- The Renaissance Times
- Posts
- NVIDIA May Be Working On a New Audio Model
NVIDIA May Be Working On a New Audio Model
And The Demos Look Unbelievable
The Renaissance Times
April 08, 2024
Dear Artisan,
One of the most mind blowing demos and products out on the market right now is Suno.ai.
In just a few minutes, you can create a full-length song with just a one sentence description, spanning across any genre, sung by any voice, and belting out any lyrics you can dream up. It feels like magic.
Well, this weekend something dramatically better surfaced on X, leading people to wonder who released it and what the product specs look like. While all of that is still unknown, we have the demos, and they look incredible.
Let’s dive in.
Hot Off The Press
A New Text to Audio Model Surfaces, Rumored to be From NVIDIA
Spotify launches personalized AI playlists that you can build using prompts
Synchron, a competitor to Neuralink, will soon launch brain implant trials for stroke and MS patients
OpenAI transcribed over a million hours of YouTube videos to train GPT-4
TSMC Gets $11.6 Billion in US Grants, Loans for Three Chip Fabs
The One Big Thing
here's an exclusive new udio AI music generation 🫡
source: anon. anon is always the source.
— ʟᴇɢɪᴛ (@legit_rumors)
7:42 PM • Apr 7, 2024
Alright, watch this and tell me how anyone has a career as a generic “singer” anymore in the future.
Yes, the world will always have personalities and stars, but the ability to be able to create fantastical sound at the click of your fingers will change everything. This is as dramatic a lowering of the barriers to creation as they come.
You are about to have an entire ensemble and music director ready to perform the craziest things you can dream up.
While nobody knows who released these demos, the rumor is that they are 10x better than what Suno is providing now, which is the state of the art. The other rumor is that this model was developed by NVDIA.
We have written in the past about how the lines will blur between hardware and software companies in the world of AI, because your models and data are only as good as the number of computers you have to run them on.
As we continue to enter that world, it makes complete sense for NVIDIA to start releasing models of all kinds, seeing what works, and doubling down on the ones that stick with additional researchers and all the chips they could ever need.
I’m going to need better speakers.
The Gallery
This is pretty crazy…
This post was deleted. Seems like suno’s competitor is nivdia 👀 btw this sounds like 2pac 🔥🔥 @apples_jimmy good call again 🐐
— moonbi⭕ (@lifeafterAi_)
11:11 AM • Apr 7, 2024
Tools
Must have tools for every artisan to add to their toolkit:
Kopia: A tool to virtually try on clothes on photos and body measurements
Fluently: A tool to assist non-native English speakers in professional communication
Off/Script: An app for aggregating community ideas into tangible products
InstaShorts: A tool to generate personalized TikTok videos with voiceovers cloning
Aqua Voice: A tool for voice-driven text editing and document creation
Deep Tech
The newest and coolest in the research world that you need to know about:
Sigma: Siamese Mamba Network for Multi-Modal Semantic Segmentation
Linear Attention Sequence Parallelism (LASP): Sequence Parallel to leverage Linear Attention right-product feature
Generative Rendering: A framework that directly renders low-fidelity animated meshes to animations
Closing Thought
Will you be an AI rapper or classical artist?
Work With Us!
The AI Renaissance is coming and we are building the best community of the people making it happen.
Contact us to sponsor your product or brand and reach the exact audience for your needs across our newsletter and podcast network.