Speech models tend to do well on clean read speech but fall apart on real-world Indian audio.
This Thursday, I'm joining @abhigyan_r, @sehaj__virk and Dhruv Rathi for a deep dive on Saaras V3, Sarvam's speech-to-text model built for Indian languages. I'll be focusing on the modelling side, especially how we handle long form multi-speaker ASR
Join us on Thursday, 21st May | 5:00 PM IST | Online
Register here: https://t.co/18jV3lMI3Y
You’ve curated the sources. You’ve researched everything. You know exactly what you want to say.
You just can't get it out of your head and onto the page. We are building Almanac exactly for this.
Experience the beta version here: https://t.co/F7Q7Pfgz7y
Here are a few things you can do with Almanac👇
Honored to have visionaries @NandanNilekani , @abhish18 , and @pramodkvarma share their thoughts at our product launch! Their insights sparked imagination and inspired our way forward.
@dalmiaman It's difficult to admit this... The pandemic definitely makes you see the world and your priorities differently. Great job writing this piece 👍
It's been a difficult night for us, exhausted after spending hours in bomb shelters, but waking up to so many messages of support from you guys helps.
Thank you ♥️.
We are excited to introduce the 15 #PPML enthusiasts
that will form the cohort of the first OpenMined #Beta#Bootcamp
Congratulations!
Read more on our blog: https://t.co/RheUOcXnso
Join our amazing community 👉 https://t.co/VLqVSAQV2B