Introducing the Common Voice-based Speech-to-Speech translation corpus, CVSS, which includes 2657 hours of speech-to-speech translation sentence pairs from 21 languages into English. Read about its development and how we used it to train baseline models ↓ goo.gle/3K6nC0L

Learn how a proof-of-concept visually-driven text-to-speech model, called VDTTS, generates speech that matches the exact timing of mouth movements in a video and can be used to replace audio that was recorded in noisy conditions. Read more ↓ goo.gle/37tUkuo

Task-oriented dialogue models power many modern conversational agents, but can be limited in their ability to generalize to new tasks. Check out two new approaches for dialogue modeling that use additional context during training for improved performance. goo.gle/3KE5SKv

Access 100+ image, video & audio datasets in seconds with one line of code & stream them while training ML models with Activeloop Hub (more at docs.activeloop.ai, description & links in the comments below) reddit.com/r/MachineLearn…

app to play with latent diffusion models reddit.com/r/MachineLearn…

📌4-5 May Montréal & online: Join the global #artificialintelligence community for 2 days of heated discussions on #AIpolicy #AIforGood #AIethics #AIRegulation and more! 👉americas.worldsummit.ai 💬Contact us to get the code for a 20% discount on the admission fee: [email protected]

🌟BIG catalog of online courses for #DataScientists & other Data Professionals via tut_ml ⬇️ #DataScience #Python / R / SQL Coding Libraries #MachineLearning #DeepLearning #AI #Statistics & Math #DataEngineering #BigData #Analytics #DataViz #100DaysOfCode mltut.com/best-online-co…

[Online eBook] Introduction to #DataScience — Learn #JuliaLang Programming, Math & Data Science from Scratch: datascience-book.gitlab.io/book.html#_pre… by Karthikeyan A K 🐧 ———— #BigData #DataScientists #AI #MachineLearning #Coding #JULIA #100DaysOfCode

Le 17 mai, en ouverture de #MobiliTAI, ne manquez pas la conférence de Lionel Briand, de l’Université d'Ottawa | University of Ottawa. MobiliTAI, c’est 3 jours de conférences et de discussions autour des usages de l’#IA dans la sphère de la #mobilité. Programme et inscription > mobilit.ai.

Fresh from Google research - a new #madeWithTFJS ML model is out! 📸 Check our new Portrait Depth API for real time portrait depth estimation and use it to generate 3D photos, AR effects, image relighting and more. → goo.gle/3M1Am9Z

J.P.Morgan’s Comprehensive 280-page Guide to #MachineLearning (with #Python code for popular #ML #algorithms) bit.ly/3LlDyfH [huge PDF] ———— #BigData #DataScience #AI #fintech #finserv #DigitalTransformation #DataStrategy #AIStrategy #coding #DataScientist #100DaysOfCode

TorchRL: PyTorch pre-release RL library is here! reddit.com/r/MachineLearn…

Read about new technology on #GoogleAssistant that enables more natural interactions by detecting if a query is related to previous context and, if so, rephrasing it internally to fill in missing information before providing an answer. goo.gle/3Ps0NI0

85+ Data Science Projects You Can Try with #Python python.plainenglish.io/85-data-scienc…

A unique ecosystem—Scale AI created a new Canadian AI ecosystem with more than 567 organizations actively participating in projects and investing in AI Supply Chain. #ArtificalIntelligence #SupplyChain #Retail #manufacturing #ISED #AI

What are some amazing machine learning projects to impress the recruiter? reddit.com/r/MachineLearn…

#AI For Beginners from Microsoft - 12 Weeks, 24 Lessons, AI for All! github.com/microsoft/ai-f…

The CIFAR Mila - Institut québécois d'IA and IVADO health symposium from 2021 fall lead to CIFAR Mitacs and Mila - Institut québécois d'IA announcing catalyst grant funding for 4 collaborative #research projects using synthetic #data & #AI to preserve patient privacy: ivado.ca/en/2022/05/30/…

📣 Introducing Coral Dev Board Micro. Hear the latest updates from Google Coral for developing AI/ML solutions at the edge. Watch the #GoogleIO session → goo.gle/3w9Okko

Just launched - nebulgym, a new open-source that accelerates AI training (~1.5-2x as of now) in a few lines of code without requiring you to change your training setup reddit.com/r/MachineLearn…