Hello everyone.
I am having an issue with the vocabulary size. as shown in the screenshot below. Can anyone help me why is this happening?
The following is my line of code:
from transformers import Wav2Vec2ProcessorWithLM
processor = Wav2Vec2ProcessorWithLM.from_pretrained(“BakhtUllah123/xlsr_ur_training”)
logits.shape
“.”.join(sorted(processor.tokenizer.get_vocab()))
transcription = processor.batch_decode (logits.numpy()).text