Contact Centers Are Turning to AI Filters to Protect Staff from Angry Customers

Service teams are levering AI-powered filters to modify the voices of customers and agents

Contact Centers Are Turning to AI Filters to Protect Staff from Angry Customers
Contact CentreInsights

Published: June 17, 2024

Charlie Mitchell

Handling angry customers is an unenviable task that every contact center must deal with.

Unfortunately, widespread issues such as short staffing, long wait times, and misfiring self-service deployments only exasperate the problem.

Last year, 43 percent of US customers admitted to yelling or raising their voices to express displeasure about their most serious problems, up from 35 percent in 2015.

As this problem perseveres, attracting staff into the space continues to be difficult, with AI not yet providing the silver bullet many expected.

Indeed, AI and generative AI (GenAI) are not proving as effective in stemming the aforementioned causes of customer anger.

Consider short-staffing. According to Gartner, 61 percent of customer service leaders expect headcount reductions of only five percent or less due to GenAI.

However, AI-driven self-service and workflow automations are not the only tools that could help to combat the causes of customer anger.

Now, contact centers are using AI-powered voice filters to tackle the problem head-on.

Using AI to Turn Angry Customer Audio Into Cool, Calm Audio

SoftBank Corp, a Japanese telecoms giant, is testing AI software that softens the tone of irate customers – according to Bloomberg.

In doing so, the organization expects to reduce its service agents’ stress levels.

SoftBank also noted that it hopes the technology – which blends voice processing tech and AI-enabled emotion recognition – will allow it to boost customer retention.

As quoted by Bloomberg, the organization stated:

With this solution, we aim to maintain good relationships with customers through sound communication while ensuring the psychological welfare of our workers.

One potential caution is that if agents can’t correctly adjudge the customer’s tone of voice, they may not deliver sufficient empathy or grasp the immediacy of the issue.

However, the testing will help identify such issues, and – if those tests are successful – SoftBank will consider commercializing the solution within the next two to three years.

In addition, the telecoms giant aims to apply AI across more of its operations, claiming to view the technology as a path to a “happier future for all”.

One thing is certain: agents who don’t have to bear the often overwhelming brunt of customers will be much happier in their jobs.

More Contact Centers Are Modifying Voice Conversations

While using AI to modify the tone of customer-agent conversations may seem like a new concept, many contact centers have leveraged such applications over the past two years.

Only last week, CX Today reported that 12 of the top 20 customer service BPOs are modifying their agents’ voices with AI, so reps from around the globe sound as if they are speaking with non-accented American English.

In doing so, these contact centers strive to swerve any potential customer bias, which fuels negative emotions like anger.

Sharath Keshava Narayana, Co-Founder & COO of Sanas – a provider of real-time accent translation technology – discussed the power of the solution, stating:

Before, these reps experienced an undercurrent of bias in every other call they handled, and we’ve taken that familiar, unpleasant feeling away.

After deploying Sanas, those BPOs – alongside other global enterprises – cut agent turnover rates by as much as 50 percent, with reps receiving significantly less abuse.

Those results emphasize the power of such voice applications and perhaps underline the potential of use cases like the AI-fueled angry customer filter.

So, while many contact centers only consider AI through the lens of bots and automation, these applications may well be the most impactful ally for agents – especially while voice remains the number one customer contact channel.



Agent AssistArtificial IntelligenceSentiment Analysis

Share This Post