You're not imagining it: Silly sounds make EVs hard to hear coming down the street

Over the past several years, electric vehicles have garnered something of a reputation for their unusual sounds on the road. Otherworldly EV warning sounds have been compared to ' a celestial choir,' a ' flying saucer hum,' and, in one TikTok with 23.5 million views, the song that might play just before ascending to heaven. But the angelic warble that's come to characterize EV acoustics might have a few drawbacks for pedestrians.
A new study conducted by researchers at Chalmers University of Technology in Sweden and published in March examined how well the average person could locate three common types of warning sounds from hybrid and electric vehicles at low speeds. It found that all three of the sounds were significantly harder for pedestrians to locate than the sound of a standard internal combustion engine.
Given that they have no combustion engine, EVs are naturally almost silent. That can be a benefit when it comes to urban noise pollution, but it's not ideal for pedestrian safety. For the past six years, all EVs in the U.S. have been legally required to emit some kind of low-level noise— a prompt that automakers have chosen to interpret in a range of creative ways. But it might be time for some automakers to take another crack at their proprietary EV acoustics.
What do Hanz Zimmer, a didgeridoo, and fighter jets have in common?
Starting in 2019, the National Highway Traffic Safety Administration ruled that all hybrid and electric cars have to be fitted with an external speaker that must 'make audible noise when traveling in reverse or forward at speeds up to 30 kilometers per hour (about 19 miles per hour).' While the law sets expectations for when these noises need to play, it largely leaves the contents of the noise itself up to automakers.
That's resulted in a variety of EV sounds on the road, from a Cadillac alert made using a didgeridoo to the Hyundai Ioniq 5 N's fighter-jet-inspired sound and BMW's portfolio of i4 electric sedan noises by composer Hans Zimmer.
This unusual symphony hitting the roads has inspired quippy commentary on social media. Under a TikTok sharing the BMX iX 50's reverse sound, one user wrote, 'is this ribs by lorde?' And in a video poking fun at Tesla's reverse audio, another commenter joked, 'Every time our neighbour pulls onto the drive with their electric car my husband says 'the spaceship has landed.'' Beyond sounding a bit silly, though, there are a few key shortcomings to the sounds that many automakers are selecting for their EVs.
Why are EVs so hard (and annoying) to hear?
Chalmers researchers examined three of the main categories of EV sounds, also known as acoustic vehicle alerting systems (AVAS): two-tone, multitone, and narrowband noise (a noise concentrated within a small band of audible frequencies, often perceived as a hissing sound).
To compare these sounds to that of an internal combustion engine, researchers studied the reactions of 52 test subjects inside a soundproof chamber. Each subject was surrounded by 24 loudspeakers and given a laser pointer fashioned out of a toy gun. When one of the speakers played a simulated vehicle sound designed to mimic the noise of an EV at a low speed, the subjects were to point the laser toward the sound as quickly as possible.
The tests demonstrated that all the AVAS categories were harder for subjects to locate than the sound of an internal combustion engine. And, according to Leon Müller, a PhD student at Chalmers and one of the paper's authors, one of the sounds was more problematic than the others.
'[The two-tone AVAS] is significantly harder to localize than other types of warning sounds, as well as combustion noise,' Müller says, noting that in a situation with just one vehicle present, these localization errors are relatively small and not particularly concerning for traffic safety. When there are two or three EVs present, though, the situation can get a bit stickier. 'In that case, the participants had much more [difficulty] localizing the cars, up to a point where most participants failed to even detect all presented EVs within an appropriate time,' Müller says.
There are a few reasons why pedestrians might have trouble locating EV sounds. First, Müller explains, combustion noise is a very broadband signal—meaning it contains a lot of frequencies, 'and hence more information for our hearing system to work with.' Second, humans have had substantially more time to acclimate to combustion sounds than artificial EV sounds.
'We humans have learned over the last 100 years or so that cars sound in a particular way and how driving behavior, such as acceleration, is reflected in this combustion noise,' Müller says. 'This potential learning effect might also contribute to differences in localization, especially when we need to 'decode' multiple sounds at the same time. One could expect that we would then also get used to EV sounds within a few years. The only problem is that they currently all sound different.'
A new sound
In the meantime, Müller believes there are two potential avenues to make EV sounds safer.
Currently, U.S. and EU regulations are limited to minimum sound levels in a specified number of frequency bands, which he argues 'allows the warning signals to be anything between a futuristic spaceship sound or a racing car engine.' In the U.S., he adds, regulations don't require a velocity pitch shift, meaning that a car might sound the same going 60 mph as it does at 25 mph. To address these problems, Müller says the regulations should 'make more clear demands on the sound characteristics.'
On the automaker side of the equation, the Chalmers study indicates that a more broadband AVAS signal, similar to the noise radiated by tires when driving faster, is preferable to a two-tone or multitone AVAS.
'[This sound] is potentially less annoying than tonal sounds and has the advantage that we already have 'learned' to interpret this noise since we hear it every day,' Müller says.
In the long term, he adds that adaptive AVAS solutions—like pedestrian detection technology—could help EVs radiate a more advanced warning sound directly in the direction of the pedestrian, thus improving safety and reducing noise pollution.
'One important bottom line here is that we are not saying EVs are bad or dangerous. With the right type of warning signal, they are not,' Müller says. 'On the contrary, they have the potential of reduced noise pollution since the warning sound can be controlled, while the combustion noise in [internal combustion engine vehicles] is always there.'

