Remaking an old lip sync test with another lip... sync test. Still Work in progress model but Imma really digging testin out lower fps / imitation stop motion.show more

This is delicious. A quick test of Bytedance's new... lip sync model. The lip sync looks really good and body movements are pretty dynamic. Both close up and medium shots work really well. Not bad at all! What do you think?show more

First test of LatentSync; an open source lip sync... tool 🔥 Link in thread 🧵 It only requires a source video and a spoken audio clip. 🔊 Google Veo2: video clips E2-F5-TTS: Text to Speech (o/s) LatentSync: Lip sync (o/s) MMAudio: Background SFX (o/s)show more

I managed to add lip sync with audio but... somehow I can't change the audio in this model so it's stuck at Neil Degrassi Tyson 😂 But okay we have a proof of concept, from AI pic -> AI video -> AI lipsync'd I wanted to add an AI voice, anyone know how to make the audio work on this model? Chenxi Whitehouseshow more

We may have cracked motion control in AI video.... I've been testing the new Kling AI tool, and it's shockingly good. It can handle complex movements I've never gotten another video-to-video model to do - like this gymnastics clip. It can also do lip sync and camera movement 👇show more

Tested out the new Grok Imagine video model ✨... The speech, sound, and lip sync upgrades are really impressive. It works with lots of face angles, multiple speakers, pets - and the prompt adherence is quite strong. A few clips I generated this morning (all one shots) 👇show more

I got to lip sync in real life with... the REAL Caitlyn, Katie Leung??!!!! This is an alt take of the one up on Riot Games, i kinda think i did better on this one but the ending of the other one is cuter! can you even believe 🤯🤯 #arcaneshow more

Our metroidvania has an arena mode. It started as... a developer tool to test out guns and enemies but it was too fun so we added it in the game. Fun place to grapple hook to your heart's content. (The UI is still a work in progress) Wishlist Iron Bramble on Steam!show more

AI video just reached a new level. Seedance 1.5... Pro by BytePlus turns static images into cinematic videos with real action and realism. Key features: • Native audio generation with ultra precise lip sync • Multilingual dialogues with multiple speakers • Natural motion, expressive emotions and cinematic quality visuals • Stable, premium output ready for narration, ads and professional creative work This does not look like an AI video. It feels like real footage, synchronized, expressive and production ready. #Seedance #AIVideo #VideoGeneration #BytePlusAIshow more

currently experimenting with WAN 2.6 I2V on GMI Cloud... in this test, I’m comparing two audio workflows and honestly both perform really well. one scene uses audio generated directly from the prompt, while the other uses manually uploaded audio taken from the film 300. visually, both deliver strong motion and solid performance. however, the version with audio coming straight from the prompt feels slightly more refined, camera movement is smoother, transitions flow more naturally and the sync between voice, facial motion, and pacing feels more cohesive. lip sync, especially for Chinese dialogue also comes across a bit cleaner. you can choose single shot for a clean, focused moment or multi shot if you want more cinematic transitions, even when working from just one reference image. one important note: always turn on prompt extension. it makes a noticeable difference in how well the model understands motion, transitions and overall scene flow. both audio approaches are totally usable, but if you’re building dialogue driven or cinematic scenes, starting with audio from the prompt gives WAN 2.6 a bit more context to work with. I’ll be pushing this further with more dynamic camera movement and transitions next. more experiments coming soon ✨ Wanshow more

Seedance 2 just dropped. This was my first test... and it changed everything. Perfect lip sync. Realistic movement. Multi-shot support. Natural voice completely human, zero robotic tone. And the SFX? It nailed the sound effects automatically. Every detail synced perfectly. This is a game-changer. From now on, automation isn't theory it's real. And yes, it's happening on Crafft. With a high-fidelity model like this + the right prompting rules and skills, you can generate endless variations that work every single time. Crafft agents will handle the whole process. One model, infinite outputs. You can try Seedance 2 on CapCut right now. RT + comment "PROMPT" and I'll send you the exact setup (follow so I can DM)show more

Avatar is really really fun but we need to... ban unemployed people from Alpha/Betas because why am I getting hit with a Katara infinite in an Alpha. No point in playing if you don't want to test the rest of the game out IMO. Zuko the truth thoughshow more

Gm ☁️✨ Yesterday I put together the first low... res draft of the collaborative piece I’m working on. Pretty damn happy with how it’s looking 👀 Still lots to do though! Here’s an animation test I rendered out a few weeks ago. Just a small section of the scene in progress 🙌🏻show more

Another score in the 60s! Round 2 of The... Manchester Open at Didsbury Golf Club Backed up yesterday’s 67 (-2) with a 69 (Level) today. Didn’t quite strike it as well as I have in previous rounds, but managed my way around and kept it tidy. Only a couple of mental errors which cost me and some horrible lip outs! Another round in the 60s and another step in the right direction. This puts me at T31 in a deep field full of top PGA pros — a solid position, but still around 10 shots off the winning score. Plenty of work ahead, and I know I need to keep pushing to close that gap. Really happy with how the season is shaping up and the progress I’m making. Big thanks to my playing partners over the past few days, always makes a difference having a good group out there. Onward and upward! #trusttheprocessshow more

