Key Takeaways
Neuralink demonstrated a Grok AI integration that allows a participant to trigger an AI agent through imagined speech and have it perform computer tasks.
A new neural voice activity detection system automatically starts and stops speech decoding, reducing the need for physical controls.
Neuralink also tested phone-call integration, brain-controlled iPhone dictation, and faster editing tools to make everyday communication easier.
Neuralink has demonstrated new features designed to help people with severe speech impairments make phone calls, control a Grok AI agent through imagined speech, and communicate without manually activating a speech decoder.
The Elon Musk-founded company also tested brain-controlled dictation across iPhone apps, allowing users to enter text without relying on conventional keyboards.
The developments are part of Neuralink’s experimental brain-computer interface (BCI) technology, which translates neural signals associated with intended speech into words.
The company demonstrated the features with a participant in its VOICE clinical study, which focuses on restoring communication for people who have lost the ability to speak.
Brain signals control AI
One of the new features allows users to activate a Grok AI agent by imagining themselves saying “Hey Neuralink,” followed by a command.
The agent then carries out the requested task autonomously on a computer. Neuralink described the integration as an early demonstration of how brain-computer interfaces could enable more direct interaction with AI-powered systems.
The company also developed a neural voice activity detection system that automatically recognizes when a participant intends to speak.
Previously, users had to press a button to start and stop recording neural activity for speech decoding. The new system detects when the user is attempting to speak and automatically starts or stops the decoding process.
More from Innovation
See AllInnovationResearchers use AI to rediscover forgotten materials from the 1970s, can boost next-gen green techInnovation97% of silver recovered from dead solar cells using acid-free molten salt bathInnovationMassive 442-foot tunneling machine building world’s longest undersea railway tunnelInnovationMIT uses tiny implanted magnets to improve control of robotic prosthetic limbsInnovationUS firm unveils Lego-like modular kits to build tailored AI, robotics, and drone hubsThis could reduce the need for physical controls and make communication more accessible to people with severe movement limitations.
Neuralink also introduced real-time visualizations that display the intensity and distribution of neural signals recorded by the implant.
These visualizations allow participants to experiment with different imagined speaking patterns and identify those that produce stronger neural activity, potentially helping improve speech-decoding performance.
Making everyday communication easier
Another development integrates BCI-generated speech directly into iPhone calls.
Instead of relying on a separate communication interface, the system feeds synthesized speech into the phone’s microphone audio stream, allowing the person on the other end to hear the generated voice.
During testing, the participant used the technology to speak with his wife over the phone. Neuralink said the couple had begun their relationship through phone conversations, making the demonstration particularly meaningful.
The company also developed an iOS keyboard integration that allows users to dictate text through neural signals directly into different applications.
The feature is intended to reduce the difficulties people with conditions such asamyotrophic lateral sclerosis(ALS) may face when using conventional keyboards or voice dictation. Neuralink also tested improvements to Bloom, its mobile application for speech-based brain-computer interfaces.
These include a faster editing system that allows users to correct incorrectly decoded text and a lock-screen feature that enables speech communication without first unlocking the phone.
The improvements were developed during a 72-hour engineering sprint at a VOICE study participant’s home, where engineers worked alongside the participant and his wife.
The team built and testedprototypesin real time, using feedback from the participant to identify problems and make changes. Some features underwent further development in the weeks following the sprint.
The demonstrations highlight Neuralink’s efforts to move brain-controlled communication beyond laboratory settings and into everyday activities.
However, the company has not provided quantitative performance results for the new features, including speech-decoding accuracy or response times.
Neuralink’sbrainimplants remain investigational and have not received approval from the US Food and Drug Administration or other regulatory authorities.
Get the latest in engineering, tech, space & science - delivered daily to your inbox.
By subscribing, you agree to our
Terms of Useand
PoliciesYou may unsubscribe at any time.
0
COMMENT
Subscribe to
Today!
Access to exclusive content, expert insights and a deeper dive into engineering and tech. No ads, no limits.
Explore Now!By
Neetika WalterWith over a decade-long career in journalism, Neetika Walter has worked with The Economic Times, ANI, and Hindustan Times, covering politics, business, technology, and the clean energy sector. Passionate about contemporary culture, books, poetry, and storytelling, she brings depth and insight to her writing. When she isn’t chasing stories, she’s likely lost in a book or enjoying the company of her dogs.
- 115 MW microreactor deployment in US advances as Hyundai partners with NANO Nuclear
- 2Solar planes at 60,000 feet could fly for months with 500 Wh/kg batteries: US study
- 324,000-ton nuclear-powered submarine fires 5,160-mile-range missile while submerged
- 46,000-ton frigate that can launch missiles, change face of naval warfare offered to US Navy
- 5Fire at US nuclear site triggers 'radiological release' alert, FAA restricts flights
Key Takeaways
Neuralink demonstrated a Grok AI integration that allows a participant to trigger an AI agent through imagined speech and have it perform computer tasks.
A new neural voice activity detection system automatically starts and stops speech decoding, reducing the need for physical controls.
Neuralink also tested phone-call integration, brain-controlled iPhone dictation, and faster editing tools to make everyday communication easier.
Neuralink has demonstrated new features designed to help people with severe speech impairments make phone calls, control a Grok AI agent through imagined speech, and communicate without manually activating a speech decoder.
The Elon Musk-founded company also tested brain-controlled dictation across iPhone apps, allowing users to enter text without relying on conventional keyboards.
The developments are part of Neuralink’s experimental brain-computer interface (BCI) technology, which translates neural signals associated with intended speech into words.
The company demonstrated the features with a participant in its VOICE clinical study, which focuses on restoring communication for people who have lost the ability to speak.
Brain signals control AI
One of the new features allows users to activate a Grok AI agent by imagining themselves saying “Hey Neuralink,” followed by a command.
The agent then carries out the requested task autonomously on a computer. Neuralink described the integration as an early demonstration of how brain-computer interfaces could enable more direct interaction with AI-powered systems.
The company also developed a neural voice activity detection system that automatically recognizes when a participant intends to speak.
Previously, users had to press a button to start and stop recording neural activity for speech decoding. The new system detects when the user is attempting to speak and automatically starts or stops the decoding process.
More from Innovation
See AllInnovationResearchers use AI to rediscover forgotten materials from the 1970s, can boost next-gen green techInnovation97% of silver recovered from dead solar cells using acid-free molten salt bathInnovationMassive 442-foot tunneling machine building world’s longest undersea railway tunnelInnovationMIT uses tiny implanted magnets to improve control of robotic prosthetic limbsInnovationUS firm unveils Lego-like modular kits to build tailored AI, robotics, and drone hubsThis could reduce the need for physical controls and make communication more accessible to people with severe movement limitations.
Neuralink also introduced real-time visualizations that display the intensity and distribution of neural signals recorded by the implant.
These visualizations allow participants to experiment with different imagined speaking patterns and identify those that produce stronger neural activity, potentially helping improve speech-decoding performance.
Making everyday communication easier
Another development integrates BCI-generated speech directly into iPhone calls.
Instead of relying on a separate communication interface, the system feeds synthesized speech into the phone’s microphone audio stream, allowing the person on the other end to hear the generated voice.
During testing, the participant used the technology to speak with his wife over the phone. Neuralink said the couple had begun their relationship through phone conversations, making the demonstration particularly meaningful.
The company also developed an iOS keyboard integration that allows users to dictate text through neural signals directly into different applications.
The feature is intended to reduce the difficulties people with conditions such asamyotrophic lateral sclerosis(ALS) may face when using conventional keyboards or voice dictation. Neuralink also tested improvements to Bloom, its mobile application for speech-based brain-computer interfaces.
These include a faster editing system that allows users to correct incorrectly decoded text and a lock-screen feature that enables speech communication without first unlocking the phone.
The improvements were developed during a 72-hour engineering sprint at a VOICE study participant’s home, where engineers worked alongside the participant and his wife.
The team built and testedprototypesin real time, using feedback from the participant to identify problems and make changes. Some features underwent further development in the weeks following the sprint.
The demonstrations highlight Neuralink’s efforts to move brain-controlled communication beyond laboratory settings and into everyday activities.
However, the company has not provided quantitative performance results for the new features, including speech-decoding accuracy or response times.
Neuralink’sbrainimplants remain investigational and have not received approval from the US Food and Drug Administration or other regulatory authorities.
Get the latest in engineering, tech, space & science - delivered daily to your inbox.
By subscribing, you agree to our
Terms of Useand
PoliciesYou may unsubscribe at any time.
0
COMMENT
Subscribe to
Today!
Access to exclusive content, expert insights and a deeper dive into engineering and tech. No ads, no limits.
Explore Now!By
Neetika WalterWith over a decade-long career in journalism, Neetika Walter has worked with The Economic Times, ANI, and Hindustan Times, covering politics, business, technology, and the clean energy sector. Passionate about contemporary culture, books, poetry, and storytelling, she brings depth and insight to her writing. When she isn’t chasing stories, she’s likely lost in a book or enjoying the company of her dogs.
- 115 MW microreactor deployment in US advances as Hyundai partners with NANO Nuclear
- 2Solar planes at 60,000 feet could fly for months with 500 Wh/kg batteries: US study
- 324,000-ton nuclear-powered submarine fires 5,160-mile-range missile while submerged
- 46,000-ton frigate that can launch missiles, change face of naval warfare offered to US Navy
- 5Fire at US nuclear site triggers 'radiological release' alert, FAA restricts flights
