Connect with us

News

Scientists use AI neural network to translate speech from brain activity

Published

on

Three recently published studies focused on using artificial intelligence (AI) neural networks to generate audio output from brain signals have shown promising results, namely by producing identifiable sounds up to 80% of the time. Participants in the studies first had their brain signals measured while they were either reading aloud or listening to specific words. All the data was then given to a neural network to “learn” how to interpret brain signals after which the final sounds were reconstructed for listeners to identify. These results represent hopeful prospects for the field of brain-computer interfaces (BCIs), where thought-based communication is quickly moving from the realm of science fiction to reality.

The idea of connecting human brains to computers is far from new. In fact, several relevant milestones have been made in recent years including enabling paralyzed individuals to operate tablet computers with their brain waves. Elon Musk has also famously brought attention to the field with Neuralink, his BCI company that essentially hopes to merge human consciousness with the power of the Internet. As brain-computer interface technology expands and develops new ways to foster communication between brains and machines, studies like these, originally highlighted by Science Magazine, will continue demonstrating the steady march of progress.

Functional areas of the human brain. | Credit: Blausen.com staff (2014) via CC BY 3.0.

In the first study conducted by researchers from Columbia University and Hofstra Northwell School of Medicine, both in New York, five epileptic participants had the brain signals from their auditory cortexes recorded as they listened to stories and numbers being read to them. The signal data was provided to a neural network for analysis which then reconstructed audio files that were accurately identified by participating listeners 75% of the time.

In the second study conducted by a team from the University of Bremen (Germany), Maastricht University (Netherlands), Northwestern University (Illinois), and Virginia Commonwealth University (Virginia), brain signal data was gathered from six patients’ speech planning and motor areas while undergoing tumor surgeries. Each patient read specific words aloud to target the data collected. After the brain data and audio data were given to their neural network for training, the program was given brain signals not included in the training set to recreate audio, the result producing words that were recognizable 40% of the time.

Finally, in a third study by a team at the University of California, San Francisco, three participants with epilepsy read text aloud while brain activity was captured from the speech and motor areas of their brains. The audio generated from their neural network’s analysis of the signal readings was presented to a group of 166 people who were asked to identify the sentences from a multiple choice test – some sentences were identified with 80% accuracy.

While the research presented in these studies shows serious progress towards connecting human brains to computers, there are still a few significant hurdles. For one, the way neuron signal patterns in the brain translate into sounds varies from person to person, so neural networks must be trained on each individual person. The best results require the best data possible, i.e., the most precise neuron signals possible, meaning this is something that can only be obtained by placing electrodes in the brain itself. The opportunities to collect data at this invasive level for research are limited, relying on voluntary participation and approval of experiments.

All three of the studies highlighted demonstrated an ability to reconstruct speech based on neural data in some significant capacity; however, also in all cases, the study participants were able to create audible sounds to use with the computer training set. In the case of patients unable to speak, the level of difficultly in interpreting the brain’s speech signals from other signals will be the biggest challenge. Also, the differences between brain signals during actual speech vs. thinking about speech will complicate matters further.

Advertisement
-

Accidental computer geek, fascinated by most history and the multiplanetary future on its way. Quite keen on the democratization of space. | It's pronounced day-sha, but I answer to almost any variation thereof.

Advertisement
Comments

News

It’s official: SpaceX takes aim at Verizon, AT&T, and T-Mobile

SpaceX is buying 800 MHz spectrum from Grain to turn Starlink Mobile into a carrier.

Published

on

By

Starlink D2D direct to device vs Verizon, AT&T (Concept render by Grok)

SpaceX has agreed to buy a nationwide block of low band wireless spectrum, a deal the company says will let Starlink Mobile operate as a full US carrier rather than a satellite add-on for someone else’s network.

The company announced the agreement on X on Thursday afternoon, saying it will “pave the way for @Starlink to become a major mobile carrier in the US.” The seller is Grain Management, a private investment firm that confirmed in a statement that SpaceX will acquire 100% of its nationwide 800 MHz portfolio. That covers up to 14 MHz of paired spectrum in the 817 to 824 MHz and 862 to 869 MHz bands. Neither side disclosed a price, and the deal still needs FCC approval.

Grain only recently picked up the licenses itself. It bought the portfolio from T-Mobile in a transaction that closed in August, paying cash plus its own 600 MHz spectrum. Rival AST SpaceMobile had been testing satellites on the same bands before SpaceX stepped in.

Advertisement
-

SpaceX said its 2 GHz spectrum will handle high bandwidth capacity, while the new 800 MHz layer “ensures Starlink Mobile’s signal penetrates through obstacles, such as walls, and can provide service to customers’ devices even when they are in buildings.” The company added that most existing phones already support the band, so customers would not need new hardware to use it.

That 2 GHz spectrum came from SpaceX’s EchoStar acquisition last year, which gave the company exclusive S band rights in the US and global Mobile Satellite Service licenses. The Grain spectrum is different in an important way: it is tailored for service from ground towers, not satellites. SpaceX said that combination would make Starlink Mobile “the first network operator to deploy both satellite and terrestrial spectrum.”

The announcement also follows a key regulatory win. Earlier this week, the FCC approved SpaceX’s plan to deploy 15,000 second generation Starlink Mobile satellites, which the company has said will carry up to 100 times the data density of the current system, as Teslarati previously reported.

Shares of AT&T, Verizon and T-Mobile fell in extended trading after the announcement. T-Mobile is currently SpaceX’s launch partner for Starlink Mobile in the US, which makes its position the most complicated of the three.

SpaceX has not been subtle about its plans. During the company’s August earnings call, President and COO Gwynne Shotwell said she expected Starlink Mobile to win over customers from the major carriers. “I anticipate us to be able to acquire quite a few of their customers because I think our service will be better,” she said, pointing to dead zone coverage and resilience during disasters. Shotwell also described plans for low cost cellular base stations that could pair with existing Starlink dishes.

SpaceX has targeted 2027 for deployment of its next generation Starlink Mobile satellites, with upgraded service expected by the end of that year. The FCC review of the Grain deal now determines when the terrestrial half of that network can come online.

Advertisement
-
Continue Reading

Elon Musk

Trump presents Elon Musk with National Medal of Science, calls him “modern-day Thomas Edison”

Trump gave Elon Musk the National Medal of Science and compared him to Thomas Edison.

Published

on

By

Elon Musk holds the National Medal of Science alongside President Donald Trump at the “Science: A New Golden Age” summit in Washington on October 8, 2026.
Elon Musk holds the National Medal of Science alongside President Donald Trump at the “Science: A New Golden Age” summit in Washington on October 8, 2026.

Elon Musk now has the highest honor the U.S. government gives to scientists and engineers. President Donald Trump presented him with the National Medal of Science on Thursday at the White House’s “Science: A New Golden Age” summit, held at the Institute of Peace in Washington, according to the White House.

Trump called Musk a “national treasure” and “our modern-day Thomas Edison.” “Few Americans in history have done more to advance America’s national interest than Elon in one field after another,” the president said. He also described Musk as an “industrial titan, brilliant engineer and one of the greatest technology founders to ever live.”

The official citation credits Musk “for engineering achievements of extraordinary ambition and scale that have pushed the frontiers of space exploration, intelligent systems, communications, and beyond, renewing American technological leadership and opening a new chapter in American manufacturing.” Tesla is not named in it, but “intelligent systems” and “American manufacturing” cover a lot of ground that Tesla and the Terafab chip project occupy.

Google co-founder Sergey Brin, Nvidia CEO Jensen Huang and AMD CEO Lisa Su received the same medal. Microsoft CEO Satya Nadella and Dell Technologies CEO Michael Dell were given the National Medal of Technology and Innovation. It was the first time Trump has handed out either award in his two terms. As Teslarati noted when the list was first reported on Wednesday, the National Science Foundation says 529 people have received the Medal of Science since Congress created it in 1959.


The ceremony closes a loop that opened in 2025, when Musk left DOGE and criticized the “Big Beautiful Bill,” sparking a public feud. Al Jazeera noted that Musk was the largest contributor to Trump’s 2024 campaign, and CNBC reported that all of the honorees have given financially to Trump or Republican causes in some form. The reconciliation has moved quickly in recent weeks. Musk sat at Trump’s left at the September 29 White House AI lunch, was named to help lead the Pentagon’s Project Meridian study on the future of warfare, and adopted the administration’s “super intelligence” wording when he said SpaceXAI will become SpaceXSI.

For Tesla and SpaceX shareholders, the practical question is what the closer relationship means for federal approvals. Tesla’s Robotaxi and Cybercab expansion depends on national autonomy rules, and SpaceX’s Starship and Starlink plans run through the FAA, FCC and NASA. Thursday’s ceremony did not change any of that directly.

Continue Reading

Elon Musk

Elon Musk’s Boring Company lands a new Middle East deal, and Nashville is about to get faster

The Boring Company signs Abu Dhabi tunnel agreement while adding more Prufrock machines in Nashville.

Published

on

By

the-boring-company-tesla-robotaxi

The Boring Company has signed an agreement with Abu Dhabi to study underground transport and utility tunnels across the emirate, adding a second UAE city to its pipeline as it prepares to also scale up tunneling back home in Nashville.

The deal was signed Thursday at the Liveability and Investment Exhibition (LIVEX 2026) by Boring Company President Steve Davis and Maysarah Mahmoud Salim Eid, director general of the Abu Dhabi Projects and Infrastructure Centre (ADPIC), according to the Abu Dhabi Media Office. Mohamed Ali Al Shorafa, chairman of the emirate’s Department of Municipalities and Transport, attended the signing.

Under the agreement, the two sides will assess feasibility, delivery and operating models for tunnels that could carry passengers or utilities. They will also look at Abu Dhabi’s potential as a regional hub for tunneling work. The current phase is exploratory, and no construction commitment or project budget has been announced.

“Abu Dhabi provides an ideal environment to explore the next generation of underground infrastructure solutions, supported by its ambitious growth vision and strong commitment to advanced technologies,”

Davis said. He added that the company wants to assess how tunnels can “expand urban capacity more efficiently, and enable better use of available space.”

The timing lines up with the money, considering last month, The Boring Company closed a $3 billion Series D led by the UAE and affiliated investors, valuing the company at $23 billion, as Teslarati reported. That round came with a commitment to build more than 150 kilometers of tunnel across the UAE, separate from the Dubai Loop pilot already under contract with Dubai’s Roads and Transport Authority. That pilot covers 6.4 kilometers and four stations linking DIFC and Dubai Mall at a cost of about $154 million.

Back home, The Boring Company projects in Nashville are also scaling up, with the company telling local NewsChannel 5 that a third Prufrock machine could start digging the Music City Loop in late October. A fourth is also targeted before the end of the year. Two machines are already mining Nashville limestone at the same time, and work is underway on a new launch site for the third.

Advertisement
-

The company said it has made more than 300 design and performance upgrades to its original Nashville machine. It is also working with property owners on more than 40 planned stations, with approvals in place for a future Nashville International Airport connection, a downtown station near the Music City Center, and stops at residential towers and the JW Marriott.

Construction on the Music City Loop began the same evening Tennessee and federal regulators approved the project’s lease in February, and the company targeted its first operational segment for late 2026. Back in Las Vegas, The Boring Company has said it plans to double its Vegas Loop station count by year’s end.

Continue Reading