Hugging Face's logo

mlx-community
/
kitten-tts-micro-0.8-4bit

Text-to-Speech
MLX
Safetensors
kitten_tts
kitten-tts
4-bit precision
Model card Files
xet
 Community
Reza2kn commited on
Commit
0c2bf85
·
verified ·
1 Parent(s): 375dc6c

Add MLX 4-bit quantized model

Browse files
Files changed (5) hide show
  1. README.md +38 -0
  2. config.json +228 -0
  3. model.safetensors +3 -0
  4. model.safetensors.index.json +634 -0
  5. voices.npz +3 -0
README.md ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - mlx
4
+ - text-to-speech
5
+ - kitten-tts
6
+ ---
7
+
8
+ # mlx-community/kitten-tts-micro-0.8-4bit
9
+
10
+ This is the **INT4 (MLX 4-bit)** MLX conversion of `KittenML/kitten-tts-micro-0.8`.
11
+
12
+ ## Usage
13
+
14
+ ```bash
15
+ pip install -U mlx-audio
16
+ ```
17
+
18
+ ```bash
19
+ python -m mlx_audio.tts.generate --model mlx-community/kitten-tts-micro-0.8-4bit --text "This is a local MLX test voice." --voice "expr-voice-5-m"
20
+ ```
21
+
22
+ ## Inference Notes
23
+
24
+ The MLX implementation includes small end-of-utterance smoothing to prevent abrupt cutoffs.
25
+ You can override it with `fade_out_ms=0` and `tail_silence_ms=0` in `Model.generate()`.
26
+
27
+ ## Conversion Notes / Fixes
28
+
29
+ - AdaIN `fc.weight` orientation was corrected (ONNX stores as `(in, out)` even when square).
30
+ - AdaIN Snake `alpha` parameters are loaded and used for generator resblocks.
31
+ - ConvTranspose output padding matches the original (right-side pad for `output_padding=1`).
32
+ - Phase slice is passed through `sin` before ISTFT, matching the ONNX graph.
33
+ - ISTFT uses normalized windowing without phase unwrap (to match original behavior).
34
+ - Tail trim + dynamic fade-out + tail silence are applied at inference time to avoid a trailing spurt.
35
+
36
+ ## Original Model
37
+
38
+ Refer to the original model card for details: https://huggingface.co/KittenML/kitten-tts-micro-0.8
config.json ADDED
@@ -0,0 +1,228 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_quant_modules": [
3
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.key",
4
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.query",
5
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.value",
6
+ "bert.encoder.embedding_hidden_mapping_in",
7
+ "bert_encoder",
8
+ "decoder.asr_res.0",
9
+ "decoder.decode.0.conv1",
10
+ "decoder.decode.0.conv1x1",
11
+ "decoder.decode.0.conv2",
12
+ "decoder.decode.0.norm1.fc",
13
+ "decoder.decode.0.norm2.fc",
14
+ "decoder.decode.1.conv1",
15
+ "decoder.decode.1.conv1x1",
16
+ "decoder.decode.1.conv2",
17
+ "decoder.decode.1.norm1.fc",
18
+ "decoder.decode.1.norm2.fc",
19
+ "decoder.decode.2.conv1",
20
+ "decoder.decode.2.conv1x1",
21
+ "decoder.decode.2.conv2",
22
+ "decoder.decode.2.norm1.fc",
23
+ "decoder.decode.2.norm2.fc",
24
+ "decoder.decode.3.conv1",
25
+ "decoder.decode.3.conv1x1",
26
+ "decoder.decode.3.conv2",
27
+ "decoder.decode.3.norm1.fc",
28
+ "decoder.decode.3.norm2.fc",
29
+ "decoder.encode.conv1",
30
+ "decoder.encode.conv1x1",
31
+ "decoder.encode.conv2",
32
+ "decoder.encode.norm1.fc",
33
+ "decoder.encode.norm2.fc",
34
+ "decoder.generator.conv_post",
35
+ "decoder.generator.noise_convs.0",
36
+ "decoder.generator.noise_convs.1",
37
+ "decoder.generator.noise_res.0.adain1.0.fc",
38
+ "decoder.generator.noise_res.0.adain1.1.fc",
39
+ "decoder.generator.noise_res.0.adain1.2.fc",
40
+ "decoder.generator.noise_res.0.adain2.0.fc",
41
+ "decoder.generator.noise_res.0.adain2.1.fc",
42
+ "decoder.generator.noise_res.0.adain2.2.fc",
43
+ "decoder.generator.noise_res.0.convs1.0",
44
+ "decoder.generator.noise_res.0.convs1.1",
45
+ "decoder.generator.noise_res.0.convs1.2",
46
+ "decoder.generator.noise_res.0.convs2.0",
47
+ "decoder.generator.noise_res.0.convs2.1",
48
+ "decoder.generator.noise_res.0.convs2.2",
49
+ "decoder.generator.noise_res.1.adain1.0.fc",
50
+ "decoder.generator.noise_res.1.adain1.1.fc",
51
+ "decoder.generator.noise_res.1.adain1.2.fc",
52
+ "decoder.generator.noise_res.1.adain2.0.fc",
53
+ "decoder.generator.noise_res.1.adain2.1.fc",
54
+ "decoder.generator.noise_res.1.adain2.2.fc",
55
+ "decoder.generator.noise_res.1.convs1.0",
56
+ "decoder.generator.noise_res.1.convs1.1",
57
+ "decoder.generator.noise_res.1.convs1.2",
58
+ "decoder.generator.noise_res.1.convs2.0",
59
+ "decoder.generator.noise_res.1.convs2.1",
60
+ "decoder.generator.noise_res.1.convs2.2",
61
+ "decoder.generator.resblocks.0.adain1.0.fc",
62
+ "decoder.generator.resblocks.0.adain1.1.fc",
63
+ "decoder.generator.resblocks.0.adain1.2.fc",
64
+ "decoder.generator.resblocks.0.adain2.0.fc",
65
+ "decoder.generator.resblocks.0.adain2.1.fc",
66
+ "decoder.generator.resblocks.0.adain2.2.fc",
67
+ "decoder.generator.resblocks.0.convs1.0",
68
+ "decoder.generator.resblocks.0.convs1.1",
69
+ "decoder.generator.resblocks.0.convs1.2",
70
+ "decoder.generator.resblocks.0.convs2.0",
71
+ "decoder.generator.resblocks.0.convs2.1",
72
+ "decoder.generator.resblocks.0.convs2.2",
73
+ "decoder.generator.resblocks.1.adain1.0.fc",
74
+ "decoder.generator.resblocks.1.adain1.1.fc",
75
+ "decoder.generator.resblocks.1.adain1.2.fc",
76
+ "decoder.generator.resblocks.1.adain2.0.fc",
77
+ "decoder.generator.resblocks.1.adain2.1.fc",
78
+ "decoder.generator.resblocks.1.adain2.2.fc",
79
+ "decoder.generator.resblocks.1.convs1.0",
80
+ "decoder.generator.resblocks.1.convs1.1",
81
+ "decoder.generator.resblocks.1.convs1.2",
82
+ "decoder.generator.resblocks.1.convs2.0",
83
+ "decoder.generator.resblocks.1.convs2.1",
84
+ "decoder.generator.resblocks.1.convs2.2",
85
+ "decoder.generator.resblocks.2.adain1.0.fc",
86
+ "decoder.generator.resblocks.2.adain1.1.fc",
87
+ "decoder.generator.resblocks.2.adain1.2.fc",
88
+ "decoder.generator.resblocks.2.adain2.0.fc",
89
+ "decoder.generator.resblocks.2.adain2.1.fc",
90
+ "decoder.generator.resblocks.2.adain2.2.fc",
91
+ "decoder.generator.resblocks.2.convs1.0",
92
+ "decoder.generator.resblocks.2.convs1.1",
93
+ "decoder.generator.resblocks.2.convs1.2",
94
+ "decoder.generator.resblocks.2.convs2.0",
95
+ "decoder.generator.resblocks.2.convs2.1",
96
+ "decoder.generator.resblocks.2.convs2.2",
97
+ "decoder.generator.resblocks.3.adain1.0.fc",
98
+ "decoder.generator.resblocks.3.adain1.1.fc",
99
+ "decoder.generator.resblocks.3.adain1.2.fc",
100
+ "decoder.generator.resblocks.3.adain2.0.fc",
101
+ "decoder.generator.resblocks.3.adain2.1.fc",
102
+ "decoder.generator.resblocks.3.adain2.2.fc",
103
+ "decoder.generator.resblocks.3.convs1.0",
104
+ "decoder.generator.resblocks.3.convs1.1",
105
+ "decoder.generator.resblocks.3.convs1.2",
106
+ "decoder.generator.resblocks.3.convs2.0",
107
+ "decoder.generator.resblocks.3.convs2.1",
108
+ "decoder.generator.resblocks.3.convs2.2",
109
+ "predictor.F0.0.conv1",
110
+ "predictor.F0.0.conv2",
111
+ "predictor.F0.0.norm1.fc",
112
+ "predictor.F0.0.norm2.fc",
113
+ "predictor.F0.1.conv1",
114
+ "predictor.F0.1.conv1x1",
115
+ "predictor.F0.1.conv2",
116
+ "predictor.F0.1.norm1.fc",
117
+ "predictor.F0.1.norm2.fc",
118
+ "predictor.F0.2.conv1",
119
+ "predictor.F0.2.conv2",
120
+ "predictor.F0.2.norm1.fc",
121
+ "predictor.F0.2.norm2.fc",
122
+ "predictor.F0_proj",
123
+ "predictor.N.0.conv1",
124
+ "predictor.N.0.conv2",
125
+ "predictor.N.0.norm1.fc",
126
+ "predictor.N.0.norm2.fc",
127
+ "predictor.N.1.conv1",
128
+ "predictor.N.1.conv1x1",
129
+ "predictor.N.1.conv2",
130
+ "predictor.N.1.norm1.fc",
131
+ "predictor.N.1.norm2.fc",
132
+ "predictor.N.2.conv1",
133
+ "predictor.N.2.conv2",
134
+ "predictor.N.2.norm1.fc",
135
+ "predictor.N.2.norm2.fc",
136
+ "predictor.N_proj",
137
+ "predictor.duration_proj.linear_layer",
138
+ "predictor.lstm",
139
+ "predictor.shared",
140
+ "predictor.text_encoder.lstms.0",
141
+ "predictor.text_encoder.lstms.1.fc",
142
+ "predictor.text_encoder.lstms.2",
143
+ "predictor.text_encoder.lstms.3.fc",
144
+ "predictor.text_encoder.lstms.4",
145
+ "predictor.text_encoder.lstms.5.fc",
146
+ "text_encoder.cnn.0.0",
147
+ "text_encoder.cnn.1.0",
148
+ "text_encoder.cnn.2.0",
149
+ "text_encoder.lstm"
150
+ ],
151
+ "asr_res_dim": 64,
152
+ "decoder_out_dim": 256,
153
+ "hidden_dim": 512,
154
+ "istftnet": {
155
+ "resblock_kernel_sizes": [
156
+ 3,
157
+ 3
158
+ ],
159
+ "upsample_rates": [
160
+ 10,
161
+ 6
162
+ ],
163
+ "upsample_initial_channel": 256,
164
+ "resblock_dilation_sizes": [
165
+ [
166
+ 1,
167
+ 3,
168
+ 5
169
+ ],
170
+ [
171
+ 1,
172
+ 3,
173
+ 5
174
+ ]
175
+ ],
176
+ "upsample_kernel_sizes": [
177
+ 20,
178
+ 12
179
+ ],
180
+ "gen_istft_n_fft": 20,
181
+ "gen_istft_hop_size": 5
182
+ },
183
+ "max_conv_dim": 256,
184
+ "max_dur": 50,
185
+ "model_type": "kitten_tts",
186
+ "n_layer": 3,
187
+ "n_mels": 80,
188
+ "n_token": 178,
189
+ "plbert": {
190
+ "num_hidden_layers": 12,
191
+ "num_attention_heads": 12,
192
+ "hidden_size": 768,
193
+ "intermediate_size": 2048,
194
+ "max_position_embeddings": 512,
195
+ "embedding_size": 128,
196
+ "inner_group_num": 1,
197
+ "num_hidden_groups": 1,
198
+ "hidden_dropout_prob": 0.0,
199
+ "attention_probs_dropout_prob": 0.0,
200
+ "type_vocab_size": 2,
201
+ "layer_norm_eps": 1e-12
202
+ },
203
+ "quantization": {
204
+ "group_size": 64,
205
+ "bits": 4,
206
+ "mode": "affine"
207
+ },
208
+ "quantization_config": {
209
+ "group_size": 64,
210
+ "bits": 4,
211
+ "mode": "affine"
212
+ },
213
+ "sample_rate": 24000,
214
+ "speed_priors": {},
215
+ "style_dim": 128,
216
+ "text_encoder_kernel_size": 5,
217
+ "voice_aliases": {
218
+ "Bella": "expr-voice-2-f",
219
+ "Jasper": "expr-voice-2-m",
220
+ "Luna": "expr-voice-3-f",
221
+ "Bruno": "expr-voice-3-m",
222
+ "Rosie": "expr-voice-4-f",
223
+ "Hugo": "expr-voice-4-m",
224
+ "Kiki": "expr-voice-5-f",
225
+ "Leo": "expr-voice-5-m"
226
+ },
227
+ "voices_path": "voices.npz"
228
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9edc51ac0f94c26fdffc2ecabcdd9153479e215122c222521bc2177ae35ecdc6
3
+ size 106623089
model.safetensors.index.json ADDED
@@ -0,0 +1,634 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 106554804,
4
+ "total_parameters": 35545426
5
+ },
6
+ "weight_map": {
7
+ "bert.embeddings.LayerNorm.bias": "model.safetensors",
8
+ "bert.embeddings.LayerNorm.weight": "model.safetensors",
9
+ "bert.embeddings.position_embeddings.biases": "model.safetensors",
10
+ "bert.embeddings.position_embeddings.scales": "model.safetensors",
11
+ "bert.embeddings.position_embeddings.weight": "model.safetensors",
12
+ "bert.embeddings.token_type_embeddings.biases": "model.safetensors",
13
+ "bert.embeddings.token_type_embeddings.scales": "model.safetensors",
14
+ "bert.embeddings.token_type_embeddings.weight": "model.safetensors",
15
+ "bert.embeddings.word_embeddings.biases": "model.safetensors",
16
+ "bert.embeddings.word_embeddings.scales": "model.safetensors",
17
+ "bert.embeddings.word_embeddings.weight": "model.safetensors",
18
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.LayerNorm.bias": "model.safetensors",
19
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.LayerNorm.weight": "model.safetensors",
20
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.dense.bias": "model.safetensors",
21
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.dense.biases": "model.safetensors",
22
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.dense.scales": "model.safetensors",
23
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.dense.weight": "model.safetensors",
24
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.key.bias": "model.safetensors",
25
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.key.biases": "model.safetensors",
26
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.key.scales": "model.safetensors",
27
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.key.weight": "model.safetensors",
28
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.query.bias": "model.safetensors",
29
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.query.biases": "model.safetensors",
30
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.query.scales": "model.safetensors",
31
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.query.weight": "model.safetensors",
32
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.value.bias": "model.safetensors",
33
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.value.biases": "model.safetensors",
34
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.value.scales": "model.safetensors",
35
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.attention.value.weight": "model.safetensors",
36
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.ffn.bias": "model.safetensors",
37
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.ffn.biases": "model.safetensors",
38
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.ffn.scales": "model.safetensors",
39
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.ffn.weight": "model.safetensors",
40
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.ffn_output.bias": "model.safetensors",
41
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.ffn_output.biases": "model.safetensors",
42
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.ffn_output.scales": "model.safetensors",
43
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.ffn_output.weight": "model.safetensors",
44
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.full_layer_layer_norm.bias": "model.safetensors",
45
+ "bert.encoder.albert_layer_groups.0.albert_layers.0.full_layer_layer_norm.weight": "model.safetensors",
46
+ "bert.encoder.embedding_hidden_mapping_in.bias": "model.safetensors",
47
+ "bert.encoder.embedding_hidden_mapping_in.biases": "model.safetensors",
48
+ "bert.encoder.embedding_hidden_mapping_in.scales": "model.safetensors",
49
+ "bert.encoder.embedding_hidden_mapping_in.weight": "model.safetensors",
50
+ "bert.pooler.bias": "model.safetensors",
51
+ "bert.pooler.biases": "model.safetensors",
52
+ "bert.pooler.scales": "model.safetensors",
53
+ "bert.pooler.weight": "model.safetensors",
54
+ "bert_encoder.bias": "model.safetensors",
55
+ "bert_encoder.biases": "model.safetensors",
56
+ "bert_encoder.scales": "model.safetensors",
57
+ "bert_encoder.weight": "model.safetensors",
58
+ "decoder.F0_conv.bias": "model.safetensors",
59
+ "decoder.F0_conv.weight_g": "model.safetensors",
60
+ "decoder.F0_conv.weight_v": "model.safetensors",
61
+ "decoder.N_conv.bias": "model.safetensors",
62
+ "decoder.N_conv.weight_g": "model.safetensors",
63
+ "decoder.N_conv.weight_v": "model.safetensors",
64
+ "decoder.asr_res.0.bias": "model.safetensors",
65
+ "decoder.asr_res.0.weight_g": "model.safetensors",
66
+ "decoder.asr_res.0.weight_v": "model.safetensors",
67
+ "decoder.decode.0.conv1.bias": "model.safetensors",
68
+ "decoder.decode.0.conv1.weight_g": "model.safetensors",
69
+ "decoder.decode.0.conv1.weight_v": "model.safetensors",
70
+ "decoder.decode.0.conv1x1.weight_g": "model.safetensors",
71
+ "decoder.decode.0.conv1x1.weight_v": "model.safetensors",
72
+ "decoder.decode.0.conv2.bias": "model.safetensors",
73
+ "decoder.decode.0.conv2.weight_g": "model.safetensors",
74
+ "decoder.decode.0.conv2.weight_v": "model.safetensors",
75
+ "decoder.decode.0.norm1.fc.bias": "model.safetensors",
76
+ "decoder.decode.0.norm1.fc.biases": "model.safetensors",
77
+ "decoder.decode.0.norm1.fc.scales": "model.safetensors",
78
+ "decoder.decode.0.norm1.fc.weight": "model.safetensors",
79
+ "decoder.decode.0.norm2.fc.bias": "model.safetensors",
80
+ "decoder.decode.0.norm2.fc.biases": "model.safetensors",
81
+ "decoder.decode.0.norm2.fc.scales": "model.safetensors",
82
+ "decoder.decode.0.norm2.fc.weight": "model.safetensors",
83
+ "decoder.decode.1.conv1.bias": "model.safetensors",
84
+ "decoder.decode.1.conv1.weight_g": "model.safetensors",
85
+ "decoder.decode.1.conv1.weight_v": "model.safetensors",
86
+ "decoder.decode.1.conv1x1.weight_g": "model.safetensors",
87
+ "decoder.decode.1.conv1x1.weight_v": "model.safetensors",
88
+ "decoder.decode.1.conv2.bias": "model.safetensors",
89
+ "decoder.decode.1.conv2.weight_g": "model.safetensors",
90
+ "decoder.decode.1.conv2.weight_v": "model.safetensors",
91
+ "decoder.decode.1.norm1.fc.bias": "model.safetensors",
92
+ "decoder.decode.1.norm1.fc.biases": "model.safetensors",
93
+ "decoder.decode.1.norm1.fc.scales": "model.safetensors",
94
+ "decoder.decode.1.norm1.fc.weight": "model.safetensors",
95
+ "decoder.decode.1.norm2.fc.bias": "model.safetensors",
96
+ "decoder.decode.1.norm2.fc.biases": "model.safetensors",
97
+ "decoder.decode.1.norm2.fc.scales": "model.safetensors",
98
+ "decoder.decode.1.norm2.fc.weight": "model.safetensors",
99
+ "decoder.decode.2.conv1.bias": "model.safetensors",
100
+ "decoder.decode.2.conv1.weight_g": "model.safetensors",
101
+ "decoder.decode.2.conv1.weight_v": "model.safetensors",
102
+ "decoder.decode.2.conv1x1.weight_g": "model.safetensors",
103
+ "decoder.decode.2.conv1x1.weight_v": "model.safetensors",
104
+ "decoder.decode.2.conv2.bias": "model.safetensors",
105
+ "decoder.decode.2.conv2.weight_g": "model.safetensors",
106
+ "decoder.decode.2.conv2.weight_v": "model.safetensors",
107
+ "decoder.decode.2.norm1.fc.bias": "model.safetensors",
108
+ "decoder.decode.2.norm1.fc.biases": "model.safetensors",
109
+ "decoder.decode.2.norm1.fc.scales": "model.safetensors",
110
+ "decoder.decode.2.norm1.fc.weight": "model.safetensors",
111
+ "decoder.decode.2.norm2.fc.bias": "model.safetensors",
112
+ "decoder.decode.2.norm2.fc.biases": "model.safetensors",
113
+ "decoder.decode.2.norm2.fc.scales": "model.safetensors",
114
+ "decoder.decode.2.norm2.fc.weight": "model.safetensors",
115
+ "decoder.decode.3.conv1.bias": "model.safetensors",
116
+ "decoder.decode.3.conv1.weight_g": "model.safetensors",
117
+ "decoder.decode.3.conv1.weight_v": "model.safetensors",
118
+ "decoder.decode.3.conv1x1.weight_g": "model.safetensors",
119
+ "decoder.decode.3.conv1x1.weight_v": "model.safetensors",
120
+ "decoder.decode.3.conv2.bias": "model.safetensors",
121
+ "decoder.decode.3.conv2.weight_g": "model.safetensors",
122
+ "decoder.decode.3.conv2.weight_v": "model.safetensors",
123
+ "decoder.decode.3.norm1.fc.bias": "model.safetensors",
124
+ "decoder.decode.3.norm1.fc.biases": "model.safetensors",
125
+ "decoder.decode.3.norm1.fc.scales": "model.safetensors",
126
+ "decoder.decode.3.norm1.fc.weight": "model.safetensors",
127
+ "decoder.decode.3.norm2.fc.bias": "model.safetensors",
128
+ "decoder.decode.3.norm2.fc.biases": "model.safetensors",
129
+ "decoder.decode.3.norm2.fc.scales": "model.safetensors",
130
+ "decoder.decode.3.norm2.fc.weight": "model.safetensors",
131
+ "decoder.decode.3.pool.bias": "model.safetensors",
132
+ "decoder.decode.3.pool.weight_g": "model.safetensors",
133
+ "decoder.decode.3.pool.weight_v": "model.safetensors",
134
+ "decoder.encode.conv1.bias": "model.safetensors",
135
+ "decoder.encode.conv1.weight_g": "model.safetensors",
136
+ "decoder.encode.conv1.weight_v": "model.safetensors",
137
+ "decoder.encode.conv1x1.weight_g": "model.safetensors",
138
+ "decoder.encode.conv1x1.weight_v": "model.safetensors",
139
+ "decoder.encode.conv2.bias": "model.safetensors",
140
+ "decoder.encode.conv2.weight_g": "model.safetensors",
141
+ "decoder.encode.conv2.weight_v": "model.safetensors",
142
+ "decoder.encode.norm1.fc.bias": "model.safetensors",
143
+ "decoder.encode.norm1.fc.biases": "model.safetensors",
144
+ "decoder.encode.norm1.fc.scales": "model.safetensors",
145
+ "decoder.encode.norm1.fc.weight": "model.safetensors",
146
+ "decoder.encode.norm2.fc.bias": "model.safetensors",
147
+ "decoder.encode.norm2.fc.biases": "model.safetensors",
148
+ "decoder.encode.norm2.fc.scales": "model.safetensors",
149
+ "decoder.encode.norm2.fc.weight": "model.safetensors",
150
+ "decoder.generator.conv_post.bias": "model.safetensors",
151
+ "decoder.generator.conv_post.weight_g": "model.safetensors",
152
+ "decoder.generator.conv_post.weight_v": "model.safetensors",
153
+ "decoder.generator.m_source.l_linear.bias": "model.safetensors",
154
+ "decoder.generator.m_source.l_linear.weight": "model.safetensors",
155
+ "decoder.generator.noise_convs.0.bias": "model.safetensors",
156
+ "decoder.generator.noise_convs.0.weight": "model.safetensors",
157
+ "decoder.generator.noise_convs.1.bias": "model.safetensors",
158
+ "decoder.generator.noise_convs.1.weight": "model.safetensors",
159
+ "decoder.generator.noise_res.0.adain1.0.fc.bias": "model.safetensors",
160
+ "decoder.generator.noise_res.0.adain1.0.fc.biases": "model.safetensors",
161
+ "decoder.generator.noise_res.0.adain1.0.fc.scales": "model.safetensors",
162
+ "decoder.generator.noise_res.0.adain1.0.fc.weight": "model.safetensors",
163
+ "decoder.generator.noise_res.0.adain1.1.fc.bias": "model.safetensors",
164
+ "decoder.generator.noise_res.0.adain1.1.fc.biases": "model.safetensors",
165
+ "decoder.generator.noise_res.0.adain1.1.fc.scales": "model.safetensors",
166
+ "decoder.generator.noise_res.0.adain1.1.fc.weight": "model.safetensors",
167
+ "decoder.generator.noise_res.0.adain1.2.fc.bias": "model.safetensors",
168
+ "decoder.generator.noise_res.0.adain1.2.fc.biases": "model.safetensors",
169
+ "decoder.generator.noise_res.0.adain1.2.fc.scales": "model.safetensors",
170
+ "decoder.generator.noise_res.0.adain1.2.fc.weight": "model.safetensors",
171
+ "decoder.generator.noise_res.0.adain2.0.fc.bias": "model.safetensors",
172
+ "decoder.generator.noise_res.0.adain2.0.fc.biases": "model.safetensors",
173
+ "decoder.generator.noise_res.0.adain2.0.fc.scales": "model.safetensors",
174
+ "decoder.generator.noise_res.0.adain2.0.fc.weight": "model.safetensors",
175
+ "decoder.generator.noise_res.0.adain2.1.fc.bias": "model.safetensors",
176
+ "decoder.generator.noise_res.0.adain2.1.fc.biases": "model.safetensors",
177
+ "decoder.generator.noise_res.0.adain2.1.fc.scales": "model.safetensors",
178
+ "decoder.generator.noise_res.0.adain2.1.fc.weight": "model.safetensors",
179
+ "decoder.generator.noise_res.0.adain2.2.fc.bias": "model.safetensors",
180
+ "decoder.generator.noise_res.0.adain2.2.fc.biases": "model.safetensors",
181
+ "decoder.generator.noise_res.0.adain2.2.fc.scales": "model.safetensors",
182
+ "decoder.generator.noise_res.0.adain2.2.fc.weight": "model.safetensors",
183
+ "decoder.generator.noise_res.0.alpha1_0": "model.safetensors",
184
+ "decoder.generator.noise_res.0.alpha1_1": "model.safetensors",
185
+ "decoder.generator.noise_res.0.alpha1_2": "model.safetensors",
186
+ "decoder.generator.noise_res.0.alpha2_0": "model.safetensors",
187
+ "decoder.generator.noise_res.0.alpha2_1": "model.safetensors",
188
+ "decoder.generator.noise_res.0.alpha2_2": "model.safetensors",
189
+ "decoder.generator.noise_res.0.convs1.0.bias": "model.safetensors",
190
+ "decoder.generator.noise_res.0.convs1.0.weight_g": "model.safetensors",
191
+ "decoder.generator.noise_res.0.convs1.0.weight_v": "model.safetensors",
192
+ "decoder.generator.noise_res.0.convs1.1.bias": "model.safetensors",
193
+ "decoder.generator.noise_res.0.convs1.1.weight_g": "model.safetensors",
194
+ "decoder.generator.noise_res.0.convs1.1.weight_v": "model.safetensors",
195
+ "decoder.generator.noise_res.0.convs1.2.bias": "model.safetensors",
196
+ "decoder.generator.noise_res.0.convs1.2.weight_g": "model.safetensors",
197
+ "decoder.generator.noise_res.0.convs1.2.weight_v": "model.safetensors",
198
+ "decoder.generator.noise_res.0.convs2.0.bias": "model.safetensors",
199
+ "decoder.generator.noise_res.0.convs2.0.weight_g": "model.safetensors",
200
+ "decoder.generator.noise_res.0.convs2.0.weight_v": "model.safetensors",
201
+ "decoder.generator.noise_res.0.convs2.1.bias": "model.safetensors",
202
+ "decoder.generator.noise_res.0.convs2.1.weight_g": "model.safetensors",
203
+ "decoder.generator.noise_res.0.convs2.1.weight_v": "model.safetensors",
204
+ "decoder.generator.noise_res.0.convs2.2.bias": "model.safetensors",
205
+ "decoder.generator.noise_res.0.convs2.2.weight_g": "model.safetensors",
206
+ "decoder.generator.noise_res.0.convs2.2.weight_v": "model.safetensors",
207
+ "decoder.generator.noise_res.1.adain1.0.fc.bias": "model.safetensors",
208
+ "decoder.generator.noise_res.1.adain1.0.fc.biases": "model.safetensors",
209
+ "decoder.generator.noise_res.1.adain1.0.fc.scales": "model.safetensors",
210
+ "decoder.generator.noise_res.1.adain1.0.fc.weight": "model.safetensors",
211
+ "decoder.generator.noise_res.1.adain1.1.fc.bias": "model.safetensors",
212
+ "decoder.generator.noise_res.1.adain1.1.fc.biases": "model.safetensors",
213
+ "decoder.generator.noise_res.1.adain1.1.fc.scales": "model.safetensors",
214
+ "decoder.generator.noise_res.1.adain1.1.fc.weight": "model.safetensors",
215
+ "decoder.generator.noise_res.1.adain1.2.fc.bias": "model.safetensors",
216
+ "decoder.generator.noise_res.1.adain1.2.fc.biases": "model.safetensors",
217
+ "decoder.generator.noise_res.1.adain1.2.fc.scales": "model.safetensors",
218
+ "decoder.generator.noise_res.1.adain1.2.fc.weight": "model.safetensors",
219
+ "decoder.generator.noise_res.1.adain2.0.fc.bias": "model.safetensors",
220
+ "decoder.generator.noise_res.1.adain2.0.fc.biases": "model.safetensors",
221
+ "decoder.generator.noise_res.1.adain2.0.fc.scales": "model.safetensors",
222
+ "decoder.generator.noise_res.1.adain2.0.fc.weight": "model.safetensors",
223
+ "decoder.generator.noise_res.1.adain2.1.fc.bias": "model.safetensors",
224
+ "decoder.generator.noise_res.1.adain2.1.fc.biases": "model.safetensors",
225
+ "decoder.generator.noise_res.1.adain2.1.fc.scales": "model.safetensors",
226
+ "decoder.generator.noise_res.1.adain2.1.fc.weight": "model.safetensors",
227
+ "decoder.generator.noise_res.1.adain2.2.fc.bias": "model.safetensors",
228
+ "decoder.generator.noise_res.1.adain2.2.fc.biases": "model.safetensors",
229
+ "decoder.generator.noise_res.1.adain2.2.fc.scales": "model.safetensors",
230
+ "decoder.generator.noise_res.1.adain2.2.fc.weight": "model.safetensors",
231
+ "decoder.generator.noise_res.1.alpha1_0": "model.safetensors",
232
+ "decoder.generator.noise_res.1.alpha1_1": "model.safetensors",
233
+ "decoder.generator.noise_res.1.alpha1_2": "model.safetensors",
234
+ "decoder.generator.noise_res.1.alpha2_0": "model.safetensors",
235
+ "decoder.generator.noise_res.1.alpha2_1": "model.safetensors",
236
+ "decoder.generator.noise_res.1.alpha2_2": "model.safetensors",
237
+ "decoder.generator.noise_res.1.convs1.0.bias": "model.safetensors",
238
+ "decoder.generator.noise_res.1.convs1.0.weight_g": "model.safetensors",
239
+ "decoder.generator.noise_res.1.convs1.0.weight_v": "model.safetensors",
240
+ "decoder.generator.noise_res.1.convs1.1.bias": "model.safetensors",
241
+ "decoder.generator.noise_res.1.convs1.1.weight_g": "model.safetensors",
242
+ "decoder.generator.noise_res.1.convs1.1.weight_v": "model.safetensors",
243
+ "decoder.generator.noise_res.1.convs1.2.bias": "model.safetensors",
244
+ "decoder.generator.noise_res.1.convs1.2.weight_g": "model.safetensors",
245
+ "decoder.generator.noise_res.1.convs1.2.weight_v": "model.safetensors",
246
+ "decoder.generator.noise_res.1.convs2.0.bias": "model.safetensors",
247
+ "decoder.generator.noise_res.1.convs2.0.weight_g": "model.safetensors",
248
+ "decoder.generator.noise_res.1.convs2.0.weight_v": "model.safetensors",
249
+ "decoder.generator.noise_res.1.convs2.1.bias": "model.safetensors",
250
+ "decoder.generator.noise_res.1.convs2.1.weight_g": "model.safetensors",
251
+ "decoder.generator.noise_res.1.convs2.1.weight_v": "model.safetensors",
252
+ "decoder.generator.noise_res.1.convs2.2.bias": "model.safetensors",
253
+ "decoder.generator.noise_res.1.convs2.2.weight_g": "model.safetensors",
254
+ "decoder.generator.noise_res.1.convs2.2.weight_v": "model.safetensors",
255
+ "decoder.generator.resblocks.0.adain1.0.fc.bias": "model.safetensors",
256
+ "decoder.generator.resblocks.0.adain1.0.fc.biases": "model.safetensors",
257
+ "decoder.generator.resblocks.0.adain1.0.fc.scales": "model.safetensors",
258
+ "decoder.generator.resblocks.0.adain1.0.fc.weight": "model.safetensors",
259
+ "decoder.generator.resblocks.0.adain1.1.fc.bias": "model.safetensors",
260
+ "decoder.generator.resblocks.0.adain1.1.fc.biases": "model.safetensors",
261
+ "decoder.generator.resblocks.0.adain1.1.fc.scales": "model.safetensors",
262
+ "decoder.generator.resblocks.0.adain1.1.fc.weight": "model.safetensors",
263
+ "decoder.generator.resblocks.0.adain1.2.fc.bias": "model.safetensors",
264
+ "decoder.generator.resblocks.0.adain1.2.fc.biases": "model.safetensors",
265
+ "decoder.generator.resblocks.0.adain1.2.fc.scales": "model.safetensors",
266
+ "decoder.generator.resblocks.0.adain1.2.fc.weight": "model.safetensors",
267
+ "decoder.generator.resblocks.0.adain2.0.fc.bias": "model.safetensors",
268
+ "decoder.generator.resblocks.0.adain2.0.fc.biases": "model.safetensors",
269
+ "decoder.generator.resblocks.0.adain2.0.fc.scales": "model.safetensors",
270
+ "decoder.generator.resblocks.0.adain2.0.fc.weight": "model.safetensors",
271
+ "decoder.generator.resblocks.0.adain2.1.fc.bias": "model.safetensors",
272
+ "decoder.generator.resblocks.0.adain2.1.fc.biases": "model.safetensors",
273
+ "decoder.generator.resblocks.0.adain2.1.fc.scales": "model.safetensors",
274
+ "decoder.generator.resblocks.0.adain2.1.fc.weight": "model.safetensors",
275
+ "decoder.generator.resblocks.0.adain2.2.fc.bias": "model.safetensors",
276
+ "decoder.generator.resblocks.0.adain2.2.fc.biases": "model.safetensors",
277
+ "decoder.generator.resblocks.0.adain2.2.fc.scales": "model.safetensors",
278
+ "decoder.generator.resblocks.0.adain2.2.fc.weight": "model.safetensors",
279
+ "decoder.generator.resblocks.0.alpha1_0": "model.safetensors",
280
+ "decoder.generator.resblocks.0.alpha1_1": "model.safetensors",
281
+ "decoder.generator.resblocks.0.alpha1_2": "model.safetensors",
282
+ "decoder.generator.resblocks.0.alpha2_0": "model.safetensors",
283
+ "decoder.generator.resblocks.0.alpha2_1": "model.safetensors",
284
+ "decoder.generator.resblocks.0.alpha2_2": "model.safetensors",
285
+ "decoder.generator.resblocks.0.convs1.0.bias": "model.safetensors",
286
+ "decoder.generator.resblocks.0.convs1.0.weight_g": "model.safetensors",
287
+ "decoder.generator.resblocks.0.convs1.0.weight_v": "model.safetensors",
288
+ "decoder.generator.resblocks.0.convs1.1.bias": "model.safetensors",
289
+ "decoder.generator.resblocks.0.convs1.1.weight_g": "model.safetensors",
290
+ "decoder.generator.resblocks.0.convs1.1.weight_v": "model.safetensors",
291
+ "decoder.generator.resblocks.0.convs1.2.bias": "model.safetensors",
292
+ "decoder.generator.resblocks.0.convs1.2.weight_g": "model.safetensors",
293
+ "decoder.generator.resblocks.0.convs1.2.weight_v": "model.safetensors",
294
+ "decoder.generator.resblocks.0.convs2.0.bias": "model.safetensors",
295
+ "decoder.generator.resblocks.0.convs2.0.weight_g": "model.safetensors",
296
+ "decoder.generator.resblocks.0.convs2.0.weight_v": "model.safetensors",
297
+ "decoder.generator.resblocks.0.convs2.1.bias": "model.safetensors",
298
+ "decoder.generator.resblocks.0.convs2.1.weight_g": "model.safetensors",
299
+ "decoder.generator.resblocks.0.convs2.1.weight_v": "model.safetensors",
300
+ "decoder.generator.resblocks.0.convs2.2.bias": "model.safetensors",
301
+ "decoder.generator.resblocks.0.convs2.2.weight_g": "model.safetensors",
302
+ "decoder.generator.resblocks.0.convs2.2.weight_v": "model.safetensors",
303
+ "decoder.generator.resblocks.1.adain1.0.fc.bias": "model.safetensors",
304
+ "decoder.generator.resblocks.1.adain1.0.fc.biases": "model.safetensors",
305
+ "decoder.generator.resblocks.1.adain1.0.fc.scales": "model.safetensors",
306
+ "decoder.generator.resblocks.1.adain1.0.fc.weight": "model.safetensors",
307
+ "decoder.generator.resblocks.1.adain1.1.fc.bias": "model.safetensors",
308
+ "decoder.generator.resblocks.1.adain1.1.fc.biases": "model.safetensors",
309
+ "decoder.generator.resblocks.1.adain1.1.fc.scales": "model.safetensors",
310
+ "decoder.generator.resblocks.1.adain1.1.fc.weight": "model.safetensors",
311
+ "decoder.generator.resblocks.1.adain1.2.fc.bias": "model.safetensors",
312
+ "decoder.generator.resblocks.1.adain1.2.fc.biases": "model.safetensors",
313
+ "decoder.generator.resblocks.1.adain1.2.fc.scales": "model.safetensors",
314
+ "decoder.generator.resblocks.1.adain1.2.fc.weight": "model.safetensors",
315
+ "decoder.generator.resblocks.1.adain2.0.fc.bias": "model.safetensors",
316
+ "decoder.generator.resblocks.1.adain2.0.fc.biases": "model.safetensors",
317
+ "decoder.generator.resblocks.1.adain2.0.fc.scales": "model.safetensors",
318
+ "decoder.generator.resblocks.1.adain2.0.fc.weight": "model.safetensors",
319
+ "decoder.generator.resblocks.1.adain2.1.fc.bias": "model.safetensors",
320
+ "decoder.generator.resblocks.1.adain2.1.fc.biases": "model.safetensors",
321
+ "decoder.generator.resblocks.1.adain2.1.fc.scales": "model.safetensors",
322
+ "decoder.generator.resblocks.1.adain2.1.fc.weight": "model.safetensors",
323
+ "decoder.generator.resblocks.1.adain2.2.fc.bias": "model.safetensors",
324
+ "decoder.generator.resblocks.1.adain2.2.fc.biases": "model.safetensors",
325
+ "decoder.generator.resblocks.1.adain2.2.fc.scales": "model.safetensors",
326
+ "decoder.generator.resblocks.1.adain2.2.fc.weight": "model.safetensors",
327
+ "decoder.generator.resblocks.1.alpha1_0": "model.safetensors",
328
+ "decoder.generator.resblocks.1.alpha1_1": "model.safetensors",
329
+ "decoder.generator.resblocks.1.alpha1_2": "model.safetensors",
330
+ "decoder.generator.resblocks.1.alpha2_0": "model.safetensors",
331
+ "decoder.generator.resblocks.1.alpha2_1": "model.safetensors",
332
+ "decoder.generator.resblocks.1.alpha2_2": "model.safetensors",
333
+ "decoder.generator.resblocks.1.convs1.0.bias": "model.safetensors",
334
+ "decoder.generator.resblocks.1.convs1.0.weight_g": "model.safetensors",
335
+ "decoder.generator.resblocks.1.convs1.0.weight_v": "model.safetensors",
336
+ "decoder.generator.resblocks.1.convs1.1.bias": "model.safetensors",
337
+ "decoder.generator.resblocks.1.convs1.1.weight_g": "model.safetensors",
338
+ "decoder.generator.resblocks.1.convs1.1.weight_v": "model.safetensors",
339
+ "decoder.generator.resblocks.1.convs1.2.bias": "model.safetensors",
340
+ "decoder.generator.resblocks.1.convs1.2.weight_g": "model.safetensors",
341
+ "decoder.generator.resblocks.1.convs1.2.weight_v": "model.safetensors",
342
+ "decoder.generator.resblocks.1.convs2.0.bias": "model.safetensors",
343
+ "decoder.generator.resblocks.1.convs2.0.weight_g": "model.safetensors",
344
+ "decoder.generator.resblocks.1.convs2.0.weight_v": "model.safetensors",
345
+ "decoder.generator.resblocks.1.convs2.1.bias": "model.safetensors",
346
+ "decoder.generator.resblocks.1.convs2.1.weight_g": "model.safetensors",
347
+ "decoder.generator.resblocks.1.convs2.1.weight_v": "model.safetensors",
348
+ "decoder.generator.resblocks.1.convs2.2.bias": "model.safetensors",
349
+ "decoder.generator.resblocks.1.convs2.2.weight_g": "model.safetensors",
350
+ "decoder.generator.resblocks.1.convs2.2.weight_v": "model.safetensors",
351
+ "decoder.generator.resblocks.2.adain1.0.fc.bias": "model.safetensors",
352
+ "decoder.generator.resblocks.2.adain1.0.fc.biases": "model.safetensors",
353
+ "decoder.generator.resblocks.2.adain1.0.fc.scales": "model.safetensors",
354
+ "decoder.generator.resblocks.2.adain1.0.fc.weight": "model.safetensors",
355
+ "decoder.generator.resblocks.2.adain1.1.fc.bias": "model.safetensors",
356
+ "decoder.generator.resblocks.2.adain1.1.fc.biases": "model.safetensors",
357
+ "decoder.generator.resblocks.2.adain1.1.fc.scales": "model.safetensors",
358
+ "decoder.generator.resblocks.2.adain1.1.fc.weight": "model.safetensors",
359
+ "decoder.generator.resblocks.2.adain1.2.fc.bias": "model.safetensors",
360
+ "decoder.generator.resblocks.2.adain1.2.fc.biases": "model.safetensors",
361
+ "decoder.generator.resblocks.2.adain1.2.fc.scales": "model.safetensors",
362
+ "decoder.generator.resblocks.2.adain1.2.fc.weight": "model.safetensors",
363
+ "decoder.generator.resblocks.2.adain2.0.fc.bias": "model.safetensors",
364
+ "decoder.generator.resblocks.2.adain2.0.fc.biases": "model.safetensors",
365
+ "decoder.generator.resblocks.2.adain2.0.fc.scales": "model.safetensors",
366
+ "decoder.generator.resblocks.2.adain2.0.fc.weight": "model.safetensors",
367
+ "decoder.generator.resblocks.2.adain2.1.fc.bias": "model.safetensors",
368
+ "decoder.generator.resblocks.2.adain2.1.fc.biases": "model.safetensors",
369
+ "decoder.generator.resblocks.2.adain2.1.fc.scales": "model.safetensors",
370
+ "decoder.generator.resblocks.2.adain2.1.fc.weight": "model.safetensors",
371
+ "decoder.generator.resblocks.2.adain2.2.fc.bias": "model.safetensors",
372
+ "decoder.generator.resblocks.2.adain2.2.fc.biases": "model.safetensors",
373
+ "decoder.generator.resblocks.2.adain2.2.fc.scales": "model.safetensors",
374
+ "decoder.generator.resblocks.2.adain2.2.fc.weight": "model.safetensors",
375
+ "decoder.generator.resblocks.2.alpha1_0": "model.safetensors",
376
+ "decoder.generator.resblocks.2.alpha1_1": "model.safetensors",
377
+ "decoder.generator.resblocks.2.alpha1_2": "model.safetensors",
378
+ "decoder.generator.resblocks.2.alpha2_0": "model.safetensors",
379
+ "decoder.generator.resblocks.2.alpha2_1": "model.safetensors",
380
+ "decoder.generator.resblocks.2.alpha2_2": "model.safetensors",
381
+ "decoder.generator.resblocks.2.convs1.0.bias": "model.safetensors",
382
+ "decoder.generator.resblocks.2.convs1.0.weight_g": "model.safetensors",
383
+ "decoder.generator.resblocks.2.convs1.0.weight_v": "model.safetensors",
384
+ "decoder.generator.resblocks.2.convs1.1.bias": "model.safetensors",
385
+ "decoder.generator.resblocks.2.convs1.1.weight_g": "model.safetensors",
386
+ "decoder.generator.resblocks.2.convs1.1.weight_v": "model.safetensors",
387
+ "decoder.generator.resblocks.2.convs1.2.bias": "model.safetensors",
388
+ "decoder.generator.resblocks.2.convs1.2.weight_g": "model.safetensors",
389
+ "decoder.generator.resblocks.2.convs1.2.weight_v": "model.safetensors",
390
+ "decoder.generator.resblocks.2.convs2.0.bias": "model.safetensors",
391
+ "decoder.generator.resblocks.2.convs2.0.weight_g": "model.safetensors",
392
+ "decoder.generator.resblocks.2.convs2.0.weight_v": "model.safetensors",
393
+ "decoder.generator.resblocks.2.convs2.1.bias": "model.safetensors",
394
+ "decoder.generator.resblocks.2.convs2.1.weight_g": "model.safetensors",
395
+ "decoder.generator.resblocks.2.convs2.1.weight_v": "model.safetensors",
396
+ "decoder.generator.resblocks.2.convs2.2.bias": "model.safetensors",
397
+ "decoder.generator.resblocks.2.convs2.2.weight_g": "model.safetensors",
398
+ "decoder.generator.resblocks.2.convs2.2.weight_v": "model.safetensors",
399
+ "decoder.generator.resblocks.3.adain1.0.fc.bias": "model.safetensors",
400
+ "decoder.generator.resblocks.3.adain1.0.fc.biases": "model.safetensors",
401
+ "decoder.generator.resblocks.3.adain1.0.fc.scales": "model.safetensors",
402
+ "decoder.generator.resblocks.3.adain1.0.fc.weight": "model.safetensors",
403
+ "decoder.generator.resblocks.3.adain1.1.fc.bias": "model.safetensors",
404
+ "decoder.generator.resblocks.3.adain1.1.fc.biases": "model.safetensors",
405
+ "decoder.generator.resblocks.3.adain1.1.fc.scales": "model.safetensors",
406
+ "decoder.generator.resblocks.3.adain1.1.fc.weight": "model.safetensors",
407
+ "decoder.generator.resblocks.3.adain1.2.fc.bias": "model.safetensors",
408
+ "decoder.generator.resblocks.3.adain1.2.fc.biases": "model.safetensors",
409
+ "decoder.generator.resblocks.3.adain1.2.fc.scales": "model.safetensors",
410
+ "decoder.generator.resblocks.3.adain1.2.fc.weight": "model.safetensors",
411
+ "decoder.generator.resblocks.3.adain2.0.fc.bias": "model.safetensors",
412
+ "decoder.generator.resblocks.3.adain2.0.fc.biases": "model.safetensors",
413
+ "decoder.generator.resblocks.3.adain2.0.fc.scales": "model.safetensors",
414
+ "decoder.generator.resblocks.3.adain2.0.fc.weight": "model.safetensors",
415
+ "decoder.generator.resblocks.3.adain2.1.fc.bias": "model.safetensors",
416
+ "decoder.generator.resblocks.3.adain2.1.fc.biases": "model.safetensors",
417
+ "decoder.generator.resblocks.3.adain2.1.fc.scales": "model.safetensors",
418
+ "decoder.generator.resblocks.3.adain2.1.fc.weight": "model.safetensors",
419
+ "decoder.generator.resblocks.3.adain2.2.fc.bias": "model.safetensors",
420
+ "decoder.generator.resblocks.3.adain2.2.fc.biases": "model.safetensors",
421
+ "decoder.generator.resblocks.3.adain2.2.fc.scales": "model.safetensors",
422
+ "decoder.generator.resblocks.3.adain2.2.fc.weight": "model.safetensors",
423
+ "decoder.generator.resblocks.3.alpha1_0": "model.safetensors",
424
+ "decoder.generator.resblocks.3.alpha1_1": "model.safetensors",
425
+ "decoder.generator.resblocks.3.alpha1_2": "model.safetensors",
426
+ "decoder.generator.resblocks.3.alpha2_0": "model.safetensors",
427
+ "decoder.generator.resblocks.3.alpha2_1": "model.safetensors",
428
+ "decoder.generator.resblocks.3.alpha2_2": "model.safetensors",
429
+ "decoder.generator.resblocks.3.convs1.0.bias": "model.safetensors",
430
+ "decoder.generator.resblocks.3.convs1.0.weight_g": "model.safetensors",
431
+ "decoder.generator.resblocks.3.convs1.0.weight_v": "model.safetensors",
432
+ "decoder.generator.resblocks.3.convs1.1.bias": "model.safetensors",
433
+ "decoder.generator.resblocks.3.convs1.1.weight_g": "model.safetensors",
434
+ "decoder.generator.resblocks.3.convs1.1.weight_v": "model.safetensors",
435
+ "decoder.generator.resblocks.3.convs1.2.bias": "model.safetensors",
436
+ "decoder.generator.resblocks.3.convs1.2.weight_g": "model.safetensors",
437
+ "decoder.generator.resblocks.3.convs1.2.weight_v": "model.safetensors",
438
+ "decoder.generator.resblocks.3.convs2.0.bias": "model.safetensors",
439
+ "decoder.generator.resblocks.3.convs2.0.weight_g": "model.safetensors",
440
+ "decoder.generator.resblocks.3.convs2.0.weight_v": "model.safetensors",
441
+ "decoder.generator.resblocks.3.convs2.1.bias": "model.safetensors",
442
+ "decoder.generator.resblocks.3.convs2.1.weight_g": "model.safetensors",
443
+ "decoder.generator.resblocks.3.convs2.1.weight_v": "model.safetensors",
444
+ "decoder.generator.resblocks.3.convs2.2.bias": "model.safetensors",
445
+ "decoder.generator.resblocks.3.convs2.2.weight_g": "model.safetensors",
446
+ "decoder.generator.resblocks.3.convs2.2.weight_v": "model.safetensors",
447
+ "decoder.generator.ups.0.bias": "model.safetensors",
448
+ "decoder.generator.ups.0.weight_g": "model.safetensors",
449
+ "decoder.generator.ups.0.weight_v": "model.safetensors",
450
+ "decoder.generator.ups.1.bias": "model.safetensors",
451
+ "decoder.generator.ups.1.weight_g": "model.safetensors",
452
+ "decoder.generator.ups.1.weight_v": "model.safetensors",
453
+ "predictor.F0.0.conv1.bias": "model.safetensors",
454
+ "predictor.F0.0.conv1.weight_g": "model.safetensors",
455
+ "predictor.F0.0.conv1.weight_v": "model.safetensors",
456
+ "predictor.F0.0.conv2.bias": "model.safetensors",
457
+ "predictor.F0.0.conv2.weight_g": "model.safetensors",
458
+ "predictor.F0.0.conv2.weight_v": "model.safetensors",
459
+ "predictor.F0.0.norm1.fc.bias": "model.safetensors",
460
+ "predictor.F0.0.norm1.fc.biases": "model.safetensors",
461
+ "predictor.F0.0.norm1.fc.scales": "model.safetensors",
462
+ "predictor.F0.0.norm1.fc.weight": "model.safetensors",
463
+ "predictor.F0.0.norm2.fc.bias": "model.safetensors",
464
+ "predictor.F0.0.norm2.fc.biases": "model.safetensors",
465
+ "predictor.F0.0.norm2.fc.scales": "model.safetensors",
466
+ "predictor.F0.0.norm2.fc.weight": "model.safetensors",
467
+ "predictor.F0.1.conv1.bias": "model.safetensors",
468
+ "predictor.F0.1.conv1.weight_g": "model.safetensors",
469
+ "predictor.F0.1.conv1.weight_v": "model.safetensors",
470
+ "predictor.F0.1.conv1x1.weight_g": "model.safetensors",
471
+ "predictor.F0.1.conv1x1.weight_v": "model.safetensors",
472
+ "predictor.F0.1.conv2.bias": "model.safetensors",
473
+ "predictor.F0.1.conv2.weight_g": "model.safetensors",
474
+ "predictor.F0.1.conv2.weight_v": "model.safetensors",
475
+ "predictor.F0.1.norm1.fc.bias": "model.safetensors",
476
+ "predictor.F0.1.norm1.fc.biases": "model.safetensors",
477
+ "predictor.F0.1.norm1.fc.scales": "model.safetensors",
478
+ "predictor.F0.1.norm1.fc.weight": "model.safetensors",
479
+ "predictor.F0.1.norm2.fc.bias": "model.safetensors",
480
+ "predictor.F0.1.norm2.fc.biases": "model.safetensors",
481
+ "predictor.F0.1.norm2.fc.scales": "model.safetensors",
482
+ "predictor.F0.1.norm2.fc.weight": "model.safetensors",
483
+ "predictor.F0.1.pool.bias": "model.safetensors",
484
+ "predictor.F0.1.pool.weight_g": "model.safetensors",
485
+ "predictor.F0.1.pool.weight_v": "model.safetensors",
486
+ "predictor.F0.2.conv1.bias": "model.safetensors",
487
+ "predictor.F0.2.conv1.weight_g": "model.safetensors",
488
+ "predictor.F0.2.conv1.weight_v": "model.safetensors",
489
+ "predictor.F0.2.conv2.bias": "model.safetensors",
490
+ "predictor.F0.2.conv2.weight_g": "model.safetensors",
491
+ "predictor.F0.2.conv2.weight_v": "model.safetensors",
492
+ "predictor.F0.2.norm1.fc.bias": "model.safetensors",
493
+ "predictor.F0.2.norm1.fc.biases": "model.safetensors",
494
+ "predictor.F0.2.norm1.fc.scales": "model.safetensors",
495
+ "predictor.F0.2.norm1.fc.weight": "model.safetensors",
496
+ "predictor.F0.2.norm2.fc.bias": "model.safetensors",
497
+ "predictor.F0.2.norm2.fc.biases": "model.safetensors",
498
+ "predictor.F0.2.norm2.fc.scales": "model.safetensors",
499
+ "predictor.F0.2.norm2.fc.weight": "model.safetensors",
500
+ "predictor.F0_proj.bias": "model.safetensors",
501
+ "predictor.F0_proj.weight": "model.safetensors",
502
+ "predictor.N.0.conv1.bias": "model.safetensors",
503
+ "predictor.N.0.conv1.weight_g": "model.safetensors",
504
+ "predictor.N.0.conv1.weight_v": "model.safetensors",
505
+ "predictor.N.0.conv2.bias": "model.safetensors",
506
+ "predictor.N.0.conv2.weight_g": "model.safetensors",
507
+ "predictor.N.0.conv2.weight_v": "model.safetensors",
508
+ "predictor.N.0.norm1.fc.bias": "model.safetensors",
509
+ "predictor.N.0.norm1.fc.biases": "model.safetensors",
510
+ "predictor.N.0.norm1.fc.scales": "model.safetensors",
511
+ "predictor.N.0.norm1.fc.weight": "model.safetensors",
512
+ "predictor.N.0.norm2.fc.bias": "model.safetensors",
513
+ "predictor.N.0.norm2.fc.biases": "model.safetensors",
514
+ "predictor.N.0.norm2.fc.scales": "model.safetensors",
515
+ "predictor.N.0.norm2.fc.weight": "model.safetensors",
516
+ "predictor.N.1.conv1.bias": "model.safetensors",
517
+ "predictor.N.1.conv1.weight_g": "model.safetensors",
518
+ "predictor.N.1.conv1.weight_v": "model.safetensors",
519
+ "predictor.N.1.conv1x1.weight_g": "model.safetensors",
520
+ "predictor.N.1.conv1x1.weight_v": "model.safetensors",
521
+ "predictor.N.1.conv2.bias": "model.safetensors",
522
+ "predictor.N.1.conv2.weight_g": "model.safetensors",
523
+ "predictor.N.1.conv2.weight_v": "model.safetensors",
524
+ "predictor.N.1.norm1.fc.bias": "model.safetensors",
525
+ "predictor.N.1.norm1.fc.biases": "model.safetensors",
526
+ "predictor.N.1.norm1.fc.scales": "model.safetensors",
527
+ "predictor.N.1.norm1.fc.weight": "model.safetensors",
528
+ "predictor.N.1.norm2.fc.bias": "model.safetensors",
529
+ "predictor.N.1.norm2.fc.biases": "model.safetensors",
530
+ "predictor.N.1.norm2.fc.scales": "model.safetensors",
531
+ "predictor.N.1.norm2.fc.weight": "model.safetensors",
532
+ "predictor.N.1.pool.bias": "model.safetensors",
533
+ "predictor.N.1.pool.weight_g": "model.safetensors",
534
+ "predictor.N.1.pool.weight_v": "model.safetensors",
535
+ "predictor.N.2.conv1.bias": "model.safetensors",
536
+ "predictor.N.2.conv1.weight_g": "model.safetensors",
537
+ "predictor.N.2.conv1.weight_v": "model.safetensors",
538
+ "predictor.N.2.conv2.bias": "model.safetensors",
539
+ "predictor.N.2.conv2.weight_g": "model.safetensors",
540
+ "predictor.N.2.conv2.weight_v": "model.safetensors",
541
+ "predictor.N.2.norm1.fc.bias": "model.safetensors",
542
+ "predictor.N.2.norm1.fc.biases": "model.safetensors",
543
+ "predictor.N.2.norm1.fc.scales": "model.safetensors",
544
+ "predictor.N.2.norm1.fc.weight": "model.safetensors",
545
+ "predictor.N.2.norm2.fc.bias": "model.safetensors",
546
+ "predictor.N.2.norm2.fc.biases": "model.safetensors",
547
+ "predictor.N.2.norm2.fc.scales": "model.safetensors",
548
+ "predictor.N.2.norm2.fc.weight": "model.safetensors",
549
+ "predictor.N_proj.bias": "model.safetensors",
550
+ "predictor.N_proj.weight": "model.safetensors",
551
+ "predictor.duration_proj.linear_layer.bias": "model.safetensors",
552
+ "predictor.duration_proj.linear_layer.biases": "model.safetensors",
553
+ "predictor.duration_proj.linear_layer.scales": "model.safetensors",
554
+ "predictor.duration_proj.linear_layer.weight": "model.safetensors",
555
+ "predictor.lstm.Wh_backward": "model.safetensors",
556
+ "predictor.lstm.Wh_forward": "model.safetensors",
557
+ "predictor.lstm.Wx_backward": "model.safetensors",
558
+ "predictor.lstm.Wx_forward": "model.safetensors",
559
+ "predictor.lstm.bias_hh_backward": "model.safetensors",
560
+ "predictor.lstm.bias_hh_forward": "model.safetensors",
561
+ "predictor.lstm.bias_ih_backward": "model.safetensors",
562
+ "predictor.lstm.bias_ih_forward": "model.safetensors",
563
+ "predictor.shared.Wh_backward": "model.safetensors",
564
+ "predictor.shared.Wh_forward": "model.safetensors",
565
+ "predictor.shared.Wx_backward": "model.safetensors",
566
+ "predictor.shared.Wx_forward": "model.safetensors",
567
+ "predictor.shared.bias_hh_backward": "model.safetensors",
568
+ "predictor.shared.bias_hh_forward": "model.safetensors",
569
+ "predictor.shared.bias_ih_backward": "model.safetensors",
570
+ "predictor.shared.bias_ih_forward": "model.safetensors",
571
+ "predictor.text_encoder.lstms.0.Wh_backward": "model.safetensors",
572
+ "predictor.text_encoder.lstms.0.Wh_forward": "model.safetensors",
573
+ "predictor.text_encoder.lstms.0.Wx_backward": "model.safetensors",
574
+ "predictor.text_encoder.lstms.0.Wx_forward": "model.safetensors",
575
+ "predictor.text_encoder.lstms.0.bias_hh_backward": "model.safetensors",
576
+ "predictor.text_encoder.lstms.0.bias_hh_forward": "model.safetensors",
577
+ "predictor.text_encoder.lstms.0.bias_ih_backward": "model.safetensors",
578
+ "predictor.text_encoder.lstms.0.bias_ih_forward": "model.safetensors",
579
+ "predictor.text_encoder.lstms.1.fc.bias": "model.safetensors",
580
+ "predictor.text_encoder.lstms.1.fc.biases": "model.safetensors",
581
+ "predictor.text_encoder.lstms.1.fc.scales": "model.safetensors",
582
+ "predictor.text_encoder.lstms.1.fc.weight": "model.safetensors",
583
+ "predictor.text_encoder.lstms.2.Wh_backward": "model.safetensors",
584
+ "predictor.text_encoder.lstms.2.Wh_forward": "model.safetensors",
585
+ "predictor.text_encoder.lstms.2.Wx_backward": "model.safetensors",
586
+ "predictor.text_encoder.lstms.2.Wx_forward": "model.safetensors",
587
+ "predictor.text_encoder.lstms.2.bias_hh_backward": "model.safetensors",
588
+ "predictor.text_encoder.lstms.2.bias_hh_forward": "model.safetensors",
589
+ "predictor.text_encoder.lstms.2.bias_ih_backward": "model.safetensors",
590
+ "predictor.text_encoder.lstms.2.bias_ih_forward": "model.safetensors",
591
+ "predictor.text_encoder.lstms.3.fc.bias": "model.safetensors",
592
+ "predictor.text_encoder.lstms.3.fc.biases": "model.safetensors",
593
+ "predictor.text_encoder.lstms.3.fc.scales": "model.safetensors",
594
+ "predictor.text_encoder.lstms.3.fc.weight": "model.safetensors",
595
+ "predictor.text_encoder.lstms.4.Wh_backward": "model.safetensors",
596
+ "predictor.text_encoder.lstms.4.Wh_forward": "model.safetensors",
597
+ "predictor.text_encoder.lstms.4.Wx_backward": "model.safetensors",
598
+ "predictor.text_encoder.lstms.4.Wx_forward": "model.safetensors",
599
+ "predictor.text_encoder.lstms.4.bias_hh_backward": "model.safetensors",
600
+ "predictor.text_encoder.lstms.4.bias_hh_forward": "model.safetensors",
601
+ "predictor.text_encoder.lstms.4.bias_ih_backward": "model.safetensors",
602
+ "predictor.text_encoder.lstms.4.bias_ih_forward": "model.safetensors",
603
+ "predictor.text_encoder.lstms.5.fc.bias": "model.safetensors",
604
+ "predictor.text_encoder.lstms.5.fc.biases": "model.safetensors",
605
+ "predictor.text_encoder.lstms.5.fc.scales": "model.safetensors",
606
+ "predictor.text_encoder.lstms.5.fc.weight": "model.safetensors",
607
+ "text_encoder.cnn.0.0.bias": "model.safetensors",
608
+ "text_encoder.cnn.0.0.weight_g": "model.safetensors",
609
+ "text_encoder.cnn.0.0.weight_v": "model.safetensors",
610
+ "text_encoder.cnn.0.1.bias": "model.safetensors",
611
+ "text_encoder.cnn.0.1.weight": "model.safetensors",
612
+ "text_encoder.cnn.1.0.bias": "model.safetensors",
613
+ "text_encoder.cnn.1.0.weight_g": "model.safetensors",
614
+ "text_encoder.cnn.1.0.weight_v": "model.safetensors",
615
+ "text_encoder.cnn.1.1.bias": "model.safetensors",
616
+ "text_encoder.cnn.1.1.weight": "model.safetensors",
617
+ "text_encoder.cnn.2.0.bias": "model.safetensors",
618
+ "text_encoder.cnn.2.0.weight_g": "model.safetensors",
619
+ "text_encoder.cnn.2.0.weight_v": "model.safetensors",
620
+ "text_encoder.cnn.2.1.bias": "model.safetensors",
621
+ "text_encoder.cnn.2.1.weight": "model.safetensors",
622
+ "text_encoder.embedding.biases": "model.safetensors",
623
+ "text_encoder.embedding.scales": "model.safetensors",
624
+ "text_encoder.embedding.weight": "model.safetensors",
625
+ "text_encoder.lstm.Wh_backward": "model.safetensors",
626
+ "text_encoder.lstm.Wh_forward": "model.safetensors",
627
+ "text_encoder.lstm.Wx_backward": "model.safetensors",
628
+ "text_encoder.lstm.Wx_forward": "model.safetensors",
629
+ "text_encoder.lstm.bias_hh_backward": "model.safetensors",
630
+ "text_encoder.lstm.bias_hh_forward": "model.safetensors",
631
+ "text_encoder.lstm.bias_ih_backward": "model.safetensors",
632
+ "text_encoder.lstm.bias_ih_forward": "model.safetensors"
633
+ }
634
+ }
voices.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:112710c1be8ad0e967c190fb0fd95cbe5848ec4791b93209f20b28b7da20dac1
3
+ size 3278902