Fast, local, low-power machine learning models that do speech transcriptions on mobile phones mess up a lot of existing business models but unlocks all sorts of value for the world.
We have open-sourced wav2letter@anywhere, an inference framework for online speech recognition that delivers state-of-the-art performance. ai.facebook.com/blog/online-sp…
Facebook AI has released Libri-light, the largest open source data set for speech recognition to date. This new benchmark helps researchers pretrain acoustic models to understand speech, with few to no labeled examples. ai.facebook.com/blog/a-new-ope…
Just released the first version of at16k - at16k.com
Open source, production quality models for speech to text. Trained with @TensorFlow and @googlecloud TPUs. Gotta say that the TPUs are really really fast. 7 days of training down to 12 hours!
151K Followers 1K FollowingWork for yourself and make $10k/mo, from wherever, whenever 🤝
• Subscribe: https://t.co/BuyZXNWzZC
• Sponsor: https://t.co/3XH0Vfet1Q
5.3M Followers 291 FollowingStarted & runs 37signals (makers of Basecamp, HEY, and ONCE). Non-serial entrepreneur, serial author. DM or email me at [email protected].
943K Followers 498 FollowingFather of three, Creator of Ruby on Rails + Omarchy, Co-owner & CTO of 37signals, Shopify director, NYT best-selling author, and Le Mans 24h class-winner.
481K Followers 854 Following"The Writing Guy" | Christian | Host: How I Write Podcast | I tweet about writing, beauty, and architecture | My writing: https://t.co/SOE9HtxXdi