{ "model_type": "autoencoder", "sample_size": 32768, "sample_rate": 16000, "audio_channels": 1, "model": { "encoder": { "type": "dac", "config": { "in_channels": 1, "latent_dim": 128, "d_model": 128, "strides": [4, 4, 5, 5] } }, "decoder": { "type": "bigvgan", "config": { "out_channels": 1, "latent_dim": 64, "resblock": "1", "upsample_rates": [5, 5, 4, 4], "upsample_kernel_sizes": [10, 10, 8, 8], "upsample_initial_channel": 512, "resblock_kernel_sizes": [3, 7, 11], "resblock_dilation_sizes": [[1, 3, 5], [1, 3, 5], [1, 3, 5]], "activation": "snakebeta", "snake_logscale": true } }, "bottleneck": { "type": "vae" }, "latent_dim": 64, "downsampling_ratio": 400, "io_channels": 1 } }