{
    "encoder_dim": 64,
    "encoder_rates": [3, 3, 7, 7],
    "decoder_dim": 1536,
    "decoder_rates": [7, 7, 3, 3],
    "codebook_size": 4096,
    "codebook_dim": 8,
    "vq_strides": [8, 4, 2, 1],
    "noise": true,
    "depthwise": true
}