisticated low-latency audio and motion classification algorithms capable of running on microcontrollers and application processors side-by-side with image processing and display workloads. With optimized algorithms that can leverage MCU, FPGA, and DSP processing cores, SensiML sensor insights can augment core image/display processing for improved UI and immersive AR/VR realism.

Smart Home and Consumer Goods Automation

Smart home automation devices include a growing array of large and small appliances, switches, HVAC, water heaters, window blinds, robotic vacuums, security devices, and more. With the advent of connected home platforms, middleware, and application integrations, the inclusion of connectivity is rapidly given way to a need for built-in sensing and intelligence to differentiate today's smart home goods.

The CHALLENGE:

The possibilities of “smart” home automation are many. However, having multiple devices streaming data to the cloud and back is power-intensive, impractical, and a concern for user privacy expectations. Instead, a more effective approach to intelligent home devices calls for each device to have local insight and processing, limiting or eliminating cloud dependency and reducing power consumption. That local intelligence must be achievable on microcontrollers appropriate to the cost and power budgets of home IoT devices.

Where SensiML Can Help

Application Summary

With years of expertise in IoT and AI for low-power embedded processor designs, SensiML has dramatically streamlined and simplified the entire algorithm development process for implementing AI on extreme edge devices for smart home automation. SensiML optimized code-generation combined with domain expertise of application developers knowledgeable in desired outcomes allows for a data-driven design flow where labeled datasets replace hand-coded algorithms. The net result is a quick, easy, and effective way to add intelligence to home automation endpoints while keeping power consumption extremely low.

Use Cases

Consumer

Are You A Maker Or Inspired AI Innovator?

Visit the SensiML Data Depot repository and review available application examples, documentation, and sample datasets. Each application includes summary information on the hardware and sensor used, number of sample captures, and sector.