Pinned
Bark Text-to-Audio Model
Full Text Input: "Why was six afraid of seven?"
Ignore Bark's "I'm done with this input" token and tell Bark to just keep generating more audio anyway.
00:00
Jonathan Fly 👾
7,427 posts
Jonathan Fly 👾
@jonathanfly
CEO of bad ideas. Using the wrong tool. The least efficient way. For no good reason. 👾
suno.com/@jonathanfly
youtube.com/c/JonathanFly
Joined April 2009
- Trying out Wav2Lip. It will happily try to lip lync sync a constantly morphing face.
00:00 
GIF- 'Pair Programming' with GPT-2 trained on 2 million GitHub files. Straightforward near-term practical use of deep learning. I wish they had better examples - for example, I bet if you name a variable Gryffindor it will suggest Hufflepuff for the next one! tabnine.com/blog/deep
00:00 - Trying depth-mapping on abstract art. I don't know what you would call 'working properly' here, it's just fascinating to see a neural network try to interpret a Jackson Pollock painting as a 3D landscape. (Convergence, 1952)
00:00 - Luma's start and end keyframes are a game changer. With a sequence of keyframes from the original film, we can seamlessly remaster stop motion classics like "Jason and the Argonauts" as modern single-take action scenes.
00:00 - NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis Input is images with known camera poses. Can't wait to disable all the sanity checks and see what this renders if I give it impossible geometry. abs: arxiv.org/abs/2003.08934 site: matthewtancik.com/nerf
00:00 - Replying to @tg_bomzeI love that your Tweet turned into a giant collaboration thread with everyone posting their own hilarious upscales. My favorite so far is this 😅 guy. (If you want to get similarly creative interpretations keep the steps low.)
- Upscaling from 16x16 pixels all the way up 4k with 3 Face Super Resolution models chained together. Genuinely impressed by how this one came out. 🙋♀️
- "Once trained on a set of videos of the same category (e.g. faces), our method can be applied to any object of this class." Well you can *apply* the method to anything. Like the Avatar the Last Airbender Intro. Sort of AI human face interpretive dance. (watch past 9 seconds)
00:00 - The good news is StarGAN v2 code is out. The better news is if you put a human into the animal model you do in fact get out a feline version of the human, and it's even wearing a suit. github.com/clovaai/starga… h/t @ak92501
- I am so excited for when @facebookai drops Vid2Game code. Imagine all the amazing things you could do with this. Maybe the models are even playable in a Colab? FMV games: The Next Generation? abs: arxiv.org/abs/1904.08379 vid: youtube.com/watch?v=GprHGp…
00:00
