chunked/nltk_chunking/model_doc_clap/chunk_13.txt

In audio classification tasks, the model achieves state-of-the-art performance in the zeroshot setting and is able to obtain performance comparable to models' results in the non-zero-shot setting.