Today in Generative Media

Media split on AI; AI musical made in 2 weeks; Text-to-speech emergent abilities

I will be taking a four-day long weekend! "I am currently off-world and will return on February 20th. For anything urgent, please contact my AI clone." See you all next week…

• Take the Cash or Fight? Media Moguls Split on AI Deals (Hollywood Reporter)

• Secret Level made this AI musical in two weeks as proof of the game-changing tech. (Ad Age)

• Largest text-to-speech AI model yet shows ’emergent abilities’ (TechCrunch)

• Stability AI tries to stay ahead of the pack with a new image-generating AI model (The Verge) What comes after Stable Diffusion? Stable Cascade could be Stability AI’s future text-to-image generative AI model (VentureBeat)

• How AI can make history: Large language models can do a lot of things. But can they write like an 18th-century fur trader? (The Verge)

• Polarr Next is a Web-Based, AI-Powered Photo Editor Made for Pros (PetaPixel)

• Solving the ‘Profile View’ Crisis in Facial Image Synthesis (MetaPhysic blog)

• Magic-Me: Identity-Specific Video Generation (GitHub)

• Masked Audio Generation using a Single Non-Autoregressive Transformer (project page)

• PRDP: Proximal Reward Difference Prediction for Large-Scale Reward Finetuning of Diffusion Models (project page)

• GhostWriter: Augmenting Collaborative Human-AI Writing Experiences Through Personalization and Agency (arXiv)

• Antagonistic AI (arXiv)

• L3GO: Language Agents with Chain-of-3D-Thoughts for Generating Unconventional Objects (arXiv)

• Like not to brag but this model runs 9x faster and 10x cheaper than OpenAI's Whisper API. That's kinda crazy (X)

• Whisper running on WatchOS! 🔥 (X)

• Generate any world you can imagine with RunwayML. Made completely with Text-to-Video. (X)