Using wav2vec2 for own usecase

I’m looking to use wav2vec2 english for own project use-case for the purpose of transcription, it is possible that I can feed large audio files (upto 1hr of duration) for the task?
Thanks in Advance!

Yes, you should be able to. Here are some references on fine-tuning wav2vec on common voice dataset. With a csv file having one column for location of audio clips and the other column the label, you can fine-tune it for your own audio files.


Thanks will definitely check this out!