Pad Input HuggingFace

I want to pad my Inputs to Finetune a Wav2Vec Modell, but i cant.

from transformers import Wav2Vec2Processor
model_name = 'maxidl/wav2vec2-large-xlsr-german'
processor = Wav2Vec2Processor.from_pretrained(model_name)

fake_batch = [{'input_values' : torch.rand([1, 112000])}, {'input_values' : torch.rand([1, 86000])}]
processor.pad(fake_batch, padding=True)

Is giving me

ValueError: Unable to create tensor, you should probably activate padding with 'padding=True' to have batched tensors with the same length.

Even tho i cleary set ‘padding=True’. Whad am i doing wrong?

Ty in advanced

It seems that Wav2Vec2FeatureExtractor.__call__ expects the padding argument, not the pad method, but I’m also not deeply familiar with all transformers calls, so you might want to post the question in the Huggingface forum. :wink: