EOS™ S3 Sensor Processing Platform

The world’s only ultra-low power SoC which combines always-on/always listening voice processing with fast, efficient and sophisticated sensor processing capabilities

This ultra-low power, processing-efficient system enables OEMs to extend battery life while designing in sophisticated, always-on sensing capabilities on mobile devices. Advanced sensor algorithms such as voice triggering, motion compensated heart rate monitoring, and indoor navigation can be achieved at significant power reduction compared to competing MCU-based solutions.

Unlike traditional MCU-based solutions, the EOS S3 is a multi-core, sensor processing system that enables sophisticated algorithm partitioning to facilitate the lowest possible power consumption for the designated task.

The EOS S3 employs not only fundamental, but also very sophisticated, always-on, context-aware sensing capabilities while staying well within the strict power budgets of smartphone, wearable, and IoT designs.

Ultra-low power always on listening

Concurrent voice recognition and hub sensor processing

Customizable Hardware (eFPGA)

Customizable Software (OPEN Software)

2-mic beam-forming and noise-suppression

Dedicated uDSP (Flexible Fusion Engine™) for sensor processing

On/Off-body detection for lowest power mode

Low-Voltage (0.85 VDD) option for 33% power reduction

EOS S3 Sensor Processing Platform Block Diagram

  • Sensor Manager — Autonomously manages and controls all sensors
  • Flexible Fusion Engine (FFE) — 10MHz DSP-like processor supports always-on computational processing at one forth the power
  • eFPGA — Enables custom logic functions and I/O expansions
  • Voice Processing — Hard-coded Low Power Sound Detector (LPSD) and PDM to PCM conversion minimizes audio processing power
  • ARM Cortex-M4 with FPU — Up to 80MHz and 512 KB SRAM for general purpose processing and running O/S
  • Serial I/O — SPI Master/Slave I2C, UART
  • System — DMA, Integrate RTC Oscillators, ADC, LDO

Voice is the New Touch for Hearables

Hearable Diagram
Challenge
Solution
As more functionality is enabled in a tiny hearable, delivering a long battery life is a challenge Dedicated Low Power Sound Detection (LPSD) hardware to listen while rest of chip sleeps
Loud ambient noise 80MHz M4F core to run 2-mic beam-forming noise suppression and leading-edge voice recognition solutions
High accuracy voice detection in noisy environments OS S3 software package running leading-edge wake word engines
Preserving battery life while always listening for wake word phraseAlways-on Low-Power Sound Detection at 50uA (@Vbatt)
Fast Time-to-Market for consumer productsFreeRTOS open source reference design package

Wearable Diagram

Wearable Diagram
Challenge
Solution
Battery life is the #1 problem as small devices cannot accommodate a large battery and consumers want long battery lifeDedicated Low Power Sound Detection (LPSD) hardware listens while rest of chip sleeps, thus saving battery life
Using multiple different sensors, at varying samples rates Dedicated uDSP (Flexible Fusion Engine) to autonomously manage multiple sensors at different data rates while M4 is idle

Voice is the new UI (IoT SmartHome)

Typical IoT Device
Challenge
Solution
Maximizing battery life of low-power IoT smart devices Dedicated Low Power Sound Detection (LPSD) hardware to listen while rest of chip sleeps
Need voice control while playing music or soundBarge-in AEC feature

Smart TV & Home Appliance Diagram

Challenge
Solution
Low standby power is required to meet national and international consumer product energy ratingsDedicated Low Power Sound Detection (LPSD) hardware to listen while rest of chip sleeps
Convert PDM mic signals to I2S format for apps processor to run far-field voice recognition after being woken by S3eFPGA block allows for flexible hardware design

EOS S3 Documentation

Datasheet
Product Brief