Hugging Face's logo

FluidInference
/
kokoro-82m-coreml

Text-to-Speech
Core ML
English
Model card Files
xet
 Community
3
alexwengg commited on
Commit
5647a89
·
verified ·
1 Parent(s): be082da

Upload 8 files

Browse files
Files changed (8) hide show
  1. ANE/KokoroNoise_v2.mlmodelc/analytics/coremldata.bin +3 -0
  2. ANE/KokoroNoise_v2.mlmodelc/coremldata.bin +3 -0
  3. ANE/KokoroNoise_v2.mlmodelc/metadata.json +111 -0
  4. ANE/KokoroNoise_v2.mlmodelc/model.mil +546 -0
  5. ANE/KokoroNoise_v2.mlmodelc/weights/weight.bin +3 -0
  6. ANE/KokoroNoise_v2.mlpackage/Data/com.apple.CoreML/model.mlmodel +3 -0
  7. ANE/KokoroNoise_v2.mlpackage/Data/com.apple.CoreML/weights/weight.bin +3 -0
  8. ANE/KokoroNoise_v2.mlpackage/Manifest.json +18 -0
ANE/KokoroNoise_v2.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:53af6bf61482f6002bdb6e3a62f30774cde6e96411aebd888222a34b369f3d04
3
+ size 243
ANE/KokoroNoise_v2.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9911047f924b41f8b92811c32c50b7c718bd7b0c14842b3bc1b66b9ffe341a19
3
+ size 440
ANE/KokoroNoise_v2.mlmodelc/metadata.json ADDED
@@ -0,0 +1,111 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Mixed (Float32, Palettized (8 bits))",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float32",
10
+ "formattedType" : "MultiArray (Float32)",
11
+ "shortDescription" : "",
12
+ "shape" : "[]",
13
+ "name" : "x_source_0",
14
+ "type" : "MultiArray"
15
+ },
16
+ {
17
+ "hasShapeFlexibility" : "0",
18
+ "isOptional" : "0",
19
+ "dataType" : "Float32",
20
+ "formattedType" : "MultiArray (Float32)",
21
+ "shortDescription" : "",
22
+ "shape" : "[]",
23
+ "name" : "x_source_1",
24
+ "type" : "MultiArray"
25
+ }
26
+ ],
27
+ "modelParameters" : [
28
+
29
+ ],
30
+ "specificationVersion" : 8,
31
+ "mlProgramOperationTypeHistogram" : {
32
+ "Ios16.cumsum" : 1,
33
+ "Ios17.concat" : 1,
34
+ "UpsampleNearestNeighbor" : 1,
35
+ "Ios17.equal" : 1,
36
+ "Ios17.logicalAnd" : 5,
37
+ "Ios17.reshape" : 12,
38
+ "Ios17.instanceNorm" : 12,
39
+ "Ios17.transpose" : 5,
40
+ "Ios17.sin" : 13,
41
+ "Split" : 12,
42
+ "Ios17.expandDims" : 4,
43
+ "Ios16.avgPool" : 1,
44
+ "Ios17.add" : 50,
45
+ "Ios17.squeeze" : 3,
46
+ "Pad" : 1,
47
+ "Ios16.upsampleBilinear" : 1,
48
+ "Ios17.sqrt" : 1,
49
+ "Ios17.sub" : 4,
50
+ "Ios16.constexprLutToDense" : 26,
51
+ "Ios17.conv" : 16,
52
+ "Ios17.tanh" : 1,
53
+ "Ios17.linear" : 13,
54
+ "Ios17.pow" : 14,
55
+ "Ios17.cast" : 6,
56
+ "Ios17.less" : 3,
57
+ "Ios17.realDiv" : 1,
58
+ "Ios17.atan" : 1,
59
+ "Ios17.greater" : 3,
60
+ "Ios17.mul" : 53
61
+ },
62
+ "computePrecision" : "Mixed (Float32, Int32)",
63
+ "isUpdatable" : "0",
64
+ "stateSchema" : [
65
+
66
+ ],
67
+ "availability" : {
68
+ "macOS" : "14.0",
69
+ "tvOS" : "17.0",
70
+ "visionOS" : "1.0",
71
+ "watchOS" : "10.0",
72
+ "iOS" : "17.0",
73
+ "macCatalyst" : "17.0"
74
+ },
75
+ "modelType" : {
76
+ "name" : "MLModelType_mlProgram"
77
+ },
78
+ "userDefinedMetadata" : {
79
+ "com.github.apple.coremltools.conversion_date" : "2026-04-27",
80
+ "com.github.apple.coremltools.source" : "torch==2.11.0",
81
+ "com.github.apple.coremltools.version" : "9.0",
82
+ "com.github.apple.coremltools.source_dialect" : "TorchScript"
83
+ },
84
+ "inputSchema" : [
85
+ {
86
+ "dataType" : "Float32",
87
+ "hasShapeFlexibility" : "1",
88
+ "isOptional" : "0",
89
+ "shapeFlexibility" : "1 × 2...4000",
90
+ "shapeRange" : "[[1, 1], [2, 4000]]",
91
+ "formattedType" : "MultiArray (Float32 1 × 240)",
92
+ "type" : "MultiArray",
93
+ "shape" : "[1, 240]",
94
+ "name" : "F0_curve",
95
+ "shortDescription" : ""
96
+ },
97
+ {
