Researchers made a sonar-equipped earphone that can capture facial expressions

Researchers at Cornell University have developed an earphone that uses sonar to detect the wearer’s facial expression to create an avatar of their face. The so-called “earable” system is called EarIO.

It works by bouncing sound off the wearer’s cheeks — the audio is emitted from speakers on each side of the earphone. A microphone captures the echoes, which change as the face moves and the wearer speaks. The system then uses a deep learning algorithm to turn the echoes into a replica of the person’s expression. EarIO can transmit the facial movements to a mobile device in real time and the avatar can be used in video calls.

Camera-based devices that track face movements are “large, heavy and energy-hungry, which is a big issue for wearables,” said Cheng Zhang, principal investigator of the Smart Computer Interfaces for Future Interactions Lab, who co-authored a paper on EarIO. “Also importantly, they capture a lot of private information.” A sonar-based approach can bolster privacy, affordability, comfort and battery life, he said.

In initial testing, the team found the device works while wearers are sitting and walking, and factors like background chatter, wind and ambient road noise don’t impact the acoustic signaling. However, the high sensitivity of the sensing method can cause some issues. “It’s good, because it’s able to track very subtle movements, but it’s also bad because when something changes in the environment, or when your head moves slightly, we also capture that,” said co-author Ruidong Zhang, an information science doctoral student. The researchers hope to mitigate such disruptions in future models.

EarIO has some limitations as things stand. The device runs for around three hours on a single charge despite being far more energy efficient than a camera-based system the team previously used. The researchers hope to improve the battery life in the future. They also aim to make EarIO a plug-and-play device but it currently needs 32 minutes of facial data training before the first use.

Judge grants Twitter expedited trial against Elon Musk

Twitter has scored a victory in its lawsuit against Elon Musk over his attempted exit from a $44 billion takeover deal. In an initial hearing, Delaware Court of Chancery chancellor Kathaleen McCormick has granted Twitter’s request for an expedited, five-day trial beginning in October. The company originally sought a four-day trial in September as part of its effort to make Musk “honor his obligations.”

Musk’s lawyers asked the court to delay the trial to February 2023. They claimed Twitter was in a sudden rush to complete a trial and force the acquisition follow two months of alleged “foot-dragging and obfuscation” over requested data relating to bots and other fake accounts. Musk accused Twitter of underreporting the volume of bogus accounts (and failing to provide enough information) in a bid to clinch the deal. As part of the call for a delay, Musk’s lawyers also claimed they needed more time to search Twitter’s raw “firehose” data.

While Twitter didn’t get exactly what it wanted, the ruling is a clear victory for the social network. It won’t have to wait long to obtain a decision, and may force Musk to rely primarily on the claims he made when he announced plans to back out of the purchase. There’s a chance Musk may have to continue with the buyout before the year is over, no matter how much he might regret making the offer.

Synchron says it’s the first to implant a human brain-computer interface in the US

Brain-computer interfaces have become a practical (if limited) reality in the US. Synchron says it has become the first in the country to implant a BCI in a human patient. Doctors in New York’s Mount Sinai West implanted the company’s Stentrode in the motor cortex of a participant in Synchron’s COMMAND trial, which aims to gauge the usefulness and safety of BCIs for providing hands-free device control to people with severe paralysis. Ideally, technology like Stentrode will offer independence to people who want to email, text and otherwise handle digital tasks that others take for granted.

Surgeons installed the implant using an endovascular procedure that avoids the intrusiveness of open-brain surgery by going through the jugular vein. The operation went “extremely well” and let the patient return home 48 hours later, according to Synchron. An ongoing Australian trial has also proven successful so far, with four patients still safe a year after receiving their implants.

It may take a long time before doctors can offer Synchron’s BCIs to patients. The company received FDA approval for human trials in July 2021, and it’s still expanding the COMMAND trial as of this writing. Still, the US procedure represents a significant step toward greater autonomy for people with paralysis. It also represents a competitive victory — Elon Musk’s Neuralink has yet to receive FDA permission for its own implant.

Google allows Android apps to use third-party payments in the EU

Android developers who distribute apps on the Google Play store can now use third-party payment systems in many European countries. The measure applies to the European Economic Area (EEA), which comprises European Union states as well as Iceland, Liechtenstein and Norway. However, the policy will not apply to gaming apps, which still need to use Google Play’s own billing system for the time being.

Google is making the move after the EU’s legislative arm, the European Commission, passed the Digital Markets Act (DMA) this month. Along with the Digital Services Act, the law is designed to rein in the power of big tech by, for instance, prohibiting major platform holders from giving their own systems preferable treatment.

