It's actually insane how good Midjourney creators are getting. You can see more of the creator's work here : http://
t.me/jmneurogallery
/2071
…
MULTIMODAL AI
-
Midjourney creators showcase exceptional generative AI artwork
By
–
-
DeepFloyd IF: Open-source text-to-image model with zero-shot inpainting
By
–
DeepFloyd IF: An open-source text-to-image model by our @DeepfloydAI team @StabilityAI
— hardmaru (@hardmaru) 28 avril 2023
Check out the examples, with amazing zero-shot inpainting results at high resolution: https://t.co/9TbHDzyGOy pic.twitter.com/YjIAyROCpWDeepFloyd IF: An open-source text-to-image model by our @DeepfloydAI team @StabilityAI Check out the examples, with amazing zero-shot inpainting results at high resolution: https://
github.com/deep-floyd/IF -
Google Photos Magic Eraser Expands to More Users
By
–
ICYMI: Google Photos Magic Eraser is rolling out to more users. It can automatically detect and suggest elements to erase on your photos to be removed with one tap.
-
FaceTime with ChatGPT: Voice Interaction Experience Enhancement
By
–
FaceTime with ChatGPT https://t.co/52qfpzChrX
— Andrej Karpathy (@karpathy) 27 avril 2023
Fun, qualitatively different experience over "texting"
Ty @CallAnnieAI for shoutout to my tweet a ~year ago pitching this as an idea https://t.co/LE6dYcDsxZ
With improvements to personality, latency, ASR/TTS could be magical ✨ https://t.co/eMeYzyum5TFaceTime with ChatGPT https://
callannie.ai
Fun, qualitatively different experience over "texting" Ty @CallAnnieAI for shoutout to my tweet a ~year ago pitching this as an idea https://
x.com/CallAnnieAI/st
atus/1651379993413558272
… With improvements to personality, latency, ASR/TTS could be magical -
GPT-4 Leads in Computational Resources and Multimodal Capabilities
By
–
GPT-4 is in a class by itself with respect to both computation used to train (FLOPs) and it being multimodal generative #AI @OurWorldInData https://
ourworldindata.org/artificial-int
elligence
… -
Microsoft NaturalSpeech 2 Advances Zero-Shot Speech Synthesis
By
–
Microsoft’s NaturalSpeech 2 Outperforms Previous TTS Systems in Zero-Shot Speech and Singing Synthesis https://
syncedreview.com/2023/04/26/mic
rosofts-naturalspeech-2-outperforms-previous-tts-systems-in-zero-shot-speech-and-singing-synthesis/
… -
API Whisper pour transcription vocale en temps réel
By
–
Whisper API
— AI Breakfast (@AiBreakfast) 27 avril 2023
Near-real-time speech-to-text transcription allows you to speak to other characters, and experience in VRhttps://t.co/hV8GO5vvVnWhisper API Near-real-time speech-to-text transcription allows you to speak to other characters, and experience in VR
-
AI Reconstructs Images from Human Brain Activity Using fMRI
By
–
Reading Minds with AI: Researchers Translate Brain Waves to Images: Two researchers from Osaka University were able to reconstruct highly accurate images from human brain activity obtained by fMRI. Read this article if you are curious to find out what… https://
kdnuggets.com/2023/03/readin
g-minds-ai-researchers-translate-brain-waves-images.html?utm_source=dlvr.it&utm_medium=twitter&utm_campaign=reading-minds-with-ai-researchers-translate-brain-waves-to-images
… -
REMEDIS: Self-Supervised Learning Framework for Medical Imaging
By
–
Learn about REMEDIS, a unified large-scale self-supervised learning framework for building foundation medical imaging models that shows significant improvement in data-efficient generalization across medical imaging tasks and modalities. Check it out → https://t.co/65GL7hR6k0 pic.twitter.com/OJRgedDd0P
— Google AI (@GoogleAI) 26 avril 2023Learn about REMEDIS, a unified large-scale self-supervised learning framework for building foundation medical imaging models that shows significant improvement in data-efficient generalization across medical imaging tasks and modalities. Check it out → https://
goo.gle/41HmauE