98
+ "hasShapeFlexibility" : "0",
99
+ "isOptional" : "0",
100
+ "dataType" : "Float32",
101
+ "formattedType" : "MultiArray (Float32 1 × 128)",
102
+ "shortDescription" : "",
103
+ "shape" : "[1, 128]",
104
+ "name" : "style_timbre",
105
+ "type" : "MultiArray"
106
+ }
107
+ ],
108
+ "generatedClassName" : "KokoroNoise",
109
+ "method" : "predict"
110
+ }
111
+ ]
ANE/KokoroNoise_v2.mlmodelc/model.mil ADDED
@@ -0,0 +1,546 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ program(1.0)
2
+ [buildInfo = dict<tensor<string, []>, tensor<string, []>>({{"coremlc-component-MIL", "3520.4.1"}, {"coremlc-version", "3520.5.1"}})]
3
+ {
4
+ func main<ios17>(tensor<fp32, [1, ?]> F0_curve, tensor<fp32, [1, 128]> style_timbre) [FlexibleShapeInformation = tuple<tuple<tensor<string, []>, dict<tensor<string, []>, tensor<int32, [?]>>>, tuple<tensor<string, []>, dict<tensor<string, []>, list<tensor<int32, [2]>, ?>>>>((("DefaultShapes", {{"F0_curve", [1, 240]}}), ("RangeDims", {{"F0_curve", [[1, 1], [2, 4000]]}})))] {
5
+ tensor<fp32, [1]> m_source_l_linear_bias = const()[name = tensor<string, []>("m_source_l_linear_bias"), val = tensor<fp32, [1]>([-0x1.e28358p-6])];
6
+ tensor<fp32, [1, 9]> m_source_l_linear_weight = const()[name = tensor<string, []>("m_source_l_linear_weight"), val = tensor<fp32, [1, 9]>([[-0x1.4dfed8p-4, -0x1.7b4864p-3, -0x1.7608cep-3, -0x1.6d4e54p-3, -0x1.946f4ap-4, 0x1.527ebcp-4, 0x1.66277ap-4, -0x1.900fdap-2, -0x1.1871f2p-1]])];
7
+ tensor<fp32, [11, 1, 20]> stft_conv_real_weight = const()[name = tensor<string, []>("stft_conv_real_weight"), val = tensor<fp32, [11, 1, 20]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(64)))];
8
+ tensor<fp32, [11, 1, 20]> stft_conv_imag_weight = const()[name = tensor<string, []>("stft_conv_imag_weight"), val = tensor<fp32, [11, 1, 20]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1024)))];
9
+ tensor<fp32, [256]> noise_convs_0_bias = const()[name = tensor<string, []>("noise_convs_0_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1984)))];
10
+ tensor<fp32, [256, 22, 12]> noise_convs_0_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [67584]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3072))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(70720))), name = tensor<string, []>("noise_convs_0_weight_palettized"), shape = tensor<uint32, [3]>([256, 22, 12])];
11
+ tensor<fp32, [1, 256, 1]> noise_res_0_alpha2_2 = const()[name = tensor<string, []>("noise_res_0_alpha2_2"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(71808)))];
12
+ tensor<fp32, [1, 256, 1]> noise_res_0_alpha1_2 = const()[name = tensor<string, []>("noise_res_0_alpha1_2"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(72896)))];
13
+ tensor<fp32, [1, 256, 1]> noise_res_0_alpha2_1 = const()[name = tensor<string, []>("noise_res_0_alpha2_1"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(73984)))];
14
+ tensor<fp32, [1, 256, 1]> noise_res_0_alpha1_1 = const()[name = tensor<string, []>("noise_res_0_alpha1_1"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(75072)))];
15
+ tensor<fp32, [1, 256, 1]> noise_res_0_alpha2_0 = const()[name = tensor<string, []>("noise_res_0_alpha2_0"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(76160)))];
16
+ tensor<fp32, [1, 256, 1]> noise_res_0_alpha1_0 = const()[name = tensor<string, []>("noise_res_0_alpha1_0"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(77248)))];
17
+ tensor<fp32, [512]> noise_res_0_adain1_0_fc_bias = const()[name = tensor<string, []>("noise_res_0_adain1_0_fc_bias"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(78336)))];
