How to use openai/whisper-base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="openai/whisper-base")
# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("openai/whisper-base") model = AutoModelForSpeechSeq2Seq.from_pretrained("openai/whisper-base")
Fixes #16, which showed that one of the merges is missing, which creates different outputs. The issue with this is that finetuned models were trained using this.
@sanchit-gandhi if you can merge this one!
Thanks for the fix @ArthurZ !
· Sign up or log in to comment
