Rendered at 21:59:58 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
orsorna 19 hours ago [-]
For anyone who was piqued by the title but couldn't stomach the unedited LLM blogpost, you can read the similarity AI-genned docs for Laya directly: https://github.com/NandhaKishorM/laya
ozozozd 18 hours ago [-]
Man, it shows that someone was getting paid by the word.
Thank you!
CROON_tv 10 hours ago [-]
Similar result in production for me. I compared Jev with a general LLM (DeepSeek) as the decider for when to cut live speech into sentences, on the same 226 segments. With Jev the longest held segment was 17 s. With DeepSeek it was 40 s.
mkrishnan 2 hours ago [-]
I tried both with a synthetic dataset (1000 items), Jev scored 98% vs Laya 15%.
Thank you!