Arm on-device voice control icon

Complex deep learning models run quickly and efficiently for speech recognition.

A simple line drawing of a cloud with sparkles.

Memory footprint of a microwave language model reduced from 17 MB to 1.6 MB.

Arm speech regonition icon

Customizable, accurate speech recognition, while adhering to consumer devices constraints.

Ebook expanded cover of Edge AI - Enabling Smart IoT Applications

Discover the Real-World Impact of Edge AI

Explore in-depth use cases that show how edge AI is powering the next generation of IoT, solving real-world problems, driving faster decisions, and lowering costs through smarter operations right at the device level.

Read eBook
KEY TAKEAWAYS
  • Sensory delivers embedded voice assistants using Arm-based processors for offline, private control.
  • Arm Cortex-M55 and Ethos-U55 powered speech models with no need for cloud connectivity.
  • Arm quantization and Helium optimizations reduced model size from 17 MB to 1.6 MB.
  • Sensory used Arm tools to partition workloads and improve speed, power, and responsiveness.
  • Arm enabled Sensory to deliver fast, private voice control within microcontroller constraints.

Stability AI

Transforming On-Device Audio AI

Stability AI partnered with Arm and used Arm KleidiAI to transform on-device audio creation, reducing response times from minutes to seconds on Arm mobile CPUs.

Learn More

ROBOTIS AI Autonomous Robots

Robot Innovation

Pioneering a new era through the development of artificial intelligence robots with physical intelligence to coexist with humans.

Learn More

Raspberry Pi for Home and Industry

Accessible Computing for Everyone

Raspberry Pi partners with Arm to make AI efficient, flexible, and cost effective with new capabilities and increased efficiency.

Learn More