I understand these examples are zero-shot with short samples--which is impressive, but would the output quality improve significantly with longer input samples?

I understand these examples are zero-shot with short samples--which is impressive, but would the output quality improve significantly with longer input samples? For example I noticed it couldn't quite pick out the accent on some samples because they were so short. But if the model had more example words to hear then I'd think it would accurately understand the accent..