Try Our AI Features

Explore what Daily8 AI can do for you:

Learn with AIFact CheckingText to SpeechAI Summary

Related Articles

The Verge

37 minutes ago

• The Verge

Tesla's first robotaxi rides will have a ‘safety monitor' in the passenger seat

Tesla has begun sending out invitations for its highly anticipated robotaxi service, but there's one significant caveat: it's installing a 'safety monitor' in the front passenger seat, as previously reported by Electrek. The invites, which were sent to Tesla influencers and investors, say the human monitor will 'accompany you on your trip' when rides begin on June 22nd — a move that's at odds with Elon Musk's promise of fully unsupervised rides. Details about Tesla's robotaxi service have been slim in the weeks leading up to its launch, but Musk said in January that the company would launch its 'unsupervised' robotaxi service with 'no one in the car' this summer. Musk expanded on this in an April earnings call, saying the 10 to 20 Model Y vehicles would be remotely operated in the event of an emergency. Tesla's invitation outlines some requirements for robotaxi rides, including that riders must request service between 6AM and 12AM within a geofenced area, 'excluding airports.' It adds that 'service may be limited or unavailable in inclement weather,' which is often a challenge for autonomous vehicles. Invitees can bring one additional guest 18 or older. Over the past couple of months, Tesla has faced pressure from the National Highway Traffic Safety Administration, which has asked the EV-maker for more information about 'the ability of Tesla's system to react appropriately to reduced roadway visibility conditions.' Texas lawmakers have requested that Tesla delay its robotaxi launch until a revision to the state's autonomous driving law takes effect in September. It will require robotaxi services to get authorization from the Department of Motor Vehicles before operating without a human driver.

Android Authority

38 minutes ago

• Android Authority

T-Mobile may soon offer 36 month installment plans, versus the current 24-month default

Joe Maring / Android Authority TL;DR A new report claims T-Mobile is at least considering 36-month EIPs, versus its current two-month contract for new devices. Not only is there an internal report claiming such a program is the in the works, a few smarwatches briefly showed the new terms live on T-Mobile's website before they were reverted back. It's possible 36-month EIP program will only apply to select devices like watches and tablets, though it's also certainly possible the carrier has postponed or cancelled the plan altogether. It has been years since Verizon and AT&T ditched two-year device installment plans in favor of 36-month payment plans, leaving T-Mobile as the only major carrier that doesn't lock customers into a three-year commitment. I've stated many times in the past that it was only a matter of time before the Uncarrier followed suit. It seems that time is finally approaching — at least if a new leak from The Mobile Report proves accurate. The publication points to multiple signs suggesting 36-month installments are coming soon, including a document privately shared with them detailing that 36-month Equipment Installment Plans (EIPs) are in the pipeline for at least some devices. Additionally, a few Galaxy Watch devices briefly showed the new 36-month EIP length on T-Mobile's website, as seen in the screenshot below. Andrew Grush / Android Authority Let's be honest: the Galaxy Watch incident was certainly no mere error. While the company may have prematurely posted it or decided to delay the official announcement, the deliberate mention of 36-month terms strongly suggests legitimacy. The bigger question is whether this change will apply broadly to all devices and plans. It's quite possible these new 36-month plans might be exclusive to accessories like tablets and watches, while smartphone installment plans remain unchanged. Still, it would also make sense for T-Mobile to completely phase out two-year EIPs. This move would not only lock customers in longer but also align with T-Mobile's recent shifts toward resembling the very carriers it once criticized. For example, the carrier recently began rolling taxes and fees into all newer plans, including watches, tablets, and nearly everything except home internet service. T-Mobile has also incrementally raised prices and fees across its services. However, T-Mobile would be taking a risk by making this shift. The carrier currently offers its New in Two program, which would need significant adjustments to accommodate the 36-month structure, potentially restricting it to specific devices or plans. Such a change would likely face backlash, further undermining the carrier's credibility and giving the world just another example of how T-Mobile isn't afraid to break its customer promises. After all, the Uncarrier has already made several recent changes that are largely considered regressive by its former fanbase, and frankly, two-year device payment plans remain one of the stronger advantages T-Mobile has left beyond its strong 5G network and its (slowly declining) T-Mobile Tuesdays program. For now, the only (near) certainty is that T-Mobile is at least considering introducing 36-month EIPs. T-Mobile customers: Would the end of 36-month EIPs drive you away? 0 votes Yes NaN % No NaN % Unsure/Other (Tell us more in comments) NaN % T-Mobile's other recent moves already drove me away! NaN % As for timing, that's anyone's guess. The product page changes and the internal document suggest T-Mobile may have intended to implement this alongside other recent fee and tax adjustments, but decided either to postpone or phase it in later to avoid making a bad situation even worse. Of course, this is purely speculation on my part. I've reached out to T-Mobile for comment, but with rumors like this we typically don't get much of an answer. Got a tip? Talk to us! Email our staff at Email our staff at news@ . You can stay anonymous or get credit for the info, it's your choice.

