audio_aqa/sample_data.jsonl

{"audio": ["../example_data/audio_aqa/test_1.wav"], "conversation": [{"from": "human", "value": "Transcribe the audio into Chinese." }]}
{"audio": ["../example_data/audio_aqa/test_2.wav"], "conversation": [{"from": "human", "value": "Describe the sound in this audio clip." }]}