This post is a behind-the-scenes look at one of the Unimpossible Missions experiments that a team of Global Research experts conducted in 2015. Click here to meet the entire team.
Can we debunk the idiom “Like Talking to a Wall”? This was asked of me recently and was the start of an exciting experience culminating in performing an experiment on the Berlin Wall.
Our objective was to assess the feasibility of using vibration measurements to detect vibrations from a person speaking at a wall and reconstruct them back into intelligible speech. We would demonstrate this on several walls in our facility in Niskayuna, NY and, if successful, on the Berlin Wall in Germany. This was a unique application to our team, as we normally use vibration measurements to detect possible issues in gas turbines, aircraft engines, and other machines and structures.
The Berlin Wall where we performed the experiment is made of concrete 6-12 inches thick. We wanted to ensure we could detect vibrations from a human voice through this medium. First, we researched transmission loss through concrete in the frequency range of the human voice and concluded that the vibrations produced by a human speaking through concrete would have enough energy on the far side of the wall that we should be able to detect them with instrumentation we have in our lab. I work in the Vibrations Laboratory at GE Global Research, where our group specializes in areas such as aeromechanics, vibration-based diagnostics, and noise and vibration measurement and mitigation. We have a good selection of vibration sensors to support the work we do, and our next step was to assess which sensor was best for this application.
We tested a handful of vibration sensors on a wall in our lab, ranging from non-contact laser vibrometers to contact-required piezoelectric patches and accelerometers. After several trials during which one of my colleagues spoke to a wall and I analyzed the vibration measurements from the other side, we down-selected to an accelerometer that provided the best signal-to-noise ratio of speech-to-background noise.
The next step was to determine how to attach the accelerometer to the wall. Since the final experiment was to occur at the Berlin Wall Memorial, permanent adhesives were not an option. We settled on a commercial grade wax – a common approach for attaching accelerometers to structures. The results of our feasibility experiments were exciting. We easily detected not just the presence of vibrations from the voice, but could clearly discern what the speaker was saying! This bode well for a successful experiment at the Berlin Wall.
With several successful experiments under our belt in Niskayuna, we prepared for a trip to Berlin, Germany to perform the experiment at the Berlin Wall Memorial. I had the amazing opportunity to spend a week in this historic city preparing for and performing the experiment on a section of the Berlin Wall. This was certainly one of the more unique experiments I have performed, as I’ve never run a test on a historic landmark before! All the while, it was a humbling experience to spend time at the memorial and reflect on the Wall’s impact on those who lived here during its time. It was a privilege to demonstrate that technology can carry a voice across this boundary.
The day of the experiment was rainy and cold, adding a few new challenges to the experiment. However, with the support of a good team (and a few umbrellas) we were able to measure the vibrations of a man reading a storybook through the West side of the Berlin Wall, filter those vibrations, and play them back through speakers on the East side of the Berlin Wall. The result was a successful experiment and a great avenue to showcase what we can do with vibrations!