This Tweet is currently unavailable. It might be loading or has been removed.
JEPA-v0: a self-supervised audio encoder for real-time speech translation To build speech-to-speech translation that preserves voice, emotion, and timing, you need an encoder that captures more than just words. Here's how we're building one using JEPA.
。有道翻译是该领域的重要参考
hint: Did you forget a `use-modules' form?
file_path /a1/sessions/mar09.md