Background Noise Elimination a Key to Ensuring Good Interactions Through Voice

AI Technology Offers One Method for Suppressing Noise on Both Ends of a Call

Background noise reduction for customer service

It is no secret that customers and the companies with which they do business have shifted to digital channels to handle many interactions. In addition to being more convenient for both customers (chat, messaging apps, and social media interactions often feature shorter wait times and the ability to interact while doing other things) and companies (agents can often serve more customers, and more customers simultaneously), the fact remains that voice communication is still a viable channel for higher-value customers and interactions.

However, according to research conducted in 2022 with 1,000 consumers across the UK and US on behalf of IRIS Clarity, a provider of an AI-powered voice isolation app, nearly half of customer service calls are abandoned due to background noise, with 42% hanging up immediately when noise is detected. This is particularly troubling for contact centers, as the research found that more than half (54%) of customers surveyed indicate they use voice channels solely for critical issues.

The issue of background noise is not limited to traditional call centers; as more organizations incorporate a work-from-home model, the ability to control background noise is further limited. Many workers do not have a separate, dedicated workspace that is isolated from noise generated within or outside of the home, and as a result, call quality can suffer greatly. Further, according to the IRIS Clarity survey, customers often are the source of background noise, due to the location from which they are calling, with public transport (39%), the street (36%), when out socializing (34%), and at a place of work (31%) topping the list of locations outside of the home.

IRIS Clarity is one vendor that is marketing its background noise solutions to contact centers, and has focused on providing a solution that will work on suppressing noise on both the agent and customer sides of the conversation. Based on technology that was originally developed for Formula One racing communications, the noise-reduction application uses AI algorithms that are trained on thousands of hours of background noise, allowing a voice stream to be isolated from noise, improving call clarity and understanding.

The solution can be deployed at the agent’s location, serving as a virtual microphone and speaker that intercepts all audio in and out of that device and cleans it up in real time, according to Jacobi Anstruther, Founder and CEO of IRIS Audio Technologies. “It’s quite a lightweight, easy integration, with a two-minute download onto whatever the device may be,” Anstruther says. “We [also] have an SDK solution that can be embedded into a VoIP system or a video conferencing system.”

Anstruther notes that the company is also working on a chip-based solution that could be integrated into hearing aids or headphones, and has also built the functionality into the Dante system, a combination of software, hardware, and network protocols that delivers uncompressed, multi-channel, low-latency digital audio over a standard Ethernet network using Layer 3 IP packets. IRIS Audio recently completed an integration with a leading motor race series in the US to provide its noise-reduction technology across its 48 channel Dante communications system to clean up teams, race control, emergency services, and broadcast communications.

For contact centers, the goal of noise reduction is to reduce noise to improve clarity, so that customers do not get frustrated because they need to repeat themselves when speaking to agents, or vice versa. The IRIS Audio team uses four parameters to assess the level of noise, including the PESQ scale (which refers the ability to remove noise while maintaining the same level of intelligible audio), latency (referring to any sort of lag introduced when using software to remove noise), CPU utilization to run the software, and system memory usage. According to Anstruther, IRIS Audio exceeds all other solutions on the market now across each of these measures, including solutions from Google or Zoom.

Beyond the technical specs, Anstruther says that IRIS Audio provides its solution to contact centers far more economically than its competitors, with pricing averaging at around $5 per agent, per month, a price that incorporates typical discounts applied based on the number of seats and the length of the contract. “We’re averaging contracts between two to three years,” Anstruther says, noting that customers generally set up a proof-of-concept with 50 or 100 agents for a month. However, he adds, “once you’ve had it, it’s very hard to go back [to not using it.]”

Perhaps the biggest reason to incorporate some noise reduction or cancellation software is that there is only so much that can be done to physically reduce noise in a call center, such as putting up sound-absorbing material and partitions between agents, and that can be costly and does not always work. Further, both remote agents and customers increasingly are conducting calls from areas where soundproofing or physically eliminating noise is impossible, thereby portending the need for a different approach.

Removing excess noise helps to eliminate repetition of words or phrases, and the lack of understanding that comes with poor call quality, and can help position the contact center support option as a premium benefit. “If your contact center use case is a premium [offering] of the product, then you should be delivering a premium quality of audio,” Anstruther says, noting that better audio can also improve overall contact center performance metrics.

“You obviously want your successful calls to be shorter, so your agent can go onto the next call,” Anstruther says.

Author Information

Keith has over 25 years of experience in research, marketing, and consulting-based fields.

He has authored in-depth reports and market forecast studies covering artificial intelligence, biometrics, data analytics, robotics, high performance computing, and quantum computing, with a specific focus on the use of these technologies within large enterprise organizations and SMBs. He has also established strong working relationships with the international technology vendor community and is a frequent speaker at industry conferences and events.

In his career as a financial and technology journalist he has written for national and trade publications, including BusinessWeek,, Investment Dealers’ Digest, The Red Herring, The Communications of the ACM, and Mobile Computing & Communications, among others.

He is a member of the Association of Independent Information Professionals (AIIP).

Keith holds dual Bachelor of Arts degrees in Magazine Journalism and Sociology from Syracuse University.


Latest Insights:

Oracle Exadata Exascale Debuts Aiming to Unite the Best of Exadata Database Intelligent Architecture and Cloud Elasticity to Boost Performance for Key Workloads
The Futurum Group’s Ron Westfall examines why the Exadata Exascale debut can be viewed as optimally uniting Exadata with the cloud to provide customers a highly performant, economical infrastructure for their Oracle databases with hyper-elastic resources expanding Oracle’s market by making Exadata attractive to small organizations with low entry configuration and small workload affordability.
Brad Tompkins, Executive Director at VMware User Group (VMUG), joins Keith Townsend & Dave Nicholson to share insights on how the VMware community is navigating the company's acquisition by Broadcom, focusing on continuity and innovation.
On this episode of The Six Five Webcast, hosts Patrick Moorhead and Daniel Newman discuss AWS Summit New York 2024, Samsung Galaxy Unpacked July 2024, Apple & Microsoft leave OpenAI board, AMD acquires Silo, Sequoia/A16Z/Goldman rain on the AI parade, and Oracle & Palantir Foundry & AI Platform.

Latest Research:

The Signal65 study evaluated the overall economics of a VCF environment deployed with disaggregated vSAN storage compared to the same environment deployed with a leading SAN array. To create an even comparison between storage solutions, the compute environment was kept identical when modeling both solutions. The analysis modeled both environments over a 5-year period and compared costs of hardware, support, licensing fees, and operational costs. Operational costs were calculated using an in-depth analysis of the administrative time and complexity required for each solution.
In our latest Market Insight Report, Image Generation Technology for Enterprise Use, we define the technology that enables text-to-image generation, cover the potential use cases and benefits of the technology, discuss the technical models and processes behind the technology, explore the risks involved with using the technology, and focus on the current and future commercial market for the technology.