Aug 25
3

Dawn of a New Generation

Presentation - EOS_Page_01

A few weeks ago QuickLogic introduced its new, revolutionary EOS Sensor Processing Platform.  (In Greek mythology, Eos is the goddess of the dawn – symbolizing that this product represents “the dawn of a new generation of sensor processing systems”.)  The announcement garnered a lot of attention in the editorial and analyst community, with over 200 articles published.  So what was all the excitement about?

 At a high level, it is about a tremendously capable sensor processing platform operating at an amazingly low level of power consumption.  Specifically, the platform integrates three dedicated processing engines – Sensor Manager, patent-pending Flexible Fusion Engine (FFE) microDSP, and an ARM Cortex-M4.   We save power by using more purpose-built hardware to do the sensor processing, and then leave the ARM core asleep longer.  We believe most of the always-on, real-time sensor processing can be done in the FFE, and then the Cortex-M4 can be used to run a Real-Time Operating System, and our sensor framework.  And if our customers need more computing capability than the first FFE?  We can build a second one using the embedded, ultra-low power in-system reprogrammable FPGA.

 Have you heard of always-listening?  Through our partnership with Sensory, we now integrate a dedicated block for always-on voice triggering and phrase recognition to enable voice-driven commands at very low power levels – meaning, always listening without the negative impact to battery life.  Finally, the platform is designed to work seamlessly with our extensive SenseMe algorithm library as well as customer-developed algorithms.

 Customers who choose the EOS Sensor Processing Platform will find that they are able to easily integrate a wide range of advanced user functions without breaking their power budget.  These include: pedestrian dead reckoning, sports monitoring, biological and environmental sensor applications, gesture and context awareness, augmented reality and gaming.  Just imagine what you could do…

This entry was posted in EOS and tagged , , . Bookmark the permalink. Follow any comments here with the RSS feed for this post. Post a comment or leave a trackback: Trackback URL.

3 Comments

  1. Richard Dohr - Aug 28, 2015 @ 5:09 am | Permalink

    An always-on microphone could determine when a noisy environment merits making the ringtone louder. It could change settings without having to unlock the phone.

    Reply
  2. danielle - Aug 25, 2015 @ 3:44 pm | Permalink

    Hi Brian,

    Fun!

    In order for an iPad to “listen” it needs to be plugged in to power. In order for the Apple Watch to “listen” it needs be “triggered” by rotating the wrist. How is the S3 trigger better than an Apple Watch trigger? Could you describe some advantages of not having to have a wrist rotation in order to trigger listening? In other words are there some cool sounds or events that can be imagined that a device could be triggered to perform from a distance or with a sound other than a human voice command?

    Also, what kind of device do you think would feature this QuickLogic technology first?

    How soon could I purchase a product that contains the S3 with “always listening”?

    Thanks, Danielle

    Reply
    • Brian
      Brian (QUICKBLOGGER) - Aug 28, 2015 @ 10:00 am | Permalink

      The S3 voice triggering capability is through our partnership with Sensory and their TrulyHandsFree software voice recognition product, which includes a set of triggers and commands. One of the value propositions is that it runs deeply embedded on the EOS S3 platform – meaning, it doesn’t need to go off to the cloud for processing the pre-defined triggers and commands. Doing processing locally means the response time is fast, and in the case of a wearable device, means it doesn’t need to be paired with a smartphone for certain information like step counts. It’s worth noting that the commands are phrase spotted which means they are natural to use. You can speak naturally – “Hey, what is my step count today” for example. Phrase spotting also means that the commands work in high noise since they can spot the command among the noise. Another direct benefit of deeply embedding in hardware – especially with our implementation that hardens the Sensory Low Power Sound Detector (LPSD) in our EOS platform – is that it runs at lower power than with a pure software implementation. Much better for battery life.

      OEMs have the choice when to enable the voice trigger – one could implement it ‘always listening’, or they could have it activated by a rotation of the wrist if they so desire. I like to go on runs and bike rides, and I can tell you that I would prefer to not need to rotate my wrist to activate the always listening feature to check things like my step count, distance or calorie burn.

      We’ve targeted the EOS S3 at smartphones and wearable devices, so we believe those will be the first form factors you will see a feature like this from QuickLogic.

      As far as how soon you can purchase a product containing the S3 – please tune in to our quarterly calls or keep an eye on our future press releases. We always strive to keep the public aware of our progress with our new solutions to the extent that customer Non-Disclosure Agreements permit.

      Reply

Leave a Comment

  • Submit a Question

    Fields marked with an * are required
  • Subscribe to Blog

    Enter your email address to subscribe to this blog and receive notifications of new posts by email.

  • QuickLogic Social Sites

  • Recent Posts

  • Recent Comments

  • Categories

  • Archives