Made with the open source veo3 - multitalk
August 7, 2025 at 1:22 AM
Made with the open source veo3 - multitalk
it “improves the pre-trained T2I model by up to 20%”.
In another paper, they present the MultiTalk dataset, comprising over 420 hours of talking videos in 20 languages scrapped entirely from YouTube, which they used to enhance 3D talking head generation. (11/23)
In another paper, they present the MultiTalk dataset, comprising over 420 hours of talking videos in 20 languages scrapped entirely from YouTube, which they used to enhance 3D talking head generation. (11/23)
March 30, 2025 at 3:27 PM
it “improves the pre-trained T2I model by up to 20%”.
In another paper, they present the MultiTalk dataset, comprising over 420 hours of talking videos in 20 languages scrapped entirely from YouTube, which they used to enhance 3D talking head generation. (11/23)
In another paper, they present the MultiTalk dataset, comprising over 420 hours of talking videos in 20 languages scrapped entirely from YouTube, which they used to enhance 3D talking head generation. (11/23)
If jellycats could talk 🥐 #Jellycat #JellycatCroissant #aianimation #aivideo #multitalk #aivideogeneration
August 2, 2025 at 6:47 PM
If jellycats could talk 🥐 #Jellycat #JellycatCroissant #aianimation #aivideo #multitalk #aivideogeneration
MultiTalk is a novel framework for audio-driven multi-person conversational video generation.
Given a multi-stream audio input, a reference image and a prompt, MultiTalk generates a video containing interactions following the prompt, with consistent lip motions aligned with the audio.
Given a multi-stream audio input, a reference image and a prompt, MultiTalk generates a video containing interactions following the prompt, with consistent lip motions aligned with the audio.
June 26, 2025 at 5:19 PM
MultiTalk is a novel framework for audio-driven multi-person conversational video generation.
Given a multi-stream audio input, a reference image and a prompt, MultiTalk generates a video containing interactions following the prompt, with consistent lip motions aligned with the audio.
Given a multi-stream audio input, a reference image and a prompt, MultiTalk generates a video containing interactions following the prompt, with consistent lip motions aligned with the audio.
Hey everyone! My brand-new MV ‘Cozy’ is now live! 🎵
Created using these cutting-edge AI tools: [Midjourney, Dreamnia, Flux Kontext, MultiTalk, Omnihuman, Skyreel, Suno].
Check it out and let me know what you think! ❤️
🧵1/2
youtu.be/JAvzyjMmrvM?...
Created using these cutting-edge AI tools: [Midjourney, Dreamnia, Flux Kontext, MultiTalk, Omnihuman, Skyreel, Suno].
Check it out and let me know what you think! ❤️
🧵1/2
youtu.be/JAvzyjMmrvM?...
MV ‘Cozy’
YouTube video by luokai
youtu.be
August 12, 2025 at 1:18 PM
Hey everyone! My brand-new MV ‘Cozy’ is now live! 🎵
Created using these cutting-edge AI tools: [Midjourney, Dreamnia, Flux Kontext, MultiTalk, Omnihuman, Skyreel, Suno].
Check it out and let me know what you think! ❤️
🧵1/2
youtu.be/JAvzyjMmrvM?...
Created using these cutting-edge AI tools: [Midjourney, Dreamnia, Flux Kontext, MultiTalk, Omnihuman, Skyreel, Suno].
Check it out and let me know what you think! ❤️
🧵1/2
youtu.be/JAvzyjMmrvM?...
A new original MV is here! Once again, everything from the music 🎵 to the video 📹 was created using AI. I hope you enjoy it! ❤️
AI tools used: Midjourney, Dreamnia, Nano-banana, Kling, MultiTalk, Omnihuman, Skyreels, Suno.
youtu.be/NP6462xhCiw
AI tools used: Midjourney, Dreamnia, Nano-banana, Kling, MultiTalk, Omnihuman, Skyreels, Suno.
youtu.be/NP6462xhCiw
Shadow Under the Neon｜霓虹下的暗影: An AI-Generated MV
YouTube video by luokai
youtu.be
August 30, 2025 at 4:20 PM
A new original MV is here! Once again, everything from the music 🎵 to the video 📹 was created using AI. I hope you enjoy it! ❤️
AI tools used: Midjourney, Dreamnia, Nano-banana, Kling, MultiTalk, Omnihuman, Skyreels, Suno.