18
+ tensor<fp32, [512, 128]> noise_res_0_adain1_0_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [65536]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(80448))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(146048))), name = tensor<string, []>("noise_res_0_adain1_0_fc_weight_palettized"), shape = tensor<uint32, [2]>([512, 128])];
19
+ tensor<fp32, [256]> noise_res_0_adain1_0_norm_bias = const()[name = tensor<string, []>("noise_res_0_adain1_0_norm_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(147136)))];
20
+ tensor<fp32, [256]> noise_res_0_adain1_0_norm_weight = const()[name = tensor<string, []>("noise_res_0_adain1_0_norm_weight"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(148224)))];
21
+ tensor<fp32, [256]> noise_res_0_convs1_0_bias = const()[name = tensor<string, []>("noise_res_0_convs1_0_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(149312)))];
22
+ tensor<fp32, [512]> noise_res_0_adain2_0_fc_bias = const()[name = tensor<string, []>("noise_res_0_adain2_0_fc_bias"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(150400)))];
23
+ tensor<fp32, [512, 128]> noise_res_0_adain2_0_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [65536]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(152512))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(218112))), name = tensor<string, []>("noise_res_0_adain2_0_fc_weight_palettized"), shape = tensor<uint32, [2]>([512, 128])];
24
+ tensor<fp32, [256]> noise_res_0_convs2_0_bias = const()[name = tensor<string, []>("noise_res_0_convs2_0_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(219200)))];
25
+ tensor<fp32, [512]> noise_res_0_adain1_1_fc_bias = const()[name = tensor<string, []>("noise_res_0_adain1_1_fc_bias"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(220288)))];
26
+ tensor<fp32, [512, 128]> noise_res_0_adain1_1_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [65536]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(222400))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(288000))), name = tensor<string, []>("noise_res_0_adain1_1_fc_weight_palettized"), shape = tensor<uint32, [2]>([512, 128])];
27
+ tensor<fp32, [256]> noise_res_0_convs1_1_bias = const()[name = tensor<string, []>("noise_res_0_convs1_1_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(289088)))];
28
+ tensor<fp32, [512]> noise_res_0_adain2_1_fc_bias = const()[name = tensor<string, []>("noise_res_0_adain2_1_fc_bias"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(290176)))];
29
+ tensor<fp32, [512, 128]> noise_res_0_adain2_1_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [65536]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(292288))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(357888))), name = tensor<string, []>("noise_res_0_adain2_1_fc_weight_palettized"), shape = tensor<uint32, [2]>([512, 128])];
30
+ tensor<fp32, [256]> noise_res_0_convs2_1_bias = const()[name = tensor<string, []>("noise_res_0_convs2_1_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(358976)))];
31
+ tensor<fp32, [512]> noise_res_0_adain1_2_fc_bias = const()[name = tensor<string, []>("noise_res_0_adain1_2_fc_bias"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(360064)))];
32
+ tensor<fp32, [512, 128]> noise_res_0_adain1_2_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [65536]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(362176))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(427776))), name = tensor<string, []>("noise_res_0_adain1_2_fc_weight_palettized"), shape = tensor<uint32, [2]>([512, 128])];
