Spell Engine

I really enjoyed making this animated episode for @Framer_X and @abydotone’s Comedy Sitcom Competition! The premise: what if our gadgets were gossiping behind our backs?

@Framer_X Was not expecting this 😂 love it

I recreated a Britney Spears music video in 3D. Here's how 👇

@matt_elevenlabs @ElevenCreative Done!

We just released Music Marketplace in ElevenCreative. To celebrate, we’re giving away 5,000 free credits so you can create and publish your first track. For the next 6 hours: retweet + follow @ElevenCreative and we’ll DM you the credits (must follow).

@kanavtwt 😂😂😂

Someone built a Google translate for Linkedin 😭

@PocketScreenAI @neco1751662 @yapper_so Makes sense about motion control! Also I wonder if certain platforms have a different approach to video ref as well 🤔

@spellengineAI @neco1751662 @yapper_so Hmm, maybe further testing is needed. Although i believe on Jimeng, motion control and video ref were different features. That or my test could be flawed too so worth trying more.

Seedance 2.0 Test 9 - Video Ref test using @yapper_so : Expand for prompt. TLDR: The video ref might not be reading the entire video but only the opening frame. As such, you would probably get similar results with audio ref + images. When I started the test, my intent was to use the original like the master shot, then each retake like a cam A/B set up. I was hoping it could generate new angles on the exact same take. As you can see, the movement timings are slightly different in each take. Unfortunate. However, voices and pacing basically came out perfectly identical each time. It is also quite clear that it understood the context of what was being said as I gave it no dialogue in my prompts. Additionally, in my original video that was made on grok the lighting gets brighter. This was not replicated in SD 2 and the lighting remained consistent to the opening frame the whole clip. As many places charge more for video ref but not audio ref. You can probably get similar results using just the audio and add the starting frame. Remember you can just prompt the image as the starting frame. Prompts: You are camera A. Your purpose is to maintain close up on the woman sitting on the right for the entire duration of this scene. Video1 . One continuous shot. No cuts. You are camera B set up on a chrane. The camera does a slow orbit starting with a wide shot, then orbiting to a behind the shouldershot behind the man on the left in this exact scene Video1 . One continuous shot. No cuts. A/N: I wanted 1 more with the camera staying on the man but he likely looked too much like Taki that it kept getting IP blocked.

@PocketScreenAI @neco1751662 @yapper_so The examples i saw didn’t seem like a prompt based recreation, it was the exact same scene (how models like Kling Omni remove objects) i’ll flick it over if I come across it again. There’s also that La La Land Dance where characters followed the exact same steps

@spellengineAI @neco1751662 @yapper_so It's hard to tell without seeing their prompt. I could prompt this saying they are drinking martinis and it would probably generate the scene with martinis.

@PocketScreenAI @neco1751662 @yapper_so I’ve seen people post the same videos but with objects removed though, I wonder how that’s possible if just first frame is used in video ref

@spellengineAI @neco1751662 @yapper_so Extending the video function exists. However, I think it is again, using the audio reference and just the last frame and any references used to generate the first clip. This is why the music seems to continue and the voices are carried forward.

@yapper_so Seedance

🚨Free Seedance 2.0 giveaway!🚨 We're giving away 20 more monthly subscriptions with Seedance 2.0 access today 🎁 Here's how to win! ⚡️Like, repost, and comment "seedance" ⏰ 20 lucky winners will be announced in the next 24h Must be following @yapper_so!

@chrisfirst ✋

Who wants early access to a new AI tool from a major AI company?👀 Drop a comment and I’ll add you to the list.

@get_artcraft Seedance

🚨Free Seedance 2.0 Film Set Giveaway!🚨 Open Source Virtual Film Set with Seedance 2.0 Giving 10 monthly subs. How to win ⚡️ Like, repost, and comment "seedance" ⏰ 10 lucky winners will be announced in the next 24 hours Must be following @get_artcraft ! Your Film Studio

@sumiturkude007 Good effort but frankly, there’s quite a few continuity issues and the lady sounds quite robotic

This short film made with Seedance 2.0 is absolutely insane. The realism looks like a real movie — no one can tell it's AI.

@Framer_X @PocketScreenAI Such a good mental position to be - I admire your maturity @Framer_X. Still kind of @PocketScreenAI to flag it though :)

Thanks for sharing buddy. But I tend not to care bc: 1. That happens often and there is nothing i can do about it so why to even care 2. These posts are targeting audience which im not oriented yet 3. These people will never build anything longterm so they are not my competitors cheers!

This is word for word @Framer_X's post. Did he get permission from you to do this? Noticed you weren't tagged anywhere. 🤨

@MayorKingAI I love it!

MidJourney V7 Style Creator I just created a new style using --sref 7129209087 I’m using it for fantasy medieval anime / 2D animation aesthetics great for RPG-style characters and cinematic fantasy scenes It’s not fully refined yet, but you can push the look by adding “2D anime style” in the prompt Feel free to try it 👇

@matchaman11 @thecodemind What do i enter into the field? It says not authenticated. Also what’s the cost per video for SD2?

@thecodemind Here muapi.ai/access-keys

Seedance 2.0 Image to Video is now on Open-Higgsfield-AI 🔥🔥 First in the world to support 9 reference images 🤯 Hollywood grade cinematic video in seconds 📸 As always ✅Free with BYOK ✅Open-source ✅No Higgsfield subscription tax Link to project below 👇

Very excited to try this. Finally a worthy Nano Banana competitor?

@AgentOpusAI Story

All I did was upload an audio clip. Agent Opus Story Mode turned it into a fully produced video - matched visuals, pacing, everything. No editing required. Want to see what it does with your content? We're giving away free credits to the first 100 people who: ♻️ Repost this 💬 Drop "Story" in the comments We'll DM you.

@PocketScreenAI I’ve had this happen since I signed up exactly a month ago 😩

I don't know what happened but Vidu Q3 can't seem to get the dialogue correct suddenly. It seems to skip sections of dialogue for no reason. I don't recall this happening even 2-3weeks ago. What changed?　Anyone heard anything? Prompt: The boy and girl are having a nostalgic friendly conversation. [0:00-0:06] The boy says gently, "あのころはそんなあたりまえなことがたのしかったな" [Cut: 0:06-0:11] Close up on the girl as she replies with happy nostalgia, "うん、そうだね。いまおもうときちょうなじかんだった。” [0:11-0:15] Continue close up on the girl as she replies with 微笑み, "そうた、いつもへんなうた、うたってたよね。” Must not have music. Friends grown up sharing a drink vibe. @ViduAI_official #viduq3

@get_artcraft Seedance

🚨Open Source Seedance 2.0 Giveaway🚨 Websites are ⚠️Lock-In⚠️ We're Open Source. We let you bring your other subs + API keys How to win: ⚡️ Like, repost, follow, comment "seedance" ⏰ 10 lucky winners in the next 24h If you want OPEN tools, infra, models: Subscribe

@MayorKingAI Still great to have though! Nice to see Turkish support even being included

@spellengineAI I believe you! Maybe not all languages have the same quality

Higgsfield just dropped Higgsfield Audio 21 voice presets, up to 3 custom voices, instant voice swap, and translate + lip-sync in +10 languages Supports 70+ input languages and 4 AI voice models including Eleven v3 Image, video, and now audio all inside Higgsfield 10 examples below 👇 #higgsfieldpartner