Enhancing audio system designs with digital signal processing

-November 13, 2016

By the way, just a quick comment on the designation of geek: I am an analog geek, and when you attain the designation of geek, you know that people are mostly referring to your steadfast dedication as an enthusiast, obsessed with a vocation toward expertise in a particular area, in my case it is analog electronics.

While at the MEMS & Sensors Executive Congress 2016 in Scottsdale, AZ I had breakfast with Will Tu, VP Sales and marketing from DSP Concepts. My goal was to understand how to effectively design audio systems with microphones to enhance the system performance if I am an analog designer.

I need to be sure that the microphones in my system can be virtually immune to a noisy background. Maybe I want to design a security system in which I can audibly detect breaking glass or other unusual noises that can identify an intruder. Or maybe I may want to design a baby monitor to detect breathing or breath stopping. I have no coding skills or DSP expertise that will be needed to optimize my microphone(s) performance in my design. Gone are the days where most companies had different design engineers for software, digital design, analog or power design. Now many small to mid-sized companies run much leaner with their design staffs.

Since microphones find their way into smartphones and other portables in areas like security, wearables, cameras, headphones (wireless types are growing quickly), and automotive applications using multiple microphones, I may now need to create directional arrays that are electronically steerable for best performance. (You can see Dr. Paul Beckmann’s white paper on enhancing the utility and performance of microphones with DSP processing here)

Will Tu introduced me to their Audio Weaver modular DSP software which is compatible with most major DSP families and embedded cores. And, if you do not have the manufacturing capability to incorporate a DSP into your design, there are DSP processing boards available from ST Microelectronics and Danville Signal.


Will Tu’s presentation regarding How MEMS Microphones will create a new class of IoT Devices: AoT


Dr. Beckman’s white paper, co-written with Gerard Andrews from Cadence Design Systems, tells us how the use of a user-friendly graphic audio development tool with a low-power audio image processing DSP can assist designers in creating an embedded audio product.

Will Tu also discussed applications such as cough detection and analysis to tell a user when they should use their inhaler before an emergency need occurs or an acoustic event detection like the NYPD ShotSpotter which can be further enhanced with good microphones and  effectively-enhance the performance by using a DSP with the Audio Weaver program. This type of design can be enhanced to possibly locate where a shot was fired from in a multi-floor building and  pin-pointing the floor and maybe the window from which it originated. Audio Zoom capability can be designed into a system as well so that you will be able to pick out a voice or sound in a crowd using multiple adaptive audio beam-forming techniques with Audio Weaver.

Will Tu also told me that there will be another exciting announcement coming shortly, which I cannot divulge at present, but I will post that news the moment that DSP Concepts announces it.

Also see:

Loading comments...

Write a Comment

To comment please Log In