Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Avatar Forcing: Real-time interactive head avatars for natural conversation, and it's not WAN. - handles both speaking & active listening + multimodal responsiveness; - 7x faster than baselines.

Wildminder

11,556 subscribers

58,867 views • 9 months ago •via X (Twitter)

Arts Education Science & Technology

Anya Rossi• Live Now

Private livecam show

15 Comments

Brie Wensleydale🧀🐭9 months ago

This is a flaw. Looks purdy good though!

tas9 months ago

Extraordinarily creepy examples, we're just building brainwashing/cult recruitment and retention tech?

Shiva Mangal9 months ago

Wow this is incredible

Prompt Father | echo9 months ago

Wow! Will you release code and when? What license?

npc9 months ago

basically @codemiko

Ahmad Zaim Hilmi9 months ago

Just wow 😳😳

Himanshu Kumar9 months ago

Impressive, the real-time interaction is a significant advancement.

Anurag Singh7 months ago

Code with be released SOON, how long is SOON

Pathetic Brook9 months ago

This is pretty good!

Josue9 months ago

Beautiful

Ha.데스9 months ago

Impressive

Stav Zilbershtein9 months ago

Very impressive

OcoeeRiversideFarm9 months ago

How to use this?

Ericreator9 months ago

dead society incoming

Lydia9 months ago

Awesome, can this be used for live streaming?

Related Videos

Large-scale natural environments are a challenge for real-time ray tracing. RTX Mega Geometry handles massive scenes and updates them up to 100x faster than previous methods, enabling full‑fidelity path tracing with advanced detail.

Large-scale natural environments are a challenge for real-time ray tracing. RTX Mega Geometry handles massive scenes and updates them up to 100x faster than previous methods, enabling full‑fidelity path tracing with advanced detail.

NVIDIA GeForce

322,657 views • 7 months ago

🚀 PixVerse just launched Interactive Live Avatars — and it’s insane! Now you can create your own AI avatar and interact with it in real time. Not just a video… a fully live, responsive experience. Here’s how it works 🧵 #PixVerseR1 PixVerse

Riya Ai

30,866 views • 6 months ago

just wrapped up the PBR path tracer for my engine, and looks like it's about 5x faster than Unreal Engine and 12x faster than Unity for the same scene. goal is to make this real time

just wrapped up the PBR path tracer for my engine, and looks like it's about 5x faster than Unreal Engine and 12x faster than Unity for the same scene. goal is to make this real time

Gabriel Dechichi

79,337 views • 7 months ago

AI AVATARS JUST GOT DANGEROUSLY REAL. Mirage just released Avatar X and it's setting the standard for preserving identity in Avatar models. Your facial movements, expressions, and the things that make you, you. All preserved with Avatar X across both horizontal and vertical generations is one of the best on the market right now. The side-by-side difference is wild ↓

Farhan

116,124 views • 2 months ago

There's more to #AI than faster work. 🎯💡⚡️ Discover how agentic AI handles the routine so you can focus on work that makes a real difference. It's time to elevate your impact, not just your output. #AWS 👉

There's more to #AI than faster work. 🎯💡⚡️ Discover how agentic AI handles the routine so you can focus on work that makes a real difference. It's time to elevate your impact, not just your output. #AWS 👉

Amazon Web Services

17,149 views • 1 year ago

Amazing AI talking head videos for socials Now faster than ever 💥 • Natural eye movement • Real facial expressions • Fluid, human-like body motion 🚨FREE 500 Credits in DM for first 50 users, who RT & Comment "VEED"

Amazing AI talking head videos for socials Now faster than ever 💥 • Natural eye movement • Real facial expressions • Fluid, human-like body motion 🚨FREE 500 Credits in DM for first 50 users, who RT & Comment "VEED"

VEED | AI Video Creation

15,887 views • 6 months ago

Concrete streets Asphalt streets LA has both. And neither are getting fixed faster than they're breaking. It's time for LA to take better care of our largest public asset: our streets.