Fast Company

39 minutes ago

• Fast Company

The AI era is here. Are you ready to lead?

Artificial intelligence (AI) is no longer a fringe experiment or a future-facing buzzword. It's a here-and-now force that's rapidly reshaping the modern business landscape. AI is already transforming how organizations operate—from how they close their books to how they engage with customers—and the pace is only accelerating. Are we ready for it? For too many leadership teams, the honest answer remains: not yet. AI's potential to improve back-office efficiency is well documented, particularly in finance functions that are often stretched thin. Intelligent automation can significantly reduce the time spent on tasks like transaction matching, reconciliations, and journal entries, freeing up teams to focus on higher-value activities like forecasting and strategic planning. Companies implementing intelligent process automation are seeing measurable impact in areas such as cash flow forecasting, anomaly detection, and real-time variance analysis. By drawing on clean, well-structured data, AI can surface patterns that eliminate risk and unlock opportunity, turning raw numbers into actionable insights. This evolution comes at a crucial time. Many finance departments are contending with a shrinking workforce and increasing expectations. Nearly 300,000 accountants and auditors have left the profession in the last two years, and the majority of CFOs now say they are responsible for enterprise-wide data and analytics. AI can help fill the gap if implemented thoughtfully. WHY RUSHING CAN LEAD TO FAILURE Obvious benefits aside, AI is not a silver bullet. Organizations that rush implementation without the right foundations risk doing more harm than good. Inaccurate data, poor governance, and a lack of human oversight can lead to flawed outputs, audit issues, and regulatory exposure. AI models trained on inconsistent or incomplete data can replicate human mistakes at scale, producing errors with more speed, not more accuracy. That's why clean, auditable data must be a non-negotiable starting point. Gartner predicted that 30% of generative AI projects will be abandoned after proof of concept by the end of 2025, in large part due to poor quality data. In other words: garbage in, garbage out. GOVERNANCE BEFORE GROWTH Before deploying AI at scale, leaders must build the proper guardrails. That starts with a robust data governance strategy that defines how data is collected, maintained, and used across the organization. Assigning data stewardship responsibilities, establishing quality standards, and ensuring alignment with regulatory frameworks are essential. Just as important is keeping human judgment in the loop. Especially in the Office of Finance, where compliance and auditability are paramount, AI tools should be explainable and traceable. The goal is not to replace decision makers, but to empower them with better information. Cross-functional collaboration between finance, IT, compliance, and legal teams is critical to achieving this balance. AI readiness is about people as much as it is about technology. Implementing AI effectively means helping employees evolve alongside it. That includes upskilling employees to become data-literate advisors, building comfort with new tools, and fostering a culture of experimentation balanced by accountability. Many of the skills teams already possess, such as data analysis, pattern recognition, and variance identification, map directly to AI-adjacent competencies. With the proper training and support, these professionals can become critical players in an AI-enabled function. LEADERS MUST GUIDE AI STRATEGY As AI becomes foundational to business success, it's no longer sufficient for executives to delegate its implementation to IT. CEOs and their leadership teams must play an active role in shaping AI strategy, from identifying high-impact use cases to defining clear KPIs. That includes assessing where AI can truly improve and where human judgment remains irreplaceable. It also means investing in the right tools—not necessarily the most complex or expensive ones, but those that simplify workflows and integrate well with existing systems. Ultimately, responsible AI leadership is about setting direction, removing friction, and creating a culture where teams feel both empowered and accountable. BUILD NOW OR SCRAMBLE LATER There's no doubt that AI will reshape how businesses operate. The only question is whether your organization will lead the way or struggle to catch up. A recent Deloitte survey of the financial services industry, an industry that typically is slow to adopt new technology, found that AI 'pioneers' or early adopters, see significantly more value from their AI deployments than 'followers' or those who are slow to bring the technology into their organization. Companies that act now by improving data quality, establishing governance, investing in their people, and piloting AI in high-impact areas will gain a durable competitive advantage. AI is not just a tool for transformation—it's a test of readiness. And in today's rapidly evolving landscape, readiness isn't optional—it's everything.