What is Whisper JAX?
Whisper JAX is an optimised implementation of the Whisper model by OpenAI. It runs on JAX with a TPU v4-8 in the backend. Compared to PyTorch on an A100 GPU, it is over 70x faster, making it the fastest Whisper API available.
Key features and advantages include:
Fast performance: Over 70x faster than PyTorch on an A100 GPU
Optimized implementation: Built on JAX with a TPU v4-8 for maximum efficiency
Accurate transcription: Provides accurate transcription of audio files
Progress bar: Displays progress of transcription through a progress bar
Create your own inference endpoint: To skip the queue, users can create their own inference endpoint using the Whisper JAX repository.