AI Video From Audio is BONKERS! @TheoreticallyMedia
AI Video From Audio is BONKERS!  @TheoreticallyMedia
Uploaded December 2025 | Updated November 2026, 4 days ago
Alibaba has officially released Wan 2.6, and it’s a massive leap forward for AI video generation. In today’s video, we dive deep into the new model to test its groundbreaking Audio-to-Video capabilities, character consistency with the "Starring" feature, and intelligent multi-shot generation.

But that’s not all—ByteDance has quietly rolled out Seedance 1.5 Pro (via CapCut), and we’re checking out EgoX, a mind-blowing new paper that turns third-person video into first-person POV shots. Plus, quick updates on GPT Image 1.5, Tencent’s Hunyuan World, Meta’s SAM Audio, and Kling 2.6 Voice Control.

👇 LINKS & RESOURCES MENTIONED 👇

Wan 2.6 (Alibaba): create.wan.video

Seedance 1.5 (ByteDance/CapCut): dreamina.capcut.com

EgoX Research Paper: keh0t0.github.io/EgoX

00:00 Intro: Wan 2.6 & AI Video News
00:32 Wan 2.6 Overview & Multimodal Features
01:07 Audio & Multi-Shot Capabilities
01:25 Audio-to-Video Music Video Test
02:14 Severance Inspiration & Lip Sync Analysis
02:50 Text Rendering & Physics
03:29 Text-to-Video Test: Twin Peaks Diner
04:09 When AI Gets Weird (Hallucinations)
04:37 Image-to-Video: Smart Multi-Shot Test
05:14 Analyzing Spatial Consistency
05:41 Audio-to-Video: Night of the Living Dead
06:19 Lip Sync Accuracy & Period Details
06:48 Wonky Outputs: Flamethrower Girl
07:28 Wan 2.6 "Starring" Feature (Character Consistency)
08:04 Testing "Niki" & "Idris" Characters
08:37 Bilingual & Multi-Character Prompts
08:59 Will Wan 2.6 Be Open Source?
09:29 ByteDance Seedance 1.5 Pro Release
10:02 The Confusing Rollout (AI Video 3.5?)
10:17 How to Use Seedance 1.5 in CapCut
10:59 Seedance 1.5 Generation Examples
11:54 EgoX: Turning Movies into First-Person POV
12:47 How EgoX Works (Geometry-Guided)
13:30 Rapid Fire: GPT Image 1.5 & Hunyuan World
14:02 Meta SAM Audio & Kling 2.6 Updates
14:18 Outro
AI Video From Audio is BONKERS!Leonardos FREE & New Flow State is Wild!Kling IPO Signals a Voltron AI Video Model?Seedance 2.0 Has Released (kinda...)Is OpenAI About to KILL the Banana?Real Time Video Editing & Luma Ray 3.14 Deep Dive!Real-Time AI Video is Finally Here (And It’s Insane!)Kling 2.5 goes BEAST MODE for AI Video!ComfyUIs App Mode is the Easy Button Weve Been Waiting For!AI Video Just Broke Real-Time — What Happens Next Is Wild!They Killed Fable. Here’s What You Need to Do NextAI QUAKE Is Here & You Can Play It! Is it a GameChanger?
Theoretically Media |

AI Video From Audio is BONKERS!

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER