Current Large Audio Language Models largely transcribe rather than listen(arxiv.org)4 points by earcar 85 days ago | 0 commentsNo comments yet