Loading video...

Video Failed to Load

Go Home

Brain-Computer Interfaces for speech just overcame their largest challenge — latency. Researchers from UC Berkeley and UC San Francisco have developed a new streaming approach using AI models that allows near real-time speech, a massive breakthrough from the previous 8-second delay for a single sentence. Read ➡️

NVIDIAAIDev's profile picture

NVIDIA AI Developer

74,183 subscribers

18,952 views • 1 year ago •via X (Twitter)

Science & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Why does UC Berkeley UC Berkeley allow a professor to show up to teach dressed as furry? Extremely bizarre and inappropriate.
0:10

Why does UC Berkeley UC Berkeley allow a professor to show up to teach dressed as furry? Extremely bizarre and inappropriate.

Libs of TikTok

3,164,707 views • 2 years ago

NEURALINK: REAL-TIME BRAIN ACTIVITY, STREAMED Neuralink's implanted neural device is already delivering what scientists have chased for decades: live visibility into human brain activity within minutes of surgery. In this demo, signals flare as a participant wakes up post-procedure - the device capturing the gradual surge of neural activity in real time. It’s a glimpse of the frontier: hardware that doesn’t just record the brain, but streams its inner rhythms instantly. From clinical care to brain-computer interfaces, the implications are massive. Source: Neuralink
0:13

NEURALINK: REAL-TIME BRAIN ACTIVITY, STREAMED Neuralink's implanted neural device is already delivering what scientists have chased for decades: live visibility into human brain activity within minutes of surgery. In this demo, signals flare as a participant wakes up post-procedure - the device capturing the gradual surge of neural activity in real time. It’s a glimpse of the frontier: hardware that doesn’t just record the brain, but streams its inner rhythms instantly. From clinical care to brain-computer interfaces, the implications are massive. Source: Neuralink

Mario Nawfal

204,494 views • 9 months ago

KATE'S SPEECH. 😂🤣🤣🤣. The BM, like the Mirror, are calling that 8-second mumble note shit below a speech.🤣🤣🤣 The fact that they call that 8-second mumble shit a speech just shows you how desperate they are for Kate to possess the qualities that great speaker Meghan has.
0:14

KATE'S SPEECH. 😂🤣🤣🤣. The BM, like the Mirror, are calling that 8-second mumble note shit below a speech.🤣🤣🤣 The fact that they call that 8-second mumble shit a speech just shows you how desperate they are for Kate to possess the qualities that great speaker Meghan has.

THE SUSSEX SOLDIER 🪖

26,907 views • 1 year ago

MIT researchers taught a robot dog to perceive a 3D world using Neural Volumetric Memory (NVM). This technique lets the bot climb stairs, step over gaps & run autonomously w/a single neural network. Paper: Video: w/UC San Diego
0:12

MIT researchers taught a robot dog to perceive a 3D world using Neural Volumetric Memory (NVM). This technique lets the bot climb stairs, step over gaps & run autonomously w/a single neural network. Paper: Video: w/UC San Diego

MIT CSAIL

507,583 views • 3 years ago

JUST IN: Meta AI introduces Voicebox, an all-in-one generative speech model. Voicebox is an impressive breakthrough! It could do for speech what other models like GPT-3 and Stable Diffusion have done for text and images. Some key details: - Voicebox can synthesize speech across 6 languages - It's a general-purpose model that can perform tasks it wasn't trained on. It can perform noise removal, content editing, style conversion, and more - Supports in-context text-to-speech synthesis and cross-lingual style transfer - It's 20x faster than current models and outperforms single-purpose models through in-context learning paper: blog:
0:13

JUST IN: Meta AI introduces Voicebox, an all-in-one generative speech model. Voicebox is an impressive breakthrough! It could do for speech what other models like GPT-3 and Stable Diffusion have done for text and images. Some key details: - Voicebox can synthesize speech across 6 languages - It's a general-purpose model that can perform tasks it wasn't trained on. It can perform noise removal, content editing, style conversion, and more - Supports in-context text-to-speech synthesis and cross-lingual style transfer - It's 20x faster than current models and outperforms single-purpose models through in-context learning paper: blog:

elvis

88,495 views • 3 years ago

Roch Cholowsky almost hit his 1st HR of the season. UC-San Diego didn't have anyone covering second base, so he ends up with a double. 1-for-4 with a double and walk
0:16

Roch Cholowsky almost hit his 1st HR of the season. UC-San Diego didn't have anyone covering second base, so he ends up with a double. 1-for-4 with a double and walk

Josh Nelson - Sox Machine

142,632 views • 3 months ago

Someone just built a AI agent that connects to blender and builds full 3d models from a single image.
0:18

Someone just built a AI agent that connects to blender and builds full 3d models from a single image.

