🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

TL;DR we use math to make synthetic speech do moar funny/expressive things. Listen to e.g. "That's right! shouted the Queen"; we're trying to get "Baseline" to sound more like "Reference": google.github.io/tacotron/publi…

Detecting deepfakes is one of the most important challenges ahead of us. Following our release of a synthetic audio dataset in Jan, we're releasing a large dataset of visual deepfakes to support researchers working on synthetic video detection #GoogleAI ai.googleblog.com/2019/09/contri…

Here's your hacking project for the weekend, folks: ai.googleblog.com/2019/09/contri…

If you train an end-to-end text-to-speech model whose attention fails to align, you can get hilarious descents into speech madness like this: google.github.io/tacotron/publi…. Sadly, Eric Battenberg has reduced the frequency that we'll get to cry with laughter in our office over these...

New research from my brilliant colleagues! At long last, direct-to-waveform Tacotron -- using multi-scale normalizing flows (read RJ Skerry-Ryan's thread if you have FOMO and want to learn what those words mean). Everyone's still smiling and kicking ass through pajamafromhome. 🚀

Google Research is excited to announce ... TACOSPAWN! It's a TTS model that can speak in human voices that don't exist: google.github.io/tacotron/publi… We learn a prior over the speaker embedding space of a many-speaker Tacotron: arxiv.org/abs/2111.05095

Yes, my team is actually actually hiring! If you love getting into the weeds of generative modeling (RS) or build machine learning models like a ninja (SWE), reach out! 🌠

New work from my fantastic colleagues Soroosh and Matt (they even share an umbrella)!

If you haven't listened to Lex's June 30 interview with George Hotz (and especially if you think you hate Hotz), do. It was very good and very entertaining. I recommend audio over video: lexfridman.com/george-hotz-3-…

We're excited to announce the Climate Change AI Summer School 2024!☀️ Are you an #AI expert who wants to tackle #ClimateChange? Are you a #climate expert trying to use #ML in your work? Are you curious about the topic? Apply & register now! Learn more👉climatechange.ai/events/summer_…

Very excited to share a side project that we're launching in time for the November election: openballot.app

I'm hosting Astral Codex X's San Francisco ballot party on October 19. Come join our engaging discussion: lu.ma/0ufpys63

We're thrilled to open sign-ups for OpenBallot in time for the November San Francisco election! See 40+ voter guides in one place, and use our tool to fill out your ballot. You can keep it private or share with friends: openballot.app/guides

Want a more bike-friendly SF? 🚲 Check out SF Bicycle Coalition's voter guide on OpenBallot along with 40+ others to help shape the city's future! 🗳️🌉 openballot.app/guides/9cba2a3…

Eric is the powerhouse that got RNN-based TTS models to generalize to utterance lengths not seen in training via Dynamic Convolution Attention (DCA). Now he's at it again with Transformer-based models, using interpolated relative position biases in a novel alignment mechanism 💪

Wow, google open source some tts models github.com/google/sequenc…