Our paper about Canary has been accepted to INTERSPEECH 2025 and is now available on Arxiv. Check it out to learn how we trained one of the most powerful open-source ASR and speech translation models! arxiv.org/abs/2406.19674

It is, indeed, a lot of pressure 😆

black on (esp elderly) asian violence has exploded in oakland, and it makes me fucking enraged. it’s like this is my grandma here, and this young man is taking her stuff and punching her in the face if asians complaining about this is racist, where are the black voices

Meeting up with Boston #speech folks over BBQ and beer 🍻 (minus Jonathan Le Roux)

New York City is living in the future!!! Look at the Mayor announcing the city will now use trash bins instead of dumping their trash on the streets like barbarians.

🚨 🔔 Paper alert in Efficient Streaming ASR 🔔🚨 I'm super excited to release our paper “XLSR-Transducer: Streaming ASR for Self-Supervised Pretrained Models” abs: arxiv.org/abs/2407.04439 [1/n]

Rohit Mittal I think the most immigrant South Bay trait is counting you and all your friends net worths, cribbing about who has the bigger house, saying you could move to Virginia and get a mansion while having no real intent to do so

How to not be a gym douchebag: if there are only 2 benches available, don't occupy both of them for your "RDL setup".

I am 46 years old. Never lived in a state for 15+ yrs My father worked in the Indian Navy. Got posted all over the country. His kids don't deserve jobs in Karnataka? I build companies. Have created 25000+ jobs across India! My kids dont deserve jobs in their home city? Shame.

Starting today, open source is leading the way. Introducing Llama 3.1: Our most capable models yet. Today we’re releasing a collection of new Llama 3.1 models including our long awaited 405B. These models deliver improved reasoning capabilities, a larger 128K token context

Huge congrats to AI at Meta on the Llama 3.1 release! Few notes: Today, with the 405B model release, is the first time that a frontier-capability LLM is available to everyone to work with and build on. The model appears to be GPT-4 / Claude 3.5 Sonnet grade and the weights are

I keep seeing more and more spam posts on this app. Trying out Threads to see how that's like. Are any of y'all on it?

Every once in a while I find a thread on this website which makes all the spam bearable. Couldn't help sharing!

The #ICASSP2025 paper submission deadline is 9 September 2024. No new submissions will be accepted after this deadline. However, you may update the submitted paper until September 12. The title, author list and EDICS of the submitted paper cannot be changed.

Paper link: arxiv.org/abs/2409.08148

``M-BEST-RQ: A Multi-Channel Speech Foundation Model for Smart Glasses,'' Yufeng Yang, Desh Raj, Ju Lin, Niko Moritz, Junteng Jia, Gil Keren, Egor Lakomkin, Yiteng Huang, Jacob Donley, Jay Mahadeokar, Ozlem Kalinli, ift.tt/EsxaHFL

📣 You can now have a conversation with Meta AI using voice. It’s super fast, connected to the web, natural and conversational and even comes with celebrity voice options from Awkwafina, Kristen Bell, John Cena, and more. What voice speaks to you? (pun intended 😆)

rdesh26.bsky.social

📢 Introducing VERSA: our new open-source toolkit for speech & audio evaluation! - 80+ metrics in one unified interface - Flexible input support - Distributed evaluation with Slurm - ESPnet compatible Check out the details wavlab.org/activities/202… github.com/wavlab-speech/…