Off topic, but not sure if this level of audio classification task can be ever done with few-shot prompting a multi-modal model. Don't think GPT4V will be there

Off topic, but not sure if this level of audio classification task can be ever done with few-shot prompting a multi-modal model. Don't think GPT4V will be there yet, but food for thought.