The DMA isn’t expected to come into effect until sometime in 2024. However, Google’s director of EU government affairs and public policy, Estelle Werth, wrote in a blog post that the company is “launching this program now to allow us to work closely with our developer partners and ensure our compliance plans serve the needs of our shared users and the broader ecosystem.”

The move partially reverses a policy that required all in-app payments to be processed through the Play Store’s billing system. Developers who opt for a different billing system won’t be able to avoid Google’s fees entirely. However, Google will lower the service fees it charges them by three percent.

Google says that 99 percent of developers qualify for a fee of 15 percent or less. The others typically pay 30 percent. The fees Google charges would drop to 12 percent (or lower) or 27 percent, respectively, if they select a third-party billing system. Developers may end up paying around the same amount in fees anyway, depending on what the alternative payment provider charges.

Developers will also “need to meet appropriate user protection requirements,” Google says. The company plans to allow gaming apps to use alternative payment systems in the EEA sometime before the DMA comes into effect.

Google has long resisted the idea of opening up the Play store to alternative payment providers in the face of pressure from multiple sides, but the EU forced its hand in this case. The company adopted a similar measure in South Korea last year after the country passed a law requiring Google and Apple to offer third-party payments.

Apple settles lawsuit over its reviled ‘butterfly’ keyboard for $50 million

Apple could soon compensate MacBook owners for their troubles with faulty “butterfly” keyboards. Reutersreports Apple has agreed to pay $50 million to settle a class-action lawsuit alleging that it knew about and concealed the unreliable designs of keyboards on MacBook, MacBook Air and MacBook Pro models released between 2015 and 2019. If a judge approves the preliminary deal, Apple would pay customers who needed repairs in California, Florida, Illinois, Michigan, New Jersey, New York and Washington.

The company won’t have to admit wrongdoing as part of the settlement. It will have to continue offering free keyboard repairs for four years after purchase.

Apple introduced the butterfly keyboard in 2015 with the 12-inch MacBook. It was meant to enable ever-slimmer laptops without compromising stability, but the design quickly developed a reputation for extreme sensitivity to debris. Keys would get stuck or lose responsiveness if even tiny dust motes or crumbs slipped underneath. The company took steps to mitigate the problem (such as membranes) and eventually began reverting to more conventional keyboards starting with the 16-inch MacBook Pro from late 2019. Apple acknowledged that some were having problems and launched repair programs, but maintained that the majority of customers had no issues.

As with many class-actions, you shouldn’t expect a windfall if you’re affected. Attorneys said they expected a $395 payout if you’ve had to replace multiple keyboards, $125 for one full replacement and $50 if you only replaced key caps. The lawyers may also claim up to $15 million of the $50 million settlement in legal fees, which could limit the money available for MacBook owners. While the payouts aren’t absolutely necessary when Apple has previously offered refunds for repairs, they’re more likely to be symbolic than practical.

Transit app Moovit rolls out more personalized trip-planning features

Transit app Moovit is aiming to be more helpful when it comes to helping users get to their destination. Starting today, the app is rolling out more personalized trip-planning features in 3,500 cities across 112 countries to build on its existing route…

Amazon sues operators of 10,000 Facebook groups over fake review schemes

Amazon’s long-running campaign against fake reviews just enjoyed some success. TechCrunchreports that Amazon has sued the administrators of more than 10,000 Facebook groups for attempting to coordinate fake reviews in exchange for free products or money. The group operators tried to recruit people in the US, UK, four European countries and Japan. One of the larger examples, Amazon Product Review, tried to avoid detection by “obfuscating” letters in phrases used to detect activity like this. Meta pulled the 43,000-member earlier this year.

The legal action comes after roughly two years of pinpointing rogue Facebook groups. Amazon said it would use info obtained during the discovery process to identify the perpetrators and pull fake reviews stemming from these communities.

Amazon has been suing fake review peddlers since 2015, and ramped up its efforts in recent years. The online shopping giant reported over 1,000 paid review groups to social networks in the first quarter of 2021, or three times the volume it reported a year earlier. The company argued that reports and lawsuits weren’t enough to stop fake reviews, however. It believed that a more permanent solution would involve a stronger “public-private partnership” between online shops, social networks and law enforcement.

Whether or not Amazon can meaningfully deter fake reviews is another matter. This latest bust underscores the scope of the problem — it’s easy to create a social media group that goes undetected for long enough to do significant damage. UK regulators certainly aren’t convinced Amazon is doing enough, and have opened an investigation into the firm’s approach to bogus review content. Amazon is fighting a battle that might be difficult to win under ideal circumstances, and it’s not clear if the company’s existing strategy is effective.