How we interact with technology, through the interfaces we use, has changed radically with people becoming more comfortable using touchscreens whether in their private lives, professionally or when interacting with public services such as banks and public transport.

When it comes to displays capacitive and multi-touch resistive touchscreens have provided users with a host of new capabilities and options.

But is that about to change in light of the Covid-19 pandemic? How can we ensure that the touchscreens we use, especially in the public realm, are safe to use?

“Even before the current Covid-19 pandemic, there had been growing concern over the hygiene of public touchscreens. Reports showed that touchscreens used on some self-service kiosks in fast-food restaurants tested positive for E. coli bacteria, for example, and that has brought into question the hygiene of publicly used touchscreens,” says Ian Crosby, Sales & Marketing Director, Zytronic.

The current medical crisis makes it clearer than ever that more must be done to ensure the use of touchscreens does not cause the spread of harmful bacteria and viruses.

“Despite concerns over the use of touchscreens, they can eliminate the need for face-to-face interaction with customers, helping maintain proper social distancing and helping contain Coronavirus and other infections,” according to Crosby. “They are also more easily cleanable than physical buttons because they provide a smooth and continuous surface. They are simple and easy to clean properly; with regular cleaning, they can provide a much more hygienic alternative to conventional input methods on publicly used machines.”

Despite these advantages, changes are needed to the way we approach public touch screen usage in order to provide the safest possible solution.

“There are several key steps Zytronic recommends, which can be taken to promote safety and hygiene in public touchscreens based on the problems currently found on public touchscreens,” says Crosby. “Installing flush edge-to-edge interfaces will make cleaning and disinfection far easier, as bacteria tends to be found in the nooks and gaps of public machines and touchscreens. Regular cleaning and disinfection of publicly used touchscreens is essential to maintain proper hygiene.

“Many touchscreen technologies, such as projected capacitive touch, can respond to the touch of a gloved hand or stylus. This means users can safely interact with touchscreens whilst minimising the chance of catching a virus or infection or spreading it themselves.”

According to Crosby, specialist coatings for touchscreens are also available which can slow the spread of, or even kill, bacteria.

“In situations where fixed function keys are necessary, solutions such as the ZyBrid VK can be implemented, where the keys are part of a single, uninterrupted glass surface which allows for easy cleaning. Where tactility is needed, touchscreens are now available which implement machined features such as dimples, grooves and dials into the display. With these options, the glass remains unbroken and proves superior to moving buttons in terms of ease of cleaning.”

Audio alternative

Health experts suggest that COVID-19 is here to stay for the foreseeable future, so are there alternatives to touchscreen kiosks because people may be wary of using them - will they really want to touch a screen that lots of other fingers have touched?

“Anti-bacterial touchscreens exist but this misses an important factor,“ explains Dave Betts, chief science officer at AudioTelligence, which specialises in blind audio signal separation. “The kiosk has to be perceived to be safe and hygienic by customers, as well as being genuinely safe and hygienic.”

Could we remain socially isolated simply by talking to the kiosk, using a contactless payment system and then picking up our order from a service counter?

“The amount of physical contact would be minimised and the process made quicker. Speech contains, on average, the equivalent of 39 bits of information per second – whereas a touchscreen menu system is an order of magnitude slower. It is much quicker to say "two burgers with chips, one with cola" than go through the laborious process of finding the right buttons to press on a touchscreen,” Betts suggests.

However, if you decide to use a voice activated system how do you overcome the problems of noise and impact of other people nearby?

Automatic speech recognition (ASR) has been around for many years and its reliability is improving, according to Betts. “It is most reliable on limited vocabulary systems. But that's fine as most kiosks only need a limited vocabulary. However, if you add in a noisy environment with lots of other people around, things are not so great.”

Kiosks tend to be found in high street shops where several kiosks will be clustered together, which raises the question – how can you ensure that your order doesn’t get mixed up with the person at the neighbouring kiosk?

“It's tempting to think that noise suppression is the answer. There have been some amazing advances in noise suppression in the last few years,

fuelled by the artificial intelligence revolution. But these techniques share a common theme – they sound amazing if the signal was already clearly intelligible,” explains Betts.

“Audio intelligibility is not the same as audio quality. If you take clean speech and add noise, the quality will sound worse and worse but the speech will still be utterly intelligible until, suddenly, it isn’t. Generally, humans struggle if speech and noise are about the same level at the microphones.” He continues, “Recently, Nvidia RTX was in the news with several online demos. Whenever you listen to a demo like these, you should ask yourself: “How easy was it to understand the noisy speech?” The answer will be "quite easy". Now if you could listen to what a microphone picks up in a noisy shopping mall, the situation will be different.

“It will be hard for someone to understand the raw speech – let alone an ASR system. And that’s exactly where noise suppression falls down – it can’t pull out the speech cleanly from such high noise levels.”

Could beamforming hold the key? Perhaps, says Betts.

“It’s surprisingly difficult to do well, however, unless you are willing to invest in a large number of expensive calibrated microphones and even then it’s nowhere near good enough for a general high-volume kiosk.”

Blind Source Separation

An alternative, according to Betts, is blind source separation (BSS).

“This technique simply needs between four and eight off-the-shelf microphones, with no calibration required. The array geometry is flexible, providing the array is between 5cm and 30cm across. Ideally there would be a clear 'line of sight' to the customer and, if space allows, a 2D array is preferable – but a linear array does also work.”

BSS is able to separate the incoming audio back into its constituent sources automatically. So not only is the customer's voice brought out of the background noise, it is clearly distinguished from the voice of the person at the neighbouring kiosk.

“This is all done with data-driven machine learning. The system is continuously analysing the sound field and can pick out the speech of the person in front of the kiosk – adapting automatically to the lunchtime rush or the quiet of a 2am motorway pit-stop. Just like a human – but with no social distance required.”

As the concepts behind BSS are mathematical, it can be implemented on any general computing device.

“The CPU cost is well within the capabilities of a modern Arm processor that supports single precision floating point. Then it needs memory – as BSS is a data-driven approach, it needs a frame store to keep all the audio it’s analysing. For 16kHz and eight microphones, that frame store could be as much as 40MB – totally within the capabilities of a modern Arm processor. Optionally, a camera with face detection could help with ensuring the correct customer is selected,” suggests Betts.

Then there’s the ASR and speech-to-intent system.

“Google managed to port its speech recognition system (currently English only) on to a pixel phone in less than 80Mb of memory. Similarly,

specialist multilingual speech-to-intent systems for a limited vocabulary can be implemented in under 500Kb of memory, depending on the size of the vocabulary.”

The COVID-19 pandemic has certainly highlighted the importance of good hygiene practice and strict implementation of health and safety procedures, and from traditional touchscreens to innovative audio solutions companies are looking to respond.