those recent, powerful music generation models make me wonder why / to what extent we need music source separation.

Excited to share SingSong, a system which can generate instrumental accompaniments to pair with input vocals! 📄arxiv.org/abs/2301.12662 🔊g.co/magenta/singso… Work co-led by myself, Antoine Caillon, and Adam Roberts as part of Google Magenta Project and the broader MusicLM project 🧵

Deadline extension for for the Call For Paper for the next Web Audio Special Issue of the Journal of the Audio Engineering Society (JAES). Deadline is now March 15, 2023. tinyurl.com/4ufdmt5n

If a person in some parts of China crosses the street in the improper location, the facial recognition technology will immediately send them to a public shame board, connect them to their bank card, and deduct the amount of the fine

Bing Chat is underrated. Most people don't know what it’s capable of, how it differs from ChatGPT and what kind of amazing content it can generate. 🚀 8 compelling ways to talk to The New Bing: 🧵👇

Might be the most impressive version of this yet

We've released the Koala into the wild🐨 The Koala is a chatbot finetuned from LLaMA that is specifically optimized for high-quality chat capabilities, using some tricky data sourcing. Our blog post: bair.berkeley.edu/blog/2023/04/0… Web demo: koala.lmsys.org

One of the most impressive CV works I've seen recently. Also huge kudos to Meta AI for sticking to open sourcing despite the trend increasingly going towards the opposite direction.

[LIVE] Forum de l'Alternance | Métiers du numérique Les entretiens vont bon train ! Beaucoup d'opportunités pour nos étudiants ! #alternance #numérique Master Stratégie Digitale Junior MIAGE Concept Université Côte d'Azur Sopra Steria France Monaco Telecom

30 years of web browser market disruption #CES2024

The Diffusion Transformer paper, by my former-FAIR-and-current-NYU colleague Saining Xie and former-Berkeley-student-and-current-OpenAI engineer William Peebles, was rejected from CVR2023 for "lack of novelty", accepted at ICCV2023, and apparently forms the basis for Sora.

The YOLO-World YouTube tutorial is out! please, let us know what you think! - model architecture - processing images and video in Colab - prompt engineering and detection refinement - pros and cons of the model watch here: youtube.com/watch?v=X7gKBG… ↓ more resources

If you want to use YOLO-World in your own project, you can do so through the Roboflow Inference Python package. github: github.com/roboflow/infer…

Robot factory in China 🇨🇳

Creating, capturing and displaying immersive 3D scenes. Talk by Adrien Boisseau and George Drettakis from the INRIA Graphdeco team. At the 3rd Journée d'études #XR2C2 polyteh@Univ_CotedAzur Centre Inria d'Université Côte d'Azur Laboratoire i3S

Musical metaverse: opportunities and challenges. Talk by Luca Turchet from Università di Trento at the 3rd Journée d'études #XR2C2 Université Côte d'Azur Laboratoire i3S

Informed virtual environments and adaptive sensorial feedback. Talk by Indira Thouvenin of @utcompiegne at the 3rd Journée d'études #XR2C2. Université Côte d'Azur Laboratoire i3S

ViAjeRo: XR and passengers of the future. Talk by Stephen Brewster from University of Glasgow at the 3rd Journée d'études #XR2C2. Université Côte d'Azur Laboratoire i3S

Computational models for natural visual perception and interaction in XR. Talk by Fabio Solari from Università di Genova at the 3rd Journée d'études #XR2C2. Université Côte d'Azur Laboratoire i3S