from transformers import Wav2Vec2Model

model = Wav2Vec2Model.from_pretrained("output", from_flax=True)
model.save_pretrained("./")