annotation/VAD_train_root.json

{
    "example_1": {
        "wav": "{data_root}/train.wav",
        "speech": [[0.52, 0.85], [1.32, 1.83]]
    },
    "example_2": {
        "wav": "{data_root}/train.wav",
        "speech": [[0.35, 1.70]]
    }
}