youtu.be/NP6462xhCiw
AI tools used: Midjourney, Dreamnia, Nano-banana, Kling, MultiTalk, Omnihuman, Skyreels, Suno.
youtu.be/NP6462xhCiw
We tested out the 🔥new open source model MultiTalk so you don’t have to (but you should - it’s pretty impressive) #multitalk #aivideo #aianimation #opensource #veo3
July 17, 2025 at 4:37 PM
We tested out the 🔥new open source model MultiTalk so you don’t have to (but you should - it’s pretty impressive) #multitalk #aivideo #aianimation #opensource #veo3
The highs and lows of software version mismanagement #aianimation #multitalk #modal #softwareengineer
August 7, 2025 at 7:36 PM
The highs and lows of software version mismanagement #aianimation #multitalk #modal #softwareengineer
Learn how AI image generation works from a polar bear in less than 1 minute #stablediffusion #aianimation #learnai #multitalk #wan2gp
July 25, 2025 at 4:57 PM
Learn how AI image generation works from a polar bear in less than 1 minute #stablediffusion #aianimation #learnai #multitalk #wan2gp
as they said in the video above, this is from a paper called MultiTalk which can generate videos of multiple people talking by using audio from different sources, a reference image, and a prompt
project page: meigen-ai.github.io/multi-talk
code (coming soon): github.com/MeiGen-AI/M...
project page: meigen-ai.github.io/multi-talk
code (coming soon): github.com/MeiGen-AI/M...
GitHub - MeiGen-AI/MultiTalk: Let Them Talk: Audio-Driven Multi-Person Conversational Video Generation
Let Them Talk: Audio-Driven Multi-Person Conversational Video Generation - MeiGen-AI/MultiTalk
github.com
June 7, 2025 at 7:46 AM
as they said in the video above, this is from a paper called MultiTalk which can generate videos of multiple people talking by using audio from different sources, a reference image, and a prompt
project page: meigen-ai.github.io/multi-talk
code (coming soon): github.com/MeiGen-AI/M...
project page: meigen-ai.github.io/multi-talk
code (coming soon): github.com/MeiGen-AI/M...
For TTS there's Namigen (Mac) on its way.
For the lip-sync stuff there's a solution for Windows:
MultiTalk Wan2GP (1.3GB model)
But even if ported to Mac and highly performant it's maybe still too limited to use it on a Mac with 10 GB VRAM: maybe we get 480px, 25/s?.
For the lip-sync stuff there's a solution for Windows:
MultiTalk Wan2GP (1.3GB model)
But even if ported to Mac and highly performant it's maybe still too limited to use it on a Mac with 10 GB VRAM: maybe we get 480px, 25/s?.
July 19, 2025 at 8:35 PM
For TTS there's Namigen (Mac) on its way.
For the lip-sync stuff there's a solution for Windows:
MultiTalk Wan2GP (1.3GB model)
But even if ported to Mac and highly performant it's maybe still too limited to use it on a Mac with 10 GB VRAM: maybe we get 480px, 25/s?.
For the lip-sync stuff there's a solution for Windows:
MultiTalk Wan2GP (1.3GB model)
But even if ported to Mac and highly performant it's maybe still too limited to use it on a Mac with 10 GB VRAM: maybe we get 480px, 25/s?.
