{ | |
"filter_channels": 768, | |
"gin_channels": 256, | |
"hidden_channels": 192, | |
"inter_channels": 192, | |
"kernel_size": 3, | |
"n_heads": 2, | |
"n_layers": 6, | |
"p_dropout": 0.0, | |
"resblock": "1", | |
"resblock_dilation_sizes": [ | |
[ | |
1, | |
3, | |
5 | |
], | |
[ | |
1, | |
3, | |
5 | |
], | |
[ | |
1, | |
3, | |
5 | |
] | |
], | |
"resblock_kernel_sizes": [ | |
3, | |
7, | |
11 | |
], | |
"segment_size": 36, | |
"spec_channels": 1025, | |
"spk_embed_dim": 109, | |
"sr": 48000, | |
"upsample_initial_channel": 512, | |
"upsample_kernel_sizes": [ | |
24, | |
20, | |
4, | |
4 | |
], | |
"upsample_rates": [ | |
12, | |
10, | |
2, | |
2 | |
] | |
} |