- Specialized NPU IP makes it easier for device makers to implement feature-rich, always-sensing camera systems.
- A dedicated processing solution addresses the power and privacy concerns of always-sensing camera deployments.
Santa Clara, California, July 18, 2023—Expedera Inc, a leading provider of scalable Neural Processing Unit (NPU) semiconductor intellectual property (IP), today announced the availability of its Origin™ E1 LittleNPU, a family of highly specialized artificial intelligence (AI) inference edge processors for use in always-sensing camera applications on smartphones, security cameras, home appliances, and other power-constrained devices.
“Device makers are constantly looking to add features that differentiate their products from the competition,” said Paul Karazuba, vice president of marketing at Expedera. “Always-sensing cameras are an example of this. They allow users to interact with their devices more naturally, enhancing the user experience and increasing owner satisfaction. The LittleNPU makes adding this easier for device makers because it provides the most power-efficient, security-aware method of adding AI processing to an always-sensing camera subsystem.”
While always-on audio applications have become commonplace in smartphones and smart speakers, always-sensing video has been more challenging due to higher power consumption and privacy concerns. The Origin E1 LittleNPU product family, optimized for the visual neural networks found in always-sensing applications, requires minimal power. For example, the LittleNPU consumes as little as 10 mW in some implementations. Additionally, always-sensing data is handled within the boundaries of the LittleNPU, minimizing the attack surface and allowing system hardware and software security resources to safeguard user data better.
Expedera’s LittleNPU is available as soft IP and is portable to any semiconductor process. For more information, visit https://www.expedera.com/LittleNPU.
Expedera provides scalable neural engine semiconductor IP that enables major performance, power, and latency improvements while reducing cost and complexity in edge AI inference applications. Third-party silicon validated and customer-proven in over 10 million devices, Expedera’s solutions produce superior performance and are scalable for wide-ranging applications from edge nodes and smartphones to automotive and data centers. Expedera’s Origin Neural Processing Unit products are easily integrated, readily scalable, and customized to unique customer use cases and requirements. The company is headquartered in Santa Clara, California, with engineering development centers and customer support offices in the United Kingdom, China, Japan, and Taiwan. Visit https://www.expedera.com
# # #
Paul Karazuba, Vice President of Marketing