Unknown
Black Video References Help Preserve Voice Delivery in AI-Generated Scenes
An ElevenLabs workflow pairs Eleven v4 voice generation with Seedance 2.5 to address a common problem in AI filmmaking: a character’s voice and delivery changing between scenes. The tutorial’s creator builds and directs the performance with audio tags, then uses it as a reference for video generation. Three methods are demonstrated, with the creator reporting more consistent delivery from a black video reference and using a timestamped transcript to delay dialogue until after the action begins.
Housing Stability Depends on Affordability, Financing, and Resilience
Housing stability, the participants at Aspen Ideas: Economy argued, depends on more than whether a family rents or owns: it requires homes people can afford to stay in and the financial support to withstand shocks. Samir Goel of Esusu, Jeff Hall of the Anne and Henry Zarrow Foundation, and Kristen Scheyder of Citi Foundation made the case for pairing housing production and preservation with tools such as rent reporting, emergency savings and flexible financing. They cautioned that credit-building alone cannot prevent displacement or make housing affordable.
Micro Duck’s Walking Policy Trains in the Cloud, Then Returns to Local Simulation
The workflow uses NVIDIA RTX Spark for scene creation and policy evaluation, while cloud compute handles parallel training in Isaac Lab; the trained policy then returns to the local system before deployment to Micro Duck. The operator says Astra uses tools including FreeCAD, Blender and Isaac Sim to build the course and robot model, and that Isaac Lab trains thousands of Micro Ducks in parallel. The account shows the robot walking after deployment but reports no separate measurement of its real-world performance.
ChatGPT Voice Lets Users Interrupt, Revise Requests, and Resume Conversations
OpenAI presents the new ChatGPT Voice, powered by GPT Live, as a more fluid way to talk with an assistant: users can interrupt it, add requests before it answers, and revise what they want translated. Its demonstrations show those interactions in a word puzzle, travel planning and a Spanish message. They illustrate conversational flexibility, but do not establish how accurately or reliably the system performs beyond the scripted exchanges.
Build an AI Ad Around a Jingle That People Remember
ElevenLabs’ tutorial argues that the jingle should be settled before an AI-made ad’s visuals, because the product, music and edit need to work as one piece. It demonstrates a five-step workflow using a fictional chocolate bar: develop a simple product concept, write and choose a roughly 30-second jingle, create consistent product references, animate storyboard images and cut the clips to the music. The company also presents The Search, a competition that scores catchiness most heavily and offers $100,000 in prizes.
Culture Care Offers an Alternative to the Zero-Sum Logic of Culture War
Artist Makoto Fujimura argues that culture should be tended rather than treated as territory in a fight for power—a practice he calls culture care. In a conversation with documentary filmmaker Laura Waters Hinson, the two consider how art and film can make room for complexity. Hinson describes the challenge of making a documentary about missionary Elisabeth Elliot that neither reduces her to a hero nor turns her into an indictment.