Oliver Prompts

53,398 views • 1 month ago

NVIDIA AI Released DiffusionRenderer: An AI Model for Editable, Photorealistic 3D Scenes from a Single Video In a groundbreaking new paper, researchers at NVIDIA, University of Toronto, Vector Institute and the University of Illinois Urbana-Champaign have unveiled a framework that directly tackles this challenge. DiffusionRenderer represents a revolutionary leap forward, moving beyond mere generation to offer a unified solution for understanding and manipulating 3D scenes from a single video. It effectively bridges the gap between generation and editing, unlocking the true creative potential of AI-driven content. DiffusionRenderer treats the “what” (the scene’s properties) and the “how” (the rendering) in one unified framework built on the same powerful video diffusion architecture that underpins models like Stable Video Diffusion..... Read full article here: Paper: GitHub Page: NVIDIA NVIDIA AI NVIDIAnewsroom NVIDIA AIDev
0:10

NVIDIA AI Released DiffusionRenderer: An AI Model for Editable, Photorealistic 3D Scenes from a Single Video In a groundbreaking new paper, researchers at NVIDIA, University of Toronto, Vector Institute and the University of Illinois Urbana-Champaign have unveiled a framework that directly tackles this challenge. DiffusionRenderer represents a revolutionary leap forward, moving beyond mere generation to offer a unified solution for understanding and manipulating 3D scenes from a single video. It effectively bridges the gap between generation and editing, unlocking the true creative potential of AI-driven content. DiffusionRenderer treats the “what” (the scene’s properties) and the “how” (the rendering) in one unified framework built on the same powerful video diffusion architecture that underpins models like Stable Video Diffusion..... Read full article here: Paper: GitHub Page: NVIDIA NVIDIA AI NVIDIAnewsroom NVIDIA AIDev

Marktechpost AI Dev News ⚡

104,741 views • 11 months ago

DIVIDED DISPLAY: Contrasting scenes from UC Berkeley show a split response to TPUSA’s campus event — fires and arrests outside as agitators clashed with police, while inside, supporters stood and cheered for Charlie Kirk.
0:14

DIVIDED DISPLAY: Contrasting scenes from UC Berkeley show a split response to TPUSA’s campus event — fires and arrests outside as agitators clashed with police, while inside, supporters stood and cheered for Charlie Kirk.

Fox News

71,106 views • 6 months ago

This AI is a storyteller. Everyone is hyping other models, but Higgsfield just made a killer audio tool. Their new Speak 2.0 feature, powered by AI, is the real deal for K-pop. This is a massive opportunity.
0:10

This AI is a storyteller. Everyone is hyping other models, but Higgsfield just made a killer audio tool. Their new Speak 2.0 feature, powered by AI, is the real deal for K-pop. This is a massive opportunity.

AGAFE

60,957 views • 9 months ago

The Breakthrough!! The FDA has approved the first blood test for Alzheimer's diagnosis. (Approval since December2025) This changes everything!!!!!!!! Alzheimer's diagnosis just went from a $5,000+ brain scan to a simple $200 blood draw. Here's why this is a MASSIVE breakthrough 👇
0:13

The Breakthrough!! The FDA has approved the first blood test for Alzheimer's diagnosis. (Approval since December2025) This changes everything!!!!!!!! Alzheimer's diagnosis just went from a $5,000+ brain scan to a simple $200 blood draw. Here's why this is a MASSIVE breakthrough 👇

Harshi Peiris, Ph.D.

28,735 views • 3 months ago

UC SAN DIEGO IS THE BIG WEST CHAMPION 🤩 The Tritons punch their ticket to the NCAA Tournament for the second straight year 🎟️
0:14

UC SAN DIEGO IS THE BIG WEST CHAMPION 🤩 The Tritons punch their ticket to the NCAA Tournament for the second straight year 🎟️

espnW

10,807 views • 2 months ago

Heading to UC Berkeley to debate abortion. Please pray for me and everyone attending. ❤️
0:14

Heading to UC Berkeley to debate abortion. Please pray for me and everyone attending. ❤️

Lila Rose

60,449 views • 1 month ago

Two 19-year-old students from the University of Washington, Thomas Pryor and Navid Azodi have developed a pair of gloves that convert sign language to speech or text.
0:13

Two 19-year-old students from the University of Washington, Thomas Pryor and Navid Azodi have developed a pair of gloves that convert sign language to speech or text.

Science girl

26,686 views • 7 months ago

A man with impaired speech from amyotrophic lateral sclerosis (ALS) uses the speech neuroprosthesis to say whatever he wants for the first time. View more videos showing speech neuroprosthesis in practice:
0:13

A man with impaired speech from amyotrophic lateral sclerosis (ALS) uses the speech neuroprosthesis to say whatever he wants for the first time. View more videos showing speech neuroprosthesis in practice:

NEJM

25,179 views • 1 year ago

This is a hostile takeover. Everyone is hyping other models, but Higgsfield just made a killer audio tool. Their new Speak 2.0 feature, powered by AI, is the real deal for K-pop. This is a massive opportunity.
0:10

This is a hostile takeover. Everyone is hyping other models, but Higgsfield just made a killer audio tool. Their new Speak 2.0 feature, powered by AI, is the real deal for K-pop. This is a massive opportunity.

Fakhr

19,785 views • 9 months ago

🚨 BREAKING: A startup just raised $55M to plug directly into the human brain and the real breakthrough isn’t the chip. It’s the material. What’s new Axoft built an implantable brain–computer interface using a bio-inspired material called Fleuron that is: • ~10,000× softer than current implants • Causes far less scarring • Captures higher quality brain signals for longer And it’s already been tested in humans. Why this matters Current brain implants fail over time because the brain rejects them. This changes that. The interface now behaves more like brain tissue than hardware Which means: • More stable neural data • Access to deeper brain regions • Longer-lasting implants The real shift This isn’t just better hardware. It’s a new direction: Design the interface to match biology — not fight it What this unlocks • Real-time brain signal decoding • New treatments for neurological disorders • Direct training data for next-gen AI systems Think about this If we can reliably read the brain… We’re no longer guessing what the brain is doing. We’re measuring it directly. Follow me I track where biology and technology stop being separate systems.
0:10

🚨 BREAKING: A startup just raised $55M to plug directly into the human brain and the real breakthrough isn’t the chip. It’s the material. What’s new Axoft built an implantable brain–computer interface using a bio-inspired material called Fleuron that is: • ~10,000× softer than current implants • Causes far less scarring • Captures higher quality brain signals for longer And it’s already been tested in humans. Why this matters Current brain implants fail over time because the brain rejects them. This changes that. The interface now behaves more like brain tissue than hardware Which means: • More stable neural data • Access to deeper brain regions • Longer-lasting implants The real shift This isn’t just better hardware. It’s a new direction: Design the interface to match biology — not fight it What this unlocks • Real-time brain signal decoding • New treatments for neurological disorders • Direct training data for next-gen AI systems Think about this If we can reliably read the brain… We’re no longer guessing what the brain is doing. We’re measuring it directly. Follow me I track where biology and technology stop being separate systems.

TheNewPhysics

42,362 views • 1 month ago

📹 Just in from #Gastech: Our CEO, Mike Wirth, delivered a keynote speech highlighting the need for a balanced approach to energy policy and underscored natural gas as a key enabler of a lower carbon future. Keep watching here:
0:12

📹 Just in from #Gastech: Our CEO, Mike Wirth, delivered a keynote speech highlighting the need for a balanced approach to energy policy and underscored natural gas as a key enabler of a lower carbon future. Keep watching here:

Chevron

1,255,905 views • 1 year ago

Meanwhile at UC Berkeley…
0:12

Meanwhile at UC Berkeley…

Libs of TikTok

273,518 views • 1 year ago