dataset_config.json

{
    "training_datasets": {
        "language-and-voice-lab/samromur_asr": {
            "splits": [
                "train",
            ],
            "dataset_config_name": None,
            "text_column_name": "normalized_text",
        },
        "language-and-voice-lab/althingi_asr": {
            "splits": ["train", "validation"],
            "dataset_config_name": None,
            "text_column_name": "normalized_text",
        },
        "language-and-voice-lab/malromur_asr": {
            "splits": ["train", "validation"],
            "dataset_config_name": None,
            "text_column_name": "normalized_text",
        },
        "language-and-voice-lab/samromur_children": {
            "splits": ["train", "validation"],
            "dataset_config_name": None,
            "text_column_name": "normalized_text",
        },
        "language-and-voice-lab/raddromur_asr": {
            "splits": ["train", "validation"],
            "dataset_config_name": None,
            "text_column_name": "normalized_text",
        },
    }
}