A quick weekend test of adding colorful plaster to... the walls 🖌 Still needs a lot of work to make it real if we decide to go there, but I was too giddy with how it turned out so I had to share 😄 (also a sneak peek of windows & improved arches! more in the upcoming newsletter💌)show more

Best B14.1 slap from the chopsticks so far: (For... those just catching up, SpaceX is using this test tank to work on their allowances with the sticks ahead of Flight 5's potential catch. So slapping B14.1 is fine for gathering data. It's much better than slapping an actual flight booster. This way, they'll work out the fastest they can grab, but without damaging the booster in the process.show more

🇨🇳 Another great Chinese Model, OmniHuman-1.5 from ByteDance Turns... 1 image plus a voice track into expressive avatar video by pairing a System 1 and System 2 inspired planner with a Diffusion Transformer, Produces coherent motion for over 1 minute with moving camera and multi character scenes. Most avatar models move to the beat of the audio but miss meaning, so gestures feel generic and emotions feel shallow. The fix here is a Multimodal LLM planner that listens to the speech and drafts a structured plan describing intent, emotions, beats, and high level actions, which gives the motion engine clear semantic targets instead of only rhythm. The motion engine is a Multimodal Diffusion Transformer that fuses the plan with audio, the single reference image, and optional text prompts, then synthesizes continuous body, face, and head motion that matches both words and tone. A key trick is a Pseudo Last Frame, a synthetic target that summarizes the next expected state, which stabilizes fusion across modalities and keeps motion consistent over long spans. From just 1 image and speech, the system outputs speaking avatars with synchronized lips, context aware gestures, and continuous camera movement, and it also supports multi character interactions without manual choreography. Reported results show strong lip sync accuracy, high video quality, natural motion, and close match to text prompts, and the same setup works on nonhuman characters too.show more

✨ Excited to share QVQ-Max, our visual reasoning model... that's still evolving We've been experimenting with this approach for a while - try it out on Qwen Chat! ( 🚀 Just upload any image or video, ask away, and hit the "Thinking" button to see how it processes visual information step-by-step. It's a work-in-progress but fascinating to watch! Your early feedback will be super helpful as we continue developing! 🙏 Blog:show more

Kling 2.6 Motion Control is absolutely insane 🤯 Take... any reference video and transfer the exact motion onto an AI character: full-body sync, facial expressions, hand gestures, everything. All with just a few clicks. Perfect for e-comm brands and agencies creating AI video ads that don't look like AI. Here's the problem: AI-generated video ads still look robotic. The movements are stiff, the expressions are flat. Your audience clocks it as AI instantly and keeps scrolling. Kling 2.6 Motion Control fixes it: → Start with any reference clip (stock footage, existing UGC, motion reference) → Upload to Kling → Map the exact movement onto any AI character → Full-body motion, hand gestures, facial expressions—all transferred → Generate up to 30 seconds of video No stiff AI movements, no uncanny valley, no instant "skip this ad" reaction. What this unlocks: - Use one winning UGC motion → swap in different AI creators - Pull reference clips from anywhere → generate branded variations - Create dynamic AI video ads with real human movement - Test multiple "creators" without filming anyone new I recorded a quick walkthrough showing how to do this step-by-step. Want access? > Comment "KLING" > Like this post And I'll send it over (must be following so I can DM)show more

The term "continual learning" has become overloaded if you... see it as an ML problem. One classic thread is about memorization: regularization-based continual learning methods, such as EWC, MAS, and SI, estimate which parameters mattered for previous tasks and resist changing them too much. One modern thread is about adaptation: test-time training and inference-time learning methods, such as TTT, adapt part of the model on the incoming test stream before making predictions. These are sometimes discussed as separate threads. But in modern scalable architectures, I think they are better seen as complementary constraints: a model that learns quickly at test time also benefits from a mechanism for deciding what not to forget. In our #ECCV2026 paper, we study this in large-scale 4D reconstruction: how to build fast spatial memory that can adapt over long observation streams while reducing collapse and forgetting. Instead of using fully plastic test-time updates, we stabilize fast-weight adaptation with an elastic prior that balances adaptation and memory. Key ideas: - Elastic Test-Time Training: Fisher-weighted consolidation for fast-weight updates - EMA anchor weights that provide a moving reference for stability - Chunk-by-chunk inference for long 3D/4D observation streams We show that this scales across large 3D/4D pretraining settings, including both LRM-style and LVSM-style models, and improves reconstruction across benchmarks including Stereo4D, NVIDIA, and DL3DV-140. We release model checkpoints across different design choices: resolution, post-training curriculum, and whether the model uses an explicit 4DGS intermediate representation. - Homepage: - Paper: - Code: - Models: This work is co-led with Xueyang Yu, contributed by Haoyu Zhen Yuncong Yang, and advised by Michigan SLED Lab Chuang Gan.show more