Upload audio or a CSV of audio URLs and the model will predict whether each speaker is a native or non-native Arabic speaker.
Expected columns: dp_id, audio_url, nativity_status, language
dp_id, audio_url, nativity_status, language