32k:
    filter_length: 513,
    a?: 32,
    inter_channels: 192,
    hidden_channels: 192,
    filter_channels: 768,
    n_heads: 2,
    kernen_layersl_size: 6,
    kernel_size: 3,
    p_dropout: 0,
    resblock: "1",
    resblock_kernel_sizes: [3, 7, 11],
    resblock_dilation_sizes: [[1, 3, 5], [1, 3, 5], [1, 3, 5]],
    upsample_rates:  [10, 4, 2, 2, 2],
    upsample_initial_channel: 512,
    upsample_kernel_sizes: [16, 16, 4, 4, 4],
    spk_embed_dim: 109,
    gin_channels: 256,
    sampling_rate: 32000,
48k:
    filter_length: 1025,
    a?: 32,
    inter_channels: 192,
    hidden_channels: 192,
    filter_channels: 768,
    n_heads: 2,
    kernen_layersl_size: 6,
    kernel_size: 3,
    p_dropout: 0,
    resblock: "1",
    resblock_kernel_sizes: [3, 7, 11],
    resblock_dilation_sizes: [[1, 3, 5], [1, 3, 5], [1, 3, 5]],
    upsample_rates: [12, 10, 2, 2],
    upsample_initial_channel: 512,
    upsample_kernel_sizes: [24, 20, 4, 4],
    spk_embed_dim: 109,
    gin_channels: 256,
    sampling_rate: 48000,