Amelia's Weekly Fish Fry

Machine Learning Megaphone: Advancements in Voice Technology

This week’s podcast is all about the advancements in voice technology. Laurent Pilati (NXP) joins me to discuss the challenges surrounding machine learning for voice tech, the details of NXP’s Voice Intelligent Technology (VIT), and what benchmark metrics are involved for the qualification for a voice design. Also this week, I take a closer look at a new special glove-like device developed at UCLA that can translate American Sign Language into English spoken speech in real-time through a smartphone app.

The Future Will Be Open Source: How QuickLogic Is Changing the FPGA Landscape

In this week’s Fish Fry podcast, Brian Faith (CEO - QuickLogic) joins me to discuss Qomu: Open Source SoC Dev Kit That Fits in Your USB Port. Brian and I delve into the details of this new kit, examine its role within the QuickLogic Open Reconfigurable Computing (QORC) initiative, and how this kit is turning FPGA development on its head in more ways than one. Also this week, we take a closer look at a new desktop 3D hologram printer developed by LitiHolo and how you can start creating your own 3D printed holograms. 

A Mesh By Any Other Name Just Isn’t the Same: Breaking Rules and Making the Impossible Possible with Ansys HFSS Mesh Fusion

In this week’s Fish Fry podcast, we're making the impossible possible! We start things off with a closer look at the world’s first metamaterial developed by a team of researchers from EPFL Labs. We investigate the unique properties of this new metamaterial and how this research could pave the way for the development of advanced forms of mechanical metamaterials. Also this week, Matt Commens (Ansys) joins us to discuss HFSS Mesh Fusion. We check out the details of this new mesh fusion technology including how it will open up new avenues for simulation, and why it will help engineers break old rules to overcome the most challenging design obstacles.

Full Stack Future Proofing: How LEVL is Changing the Device Identity Game

Step right up ladies and gentlemen! We’ve got all sorts of EE fun lined up for this week’s Fish Fry podcast. Tim Colleran (LEVL) joins us to discuss LEVL-ID, what sets this device identity technology apart from MAC randomization, and why full stack device Intelligence is key to its success. Also this week, we check out a new research study from Georgetown University that contends that mirroring human visual learning will be key to helping AI software to think more like a human brain. 

 

Brain-a-Palooza! Advancements in Brain-Hand Communication and Your Brain on Code

In our first Fish Fry podcast of 2021, we sit down with Dr. Scott Frey (University of Missouri) and chat about the challenges and newest advancements in brain-hand communication. We take a closer look at how Dr. Frey and his team are helping to personalize recovery from ‘lost’ limbs and the details of a new wireless device that could help people with severe upper limb injuries live better lives. Keeping with our brain-related theme this week, we also investigate a recent study from MIT that reveals that reading computer code is a brain beast of its own and does not precisely replicate the cognitive demands of language or math.

 

It’s All About Confidence: System on Chip Verification and Building Trust into Neural Networks

How do we know that what our neural networks are telling us should be trusted? Can we build confidence into our neural networks so they can answer that for us? According to a new study out of MIT and Harvard, we can and it won’t break the computational bank! In this week’s Fish Fry podcast, we first check out a new way for deep learning neural networks to quickly estimate confidence levels in their output. Keeping with our verification theme, Moshik Rubin (Cadence Design Systems) also joins us to discuss the challenges of SoC verification, the increasing need for system level solutions today, and where he thinks chip level verification is headed in the future. 

 

AI for Everything!

Artificial intelligence and machine learning take center stage in this week’s Fish Fry podcast. First, we take a closer look at how a new algorithm called ART (Automated Recommendation Tool) is ushering in a new age of enlightenment in the world of synthetic biology. We investigate how this revolutionary new algorithm was tested and what it means for the future of bioengineered cells. Aaron Tersteeg (Intel) also joins us this week to discuss Intel’s OpenVINO tool kit. Aaron and I chat about the biggest AI inferencing challenges facing us today, the details of the OpenVINO “write once, deploy anywhere” approach, and what you should keep in mind when starting a new AI project.

Terrible Things to Scare the Children: #adifferenthalloween with element14 and Black Hole Fuel

It’s creepy! It’s crawly! It’s a design contest too! In this week’s Fish Fry podcast, Phil Hutchinson (element14) brings us the goods on this year’s spookily spectacular element14 Halloween design competition. Phil and I chat about some of last year’s creepy creations, why this year’s contest is slightly different than the year before, and how you can enter this unique Halloween design competition. Also this week, I check out how a group of astronomers discovered the oldest known black hole and why this discovery may help unlock the mysteries of the evolution of black holes. 

 

Radar to the Rescue: How Ainstein is Improving Our Safety through mmWave IoT Sensing

In this week’s Fish Fry podcast, we start things off with a very special News You May Have Missed. In this segment, we take a closer look at how a team of researchers at the University of California San Diego School of Medicine (in collaboration with IBM) have identified a “lonely” speech pattern using machine-learning models that can be used to detect loneliness in older adults.  We investigate how machine learning can help us unlock the mysteries of natural speech patterns and why this type of research may help us better understand a variety of psychological ailments. Also this week, Andrew Boushie (VP of Strategy & Partnerships - Ainstein) joins us to discuss the future of mm wave radar technology and the super cool stuff under the hood of their new over-the-door sensor called WAYV Air. 

 

Intelligent Senses: Cyborg Locusts and First Humanoid Robot with Intelligent Vision

Cyborgs, Robots, and Locusts! Oh My! Did you know that locusts can smell explosives? Did you know also that we can now control their brains to pinpoint to exactly what they smelling?  In this week’s podcast, we start things off with an investigation into how two teams of research scientists have created a set of specialized cyborg bomb sniffing locusts. We check out how these locusts are able to  explosive chemicals in the air, how they are able to track these explosive smells and how we can now identify exactly what they are smelling by reading their brain waves. Also this week, Alessandro Gasparini and Alain Pacquin (Immervision) join us to discuss JOYCE: the first humanoid robot for the computer vision community. We discuss the details of about how Immervision and their partners will enable JOYCE with human perceptions and how design contests will assist with the evolution of this first of her kind humanoid robot.