👉 Upload a pic + separate audio streams for each character
👉 It animates each character, even in cartoons or singing
👉Open source
#MultiTalk #AI #AIvideo #opensource #GenAI
👉 It animates each character, even in cartoons or singing
👉Open source
#MultiTalk #AI #AIvideo #opensource #GenAI
July 9, 2025 at 8:41 PM
👉 Upload a pic + separate audio streams for each character
👉 It animates each character, even in cartoons or singing
👉Open source
#MultiTalk #AI #AIvideo #opensource #GenAI
👉 It animates each character, even in cartoons or singing
👉Open source
#MultiTalk #AI #AIvideo #opensource #GenAI
Kim Sung-Bin, Lee Chae-Yeon, Gihun Son, Oh Hyun-Bin, Janghoon Ju, Suekyeong Nam, Tae-Hyun Oh
MultiTalk: Enhancing 3D Talking Head Generation Across Languages with Multilingual Video Dataset
https://arxiv.org/abs/2406.14272
MultiTalk: Enhancing 3D Talking Head Generation Across Languages with Multilingual Video Dataset
https://arxiv.org/abs/2406.14272
June 21, 2024 at 4:02 PM
Kim Sung-Bin, Lee Chae-Yeon, Gihun Son, Oh Hyun-Bin, Janghoon Ju, Suekyeong Nam, Tae-Hyun Oh
MultiTalk: Enhancing 3D Talking Head Generation Across Languages with Multilingual Video Dataset
https://arxiv.org/abs/2406.14272
MultiTalk: Enhancing 3D Talking Head Generation Across Languages with Multilingual Video Dataset
https://arxiv.org/abs/2406.14272
Venkata Naren Devarakonda, Ali Umut Kaypak, Shuaihang Yuan, Prashanth Krishnamurthy, Yi Fang, Farshad Khorrami
MultiTalk: Introspective and Extrospective Dialogue for Human-Environment-LLM Alignment
https://arxiv.org/abs/2409.16455
MultiTalk: Introspective and Extrospective Dialogue for Human-Environment-LLM Alignment
https://arxiv.org/abs/2409.16455
September 26, 2024 at 4:02 AM
Venkata Naren Devarakonda, Ali Umut Kaypak, Shuaihang Yuan, Prashanth Krishnamurthy, Yi Fang, Farshad Khorrami
MultiTalk: Introspective and Extrospective Dialogue for Human-Environment-LLM Alignment
https://arxiv.org/abs/2409.16455
MultiTalk: Introspective and Extrospective Dialogue for Human-Environment-LLM Alignment
https://arxiv.org/abs/2409.16455
July 18, 2025 at 12:49 AM
Made with multitalk (open source) youtube.com/shorts/QTij-...
capybara fever
YouTube video by Fuzz Puppy
youtube.com
July 16, 2025 at 7:54 PM
Made with multitalk (open source) youtube.com/shorts/QTij-...
New AI Model dropped - MultiTalk lets you provide audio for MULTIPLE characters
July 9, 2025 at 8:40 PM
New AI Model dropped - MultiTalk lets you provide audio for MULTIPLE characters
September 25, 2025 at 11:26 AM
for preserving the instruction-following ability of the base model. MultiTalk achieves superior performance compared to other methods on several datasets, including talking head, talking body, and multi-person datasets, demonstrating the powerful [5/6 of https://arxiv.org/abs/2505.22647v1]
May 29, 2025 at 6:23 AM
for preserving the instruction-following ability of the base model. MultiTalk achieves superior performance compared to other methods on several datasets, including talking head, talking body, and multi-person datasets, demonstrating the powerful [5/6 of https://arxiv.org/abs/2505.22647v1]
To solve this problem, in this paper, we propose a novel task: Multi-Person Conversational Video Generation, and introduce a new framework, MultiTalk, to address the challenges during multi-person generation. Specifically, for audio injection, we [3/6 of https://arxiv.org/abs/2505.22647v1]
May 29, 2025 at 6:23 AM
To solve this problem, in this paper, we propose a novel task: Multi-Person Conversational Video Generation, and introduce a new framework, MultiTalk, to address the challenges during multi-person generation. Specifically, for audio injection, we [3/6 of https://arxiv.org/abs/2505.22647v1]
June 27, 2025 at 4:32 AM
meigen-ai.github.io/multi-talk/?...
MultiTalk, a novel framework for audio-driven multi-person conversational video generation.
MultiTalk, a novel framework for audio-driven multi-person conversational video generation.
June 16, 2025 at 7:50 PM
meigen-ai.github.io/multi-talk/?...
MultiTalk, a novel framework for audio-driven multi-person conversational video generation.
MultiTalk, a novel framework for audio-driven multi-person conversational video generation.
The MultiTalk project enhances animation quality for lip-sync AI workflows using GPUs like the A6000. Discover updated capabilities for image-video generation via advanced tools. Learn more here: https://huggingface.co/blog/MonsterMMORPG/multitalk-levelled-up-way-better-animation
July 17, 2025 at 6:00 AM
The MultiTalk project enhances animation quality for lip-sync AI workflows using GPUs like the A6000. Discover updated capabilities for image-video generation via advanced tools. Learn more here: https://huggingface.co/blog/MonsterMMORPG/multitalk-levelled-up-way-better-animation