Concrete streets Asphalt streets LA has both. And neither are getting fixed faster than they're breaking. It's time for LA to take better care of our largest public asset: our streets.

Streets For All

256,439 views • 13 days ago

Vivix.AI just created the most realistic live AI avatar I have ever used with A1. You can create & deploy AI avatars that move, speak, & interact just like humans in REAL time. I've been searching for this exact tool for my new AI companion project and I finally found it.

Mark Gadala-Maria

68,145 views • 17 days ago

Users define the experience. Bubble Code brings the AI host live. Through xBubble, you can build and deploy an interactive avatar livestream room with real-time video, audience chat, live responses, and session controls. Not a prerecorded digital human, a ready-to-use AI livestream product built for real interaction.

DAPPOS

106,790 views • 2 months ago

Introducing Interactive Build. Build faster. Build smarter. Interactive Build lets you collaborate directly with Alchemist AI via text or voice, powered by OpenAI’s real-time APIs. No need to craft the perfect prompt from scratch, just start talking. Describe your idea, and Alchemist AI will guide you with intelligent, dynamic questions, transforming your vision into a structured, optimized prompt ready for generation. Whether you’re starting fresh or refining an existing project, Interactive Build streamlines the process, taking you from concept to execution faster, with less effort and no guesswork. One conversation. One creation. All in real time.

Introducing Interactive Build. Build faster. Build smarter. Interactive Build lets you collaborate directly with Alchemist AI via text or voice, powered by OpenAI’s real-time APIs. No need to craft the perfect prompt from scratch, just start talking. Describe your idea, and Alchemist AI will guide you with intelligent, dynamic questions, transforming your vision into a structured, optimized prompt ready for generation. Whether you’re starting fresh or refining an existing project, Interactive Build streamlines the process, taking you from concept to execution faster, with less effort and no guesswork. One conversation. One creation. All in real time.

ALCHEMIST AI 🔮

19,789 views • 1 year ago

Miso One is an open-source voice model built for real-time conversations, responding in about 110 milliseconds while also cloning voices from short audio samples. Its speed makes Al voice chats feel smoother, faster, and much closer to natural human conversation

Miso One is an open-source voice model built for real-time conversations, responding in about 110 milliseconds while also cloning voices from short audio samples. Its speed makes Al voice chats feel smoother, faster, and much closer to natural human conversation

Hugging Models

15,754 views • 4 months ago

#ClimateChange 🌡️ exceptional and caused by human greenhouse gas emissions 🏭 We are causing #globalwarming 100 times faster than past natural changes We are taking Earth’s 🌎 climate beyond natural limits, with carbon dioxide & temperatures levels not seen for 3 million years

#ClimateChange 🌡️ exceptional and caused by human greenhouse gas emissions 🏭 We are causing #globalwarming 100 times faster than past natural changes We are taking Earth’s 🌎 climate beyond natural limits, with carbon dioxide & temperatures levels not seen for 3 million years

Professor Mark Maslin 👋 𝕏

508,072 views • 3 years ago

Someone used claude to build an interactive virtual lounge and digital hangout space tailored for vibe coders. → builds an ambient, browser-based gathering space for developers to connect while building with ai → integrates lightweight real-time interaction, avatar presence, and chat functionality → scaffolds the frontend environment and interactive elements directly from natural language prompts → eliminates manual networking and backend configuration through ai-assisted boilerplate generation → demonstrates how builders can rapidly prototype and launch niche social web apps using conversational workflows

0xMarioNawfal

54,977 views • 1 month ago

There is a subtle architecture shift happening in voice AI. The voice stack is becoming part of the agent's execution loop. Cartesia is combining the listening and speaking paths around that loop. Sonic-3.6 turns text into speech (90ms latency) and Ink-2 turns speech into text (100ms transcript latency), faster than anything else streaming. now holds the #1 spot for both speaking and listening models. Becasue, voice agents are one of those systems where 100ms in the wrong place is very noticeable. Cartesia is attacking both sides of that loop at once. super interesting work by Karan Goel and team.

