We are releasing pretrained HuBERT speech representation models and code for recognition and generation. By alternating clustering and prediction steps, HuBERT learns to invent discrete tokens representing continuous spoken input. Learn more: ai.facebook.com/blog/hubert-se…

Lots of interest at NeurIPS Conference in VAEs today. The latest FnT Machine Learning review Dynamical Variational Autoencoders by Laurent Girin, Simon Leglaive, XIAOYU BIE, Julien Diard, Thomas Hueber and Xavier Alameda-Pineda (ell/he/lui) is available for free download from nowpublishers.com/conference/Neu… (1/2)👉

Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cut abs: arxiv.org/abs/2202.11539 project page: m-psi.fr/Papers/TokenCu…

[#PhD 👨‍🎓] Congratulations to XIAOYU BIE (RobotLearn Research Team @ Inria Grenoble team) for his thesis defense "Dynamical Variational Autoencoders for Multimedia Processing" #robotics 👏 ➕ More informations : theses.fr/s225890 inria.fr/en/robotlearn

🚀 Excited to announce a #PhD opportunity in #DeepLearning for #ConversationalAI! Details here: drive.google.com/file/d/1_y5UK9… The student will be enrolled at Concordia University and affiliated with Mila - Institut québécois d'IA. Get ready for open #AI #research! #SpeechBrain #MachineLearning #opensource

Just 1⃣ line to improve your Classifier-Free Guidance (CFG)! Should definitely consider it if you are using CFG.

At the end of the month, Overleaf won't allow adding co-authors via link sharing any longer without a paid account. It will in fact also roll back on current projects

I'm excited to introduce the Source-Disentangled Neural Audio Codec (SD-Codec), a new codec model that can disentangle arbitrary audio sources into distinct latent codes for speech, music, and SFX. Check our paper below 👇 Paper: arxiv.org/abs/2409.11228

You can also check our demo page: xiaoyubie1994.github.io/sdcodec/

Curious to know a little bit more but don’t have enough time (yet!) for the detailed paper? Our blog summarizes the Movie Gen Audio work with video links, challenges, model details, and results: apoorv2904.github.io/moviegen-audio/

Our LLaMA Speech team at Meta GenAI is hiring Research Scientist Interns this year! Our research interest spans various aspects of speech LLMs and GenAI for audio/speech. If you are interested in, please DM with CV.

📢 Join our Conversational AI Reading Group! 📅 Thursday, March 27 | 11 AM - 12 PM EST 🎙Speaker: Xiaoyu Bie(XIAOYU BIE ) from Télécom Paris 📖 Topic: "Learning Source Disentanglement in Neural Audio Codec" 🔗 Details: (poonehmousavi.github.io/rg)