33
+ tensor<fp32, [256]> noise_res_0_convs1_2_bias = const()[name = tensor<string, []>("noise_res_0_convs1_2_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(428864)))];
34
+ tensor<fp32, [512]> noise_res_0_adain2_2_fc_bias = const()[name = tensor<string, []>("noise_res_0_adain2_2_fc_bias"), val = tensor<fp32, [512]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(429952)))];
35
+ tensor<fp32, [512, 128]> noise_res_0_adain2_2_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [65536]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(432064))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(497664))), name = tensor<string, []>("noise_res_0_adain2_2_fc_weight_palettized"), shape = tensor<uint32, [2]>([512, 128])];
36
+ tensor<fp32, [256]> noise_res_0_convs2_2_bias = const()[name = tensor<string, []>("noise_res_0_convs2_2_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(498752)))];
37
+ tensor<fp32, [128]> noise_convs_1_bias = const()[name = tensor<string, []>("noise_convs_1_bias"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(499840)))];
38
+ tensor<fp32, [128, 22, 1]> noise_convs_1_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [2816]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(500416))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(503296))), name = tensor<string, []>("noise_convs_1_weight_palettized"), shape = tensor<uint32, [3]>([128, 22, 1])];
39
+ tensor<fp32, [1, 128, 1]> noise_res_1_alpha2_2 = const()[name = tensor<string, []>("noise_res_1_alpha2_2"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(504384)))];
40
+ tensor<fp32, [1, 128, 1]> noise_res_1_alpha1_2 = const()[name = tensor<string, []>("noise_res_1_alpha1_2"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(504960)))];
41
+ tensor<fp32, [1, 128, 1]> noise_res_1_alpha2_1 = const()[name = tensor<string, []>("noise_res_1_alpha2_1"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(505536)))];
42
+ tensor<fp32, [1, 128, 1]> noise_res_1_alpha1_1 = const()[name = tensor<string, []>("noise_res_1_alpha1_1"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(506112)))];
43
+ tensor<fp32, [1, 128, 1]> noise_res_1_alpha2_0 = const()[name = tensor<string, []>("noise_res_1_alpha2_0"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(506688)))];
44
+ tensor<fp32, [1, 128, 1]> noise_res_1_alpha1_0 = const()[name = tensor<string, []>("noise_res_1_alpha1_0"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(507264)))];
45
+ tensor<fp32, [256]> noise_res_1_adain1_0_fc_bias = const()[name = tensor<string, []>("noise_res_1_adain1_0_fc_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(507840)))];
46
+ tensor<fp32, [256, 128]> noise_res_1_adain1_0_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [32768]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(508928))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(541760))), name = tensor<string, []>("noise_res_1_adain1_0_fc_weight_palettized"), shape = tensor<uint32, [2]>([256, 128])];
47
+ tensor<fp32, [128]> noise_res_1_adain1_0_norm_bias = const()[name = tensor<string, []>("noise_res_1_adain1_0_norm_bias"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(542848)))];
48
+ tensor<fp32, [128]> noise_res_1_adain1_0_norm_weight = const()[name = tensor<string, []>("noise_res_1_adain1_0_norm_weight"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(543424)))];
49
+ tensor<fp32, [128]> noise_res_1_convs1_0_bias = const()[name = tensor<string, []>("noise_res_1_convs1_0_bias"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(544000)))];
50
+ tensