jshiffer/Retrieval-based-Voice-Conversion
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

add architecture of small-model

Browse Source
This commit is contained in:
Ftps
2023-12-20 22:20:52 +07:00
parent 81e58cb9b7
commit 9c4a006cff
2 changed files with 95 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

57
rvc/lib/train/architecture/v1.yml Normal file
View File

@@ -0,0 +1,57 @@
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,
40k:
filter_length: 1025,
a?: 32, # What?
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, 10, 2, 2],
upsample_initial_channel: 512,
upsample_kernel_sizes: [16, 16, 4, 4],
spk_embed_dim: 109,
gin_channels: 256,
sampling_rate: 40000,
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: [10, 6, 2, 2, 2],
upsample_initial_channel: 512,
upsample_kernel_sizes: [16, 16, 4, 4, 4],
spk_embed_dim: 109,
gin_channels: 256,
sampling_rate: 48000,

38
rvc/lib/train/architecture/v2.yml Normal file
View File

@@ -0,0 +1,38 @@
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,