The changing nature of work

Image supplied by EPOS Enterprise
Image supplied by EPOS Enterprise

COVID-19 is driving new ways of working, and the changes to work styles that it is bringing will be permanent. Gartner has predicted that 41 per cent of employees will likely continue to work from home in some capacity post-pandemic (whether that is full time remote working or a hybrid blend of office and remote time). A later study found that 82 per cent of company leaders will develop a formal remote work policy and will allow employees to take advantage of it.

What this “new normal” means is that the ongoing productivity and efficiency of the organisation will hinge on employees having access to quality audio experiences.

Remote communication challenges

Productivity suffers when noise is distracting. The audio solutions brand EPOS recently conducted research into the impact that audio has on the working experience in order to gauge just how significant the impact could be and found that  that 69 per cent of workers end up spending extra time on tasks due to poor sound quality, such as misheard communication or direction. Further, 95 per cent admit that concentration and efficiency at work suffers a setback due to poor audio.

At the same time, poor sound quality that has previously been a mild irritation for many employees due to the low frequently use of video conferencing tools is quickly becoming disruptive to the core of how people are working. There has been a boom in video conferencing use: due to the pandemic, just one player in the video conferencing space – Zoom – saw a lift in daily users from 10 million to 300 million, and the company revenue jumped by 169 per cent in a single quarter.

Putting it in real numbers, businesses that employ over 100 people risk losing $US70,000 per year from lost productivity. Organisations that use poor audio to communicate with their customers risk even more – bad audio outcomes mean 23 per cent of clients walk away dissatisfied, and 18 per cent of businesses lose a key piece of work. Bad audio is bad business, in other words, and an investment in quality audio experiences should be a priority for any organisation while enabling working from home.

Bringing new smarts to headsets

What is causing these low-quality audio experiences? In many cases, it’s the ambient environment, with most home offices not set up to deliver a pristine audio experience. Ambient noise from neighbours, the street, family and pets can all create distracting noise, which can then be caught in the microphone, reducing the quality of the call for everyone.

By understanding the human environment and way that we interact, EPOS has been able to develop start-of-the-art solutions and approaches to audio equipment that suit the modern working environment and work styles. While we all might reach for a noise-cancelling headset in an effort to improve the audio that we hear on a conference call, that is not going to be the solution in itself, because while noise-cancelling technology benefits the user with the headset, the ambient noise environment can distract and frustrate those that they are talking to.

EPOS’ drive to craft premium audio experiences with a focus on craftmanship and design, performance and reliability has led it to innovative AI as the solution for improving the audio experience for all participants in an audio or video meeting. The ADAPT 660 is the world’s first  UC-certified headset that incorporates a machine learning-enhanced voice pickup. In application what this means is that the ADAPT 660 EPOS AITM continually monitors the audio surroundings and instantly adjusts the microphone settings to the ambient environment based on a virtually infinite number of parameters.

For AI to work in the context of a headset, it is important that its implementation does not result in an otherwise poor experience – for example, by weighing the headset down or making it too bulky. The size of AI applications has become something of a battleground of innovation among developers – as noted by an MIT study: “AI has a problem: in the quest to build more powerful algorithms, researchers are using ever greater amounts of data and computing power, and relying on centralized cloud services. This not only generates alarming amounts of carbon emissions, but also limits the speed and privacy of AI applications.”

The “holy grail” in response to this challenge has been termed “Tiny AI”. Tiny AI refers to the research that goes into shrinking existing deep-learning models without losing their power or capability. Tiny AI is what runs Google Assistant on smartphones without having to first sent a request to a remote server. It’s what runs Siri’s speech recognition capabilities locally on iPhone.

To maintain the comfortable use of the ADAPT 660 headset, EPOS has created an extreme form of Tiny AI, in which over 50 of its neural networks fit within the same memory space as a single JPG photo, meaning that this technology can be implemented into any standard-sized headset moving forwards.

To further assist professionals, EPOS provides a software tool – EPOS Connect, that allows for the easy management of firmware and updates, as well as the ability to further personalise the device settings to suit the working situation – ideal for the professional that finds themselves using a number of different environments (such as the home office and the workplace post-pandemic).

As office environments start opening up, professionals are still going to find themselves spending more time video conferencing. Individuals within teams will be more flexible on whether they come into the office, and enterprises will progressively tap overseas and remote expertise to fill skills gaps. The health and productivity of organisations is going to be very closely tied to the quality of the audio equipment that the organisation uses, and it’s audio equipment developed by experts in understanding the human nature of communication and collaboration that will deliver the best results.

Click here for more information on EPOS solutions. 

Copyright © 2020 IDG Communications, Inc.