How to use the wav2vec2-large-TIMIT-IPA2 model?

I need to use the model to convert speech to International Phonetic Alphabet. I don’t know how can I pass in the audio file and get the string output:

from transformers import AutoProcessor, AutoModelForCTC

processor = AutoProcessor.from_pretrained("speech31/wav2vec2-large-TIMIT-IPA2")

model = AutoModelForCTC.from_pretrained("speech31/wav2vec2-large-TIMIT-IPA2")

audio_file = "./male-voice.wav"

ipa_output = # what should I do next?


Thank you for your help!