JAX implementation of OpenAI's Whisper model for up to 70x speed-up on TPU.
updated at May 26, 2024, 3:57 a.m.
42 +1
4,138 +10
352 +3