Nvidia Patents an AI Audio Model That Restores and Separates Speech Without Labeled Training Data
Teaching an AI to clean up audio is hard partly because you need huge amounts of labeled examples showing 'dirty' audio next…
Every newly published filing from Apple, Google, Microsoft, Meta, Nvidia, Tesla, OpenAI, and Amazon — broken down in plain English. Sorted by publication date, newest first.
Teaching an AI to clean up audio is hard partly because you need huge amounts of labeled examples showing 'dirty' audio next…
Nvidia is patenting a smarter way to compress video: scan the footage once to build a plan, then scan it again to…
Nvidia is patenting a system that lets a neural network look at ordinary flat images and reconstruct the three-dimensional shape of objects…
Most AI models panic when data goes missing. Nvidia is patenting an approach that turns the size of the gap itself into…
Most AI tools that let you ask questions about a video only look at the frames. Nvidia's new patent describes a system…
Checking images for defects is tedious, manual work. Nvidia's new patent hands that job to an AI that can not only find…
Rendering realistic surfaces in 3D scenes, from scratched metal to wet concrete, is computationally expensive. Nvidia's new patent describes a way to…
Nvidia is patenting a general-purpose AI model that can clean up, restore, or isolate voices in an audio signal, and it's designed…
Nvidia has patented a processor-level system that uses neural networks to generate images from noise values, the same core idea behind modern…
Before an AI can generate a picture, it has to learn how to compress one. Nvidia's latest patent attacks that compression step…
Plain English, intelligent commentary, no hype.