There is a subtle architecture shift happening in voice AI. The voice stack is becoming part of the agent's execution loop. Cartesia is combining the listening and speaking paths around that loop. Sonic-3.6 turns text into speech (90ms latency) and Ink-2 turns speech into text (100ms transcript latency), faster than anything else streaming. now holds the #1 spot for both speaking and listening models. Becasue, voice agents are one of those systems where 100ms in the wrong place is very noticeable. Cartesia is attacking both sides of that loop at once. super interesting work by Karan Goel and team.

Rohan Paul

310,545 views • 1 month ago

It's time for Plaid Cymru to listen. When Welsh councils & teaching unions are saying the same thing - that more funding should go to our schools for ALN - you have to wonder why they're not listening. Our public services deserve better than Plaid Cymru's political games.

It's time for Plaid Cymru to listen. When Welsh councils & teaching unions are saying the same thing - that more funding should go to our schools for ALN - you have to wonder why they're not listening. Our public services deserve better than Plaid Cymru's political games.

Huw Thomas MS

16,386 views • 2 months ago

stumbled onto something genuinely impressive today - real-time interactive AI avatars with 180ms latency let me break down why I think this is one of the more underrated AI products right now 👇 most AI agents are either a chat box or a robotic voice. anam gives them a face. photorealistic. micro-expressions. no looping, no mouth-dubbing nonsense. the numbers are hard to ignore: → +44% engagement when your agent has a face vs text/voice only → CARA-3 (their latest model) is #1 across ALL avatar benchmarks in 2025 → 33% faster than the next-best competitor → 180ms avg response time, literally faster than a blink → 70+ languages with native voices → plug in your own LLM, your own voice, your own avatar image use cases they're already disrupting: customer support, sales, medical services, language tutoring, skill training and it's actually developer-friendly. clean SDK, LiveKit + Pipecat integrations, no-code option if you don't want to touch the API if you're building AI products or advising companies on AI strategy, this is one to watch

Vyom

268,158 views • 4 months ago

We just wrapped Synthesia 3.0! Here’s what we announced: Video Agents: Video you can talk to in real-time. Add a Video Agent at any point in your video - and let them run training roleplays, do knowledge checks, interview candidates and much more. Avatars: Now they move and talk like professional speakers. Create an Avatar from a single prompt or image - with facial expressions, perfect lip sync, and natural hand & body movements. Copilot: Your very own professional video editor. Writes your script in seconds, connects to knowledge bases, and even suggests visual elements. Interactivity 2.0: Get viewers to engage and take action with interactive elements like quizzes, polls, hotspots, and forms. Courses: Design interactive learning experience that combine Avatars, Video Agents, and Interactivity - and measure performance data. I’ll be sharing more about each specific update in the coming days. More about Synthesia 🎥 next chapter here:

Victor Riparbelli

29,006 views • 1 year ago

Most AI avatars still struggle with the same problems: lip-sync breaks during head turns, faces become unstable when partially covered, and movements often feel robotic. After looking into Wizstar's approach, the technical side is what stands out. Instead of directly generating facial motion from audio, Wizstar uses a two-stage, audio-driven animation pipeline: • Separates speech, mouth motion, and head pose into independent signals • Reconstructs facial textures and expressions afterward for greater realism and stability That architecture helps maintain accurate lip-sync even when the mouth is partially obscured, the camera angle changes dramatically, or the speaker makes fast movements. What impressed me most is that the avatars don't just talk—they gesture, interact with objects, and move more like real presenters than traditional AI avatars. And it’s not just the tech that caught attention—Wizstar has now claimed the #1 spot among Product Hunt’s Top Products. 🏆 🧵 For creators and businesses, the bigger advantage is scale: → Create a digital avatar once → Generate multilingual content in 75+ languages → Produce localized videos without filming, reshoots, or studio setups Marketing videos, product demos, training content, sales outreach, global campaigns—everything becomes significantly faster to produce. Definitely one of the more interesting AI avatar systems I've seen recently. Try it here: Wizstar_official #Wizstar

ɱҽԃι✨

49,060 views • 1 month ago