There were 285 million calls to European Union (EU) emergency numbers in 2023, representing a 6% increase since 2021. In the UK alone, there were 37.7 million calls to 999 and 112 numbers in 2024, an increase of 2 million calls since 2022. From the moment an emergency call is placed, to when the incident is resolved, a dedicated and coordinated team of call handlers lie in between to ensure this process is executed smoothly. Day in and day out, these professionals manage an immense volume of calls, each demanding their utmost attention and skill in navigating a diverse range of often high-stakes situations unfolding remotely. Whether it be alarms blaring in a crowded area or loud fighting in domestic disputes, when a person’s life is at stake, a call handler has no choice but to try to extract as much information as they can. The person making these calls is typically having the worst days of their lives and in such cases, the audio quality is often less than ideal – presenting significant challenges for effective communication.
An operator must deal with numerous pieces of information simultaneously and in the moment of a call, the luxury of a break does not exist. Needing callers to repeat information slows down the rescue process and endangers the safety of the victims.
What is AI Audio Enhancement?
AI audio enhancement is a new feature which can be implemented in control room software to help mitigate the existing auditory problems faced by call handlers when receiving calls from busy, noisy environments. Basic denoising technologies have existed for a while now, found on most consumer cellular devices, but with new AI call noise suppression, operators are now given the simple, yet powerful ability to enable real-time de-noising on an as-needed basis. Motorola Solutions Guardian ICCS – telephony and radio dispatch software for British and European emergency services – presents AI noise suppression as a simple button on its user interface: call handlers can choose to toggle on or off the noise reduction feature, allowing immediate usage within 16 milliseconds.
How Effective is AI Audio Enhancement?
In a field like emergency services, where mistakes can have serious real-life consequences, incorporating a technology like AI into that delicate process requires thorough management in order to successfully integrate. The line of communication between caller and call handlers is crucial and the introduction of new technology into this space cannot hinder this link in any way.
A spectrogram comparing the before and after of AI audio enhancement. Top: shows blurred frequency levels throughout entire call duration. Bottom: shows specific frequency only when there is communication occurring between caller and call handler.
Guardian ICCS’s AI audio enhancement has been proven to work in a multitude of situations from noisy call centers to motor racing teams’ comms. We discussed the topic with Chris Redman, an ex-call handler for West Yorkshire Police, one of the busiest police forces in the UK, to get the perspective of someone who has directly dealt with emergency calls. Chris mentioned his worst enemies during calls were alarms blaring, wind noises, crying and yelling, and vehicles speeding by on motorways. These background noises not only severely impact call handlers’ ability to assist callers, but also compound further stress onto their already mentally demanding day-to-day.
The Guardian ICCS team wanted expert feedback on AI audio enhancement, so Chris was involved in the initial testing of this feature: some of his immediate questions were, “How would you use it as an operator?” “Will it actually be any good?” “Is it just a gimmick?” These are all valid and common concerns but upon usage, Chris shared that he was “very impressed with it right off the bat,” noting the ability to turn off background noise being a “game-changer.” The option to switch the noise reduction on and off is crucial as Chris mentioned how background noise can be used both as evidence and valuable contextual information for better understanding the situation.
We also talked with a former call handler for Thames Valley Police, Carl Warner, who provided insight into additional concerns about the ease of adoption and deployment for current phone operators. Many agencies are working with older equipment and software, and as a result, there could be roadblocks to incorporating new technology into existing systems – especially for smaller forces working with limited budgets. Having worked in such an agency, Carl acknowledges the difficulty of innovating in spite of large-scale infrastructure problems; however, he ultimately views AI audio enhancement as “beneficial technology stepping in the right direction of embracing innovation in a responsible manner.” The issues surrounding upgrading existing police force systems mainly lies in concerns with installation and deployment of new software. However, because Guardian ICCS is cloud-hosted, it eliminates the need for costly or proprietary hardware – enabling organizations to receive regular enhancements over the internet. As such, control rooms can benefit from minimal downtime, receiving cutting-edge features promptly.
What is in Store for the Future of AI Audio Enhancement?
Being a new technology, as features are considered and adopted, more changes are set to come. In its current state, AI audio enhancement is used in telephony denoising, but Motorola Solutions is exploring implementation options for radios in Europe and the UK too. Radios are a key component in emergency service communications but outside of current adaptive noise suppression technology, AI audio enhancement technology has yet to be fully utilized in the control room, leaving space to expand. Another feature currently being developed involves maintaining both the original and AI noise suppressed audio tracks. In our talks with ex-call handlers, a shared concern was the preservation of core evidence. Capturing dual streams of audio, both altered and unaltered, can be crucial in court as it provides evidence against tampering or placing blame on technological issues. On top of that, to further an operator’s call-to-call adaptability, our team has been working on the ability to adjust the degree of noise suppression, providing operators with an additional layer of control. These changes are paving the way for even more exciting innovations, and it is inspiring to see how new features such as AI noise suppression are making their way into the public safety space.
In the midst of falling rubble and distressed crowds, improvements in noise suppression technology would be transformative.
