{ "sampling_rate": 32000, "encoder_dim": 64, "encoder_rates": [2, 3, 8, 8], "decoder_dim": 1536, "decoder_rates": [8, 8, 3, 2], "attn_window_size": 32, "codebook_size": 4096, "codebook_dim": 8, "vq_strides": [8, 4, 2, 1], "noise": true, "depthwise": true }