HN user
There is a fairly new ASIC named "Sohu" that is purpose-built for transformers. They have some bold claims that are impressive if true.
I found a short discussion[2] you may find useful.
[2] https://www.lesswrong.com/posts/qhpB9NjcCHjdNDsMG/new-fast-t...
Allowing users to register to your app using another service (e.g. Facebook, Spotify, Google) requires OAuth. That’s what I use authlib for, anyways. It simplifies the OAuth process so much that I don’t have to think about the authentication process when designing an app.
80x faster than Whisper is an incredible feat. How is Deepgram's transcription accuracy?
Also, have you heard of Conformer-1 by Assembly-AI[1]? It released a few days ago and supposedly scored higher than Whisper on various benchmarks.
Can you explain your process in more depth?
This is awesome and heavily resonates with me. Specifically this line:
Sometimes when I get depressed I stop caring for myself, but I'm always good about showing up for others.
What stage of the project are you on? How can I track or contribute? I would love to see this come to fruition.
It was an enjoyable and enlightening read. Now I’m thinking how I can rewrite my tile-based path-finding RL project to utilize more Tensors and less for loops.
Well done! I’m looking forward to reading how you concurrently ran 100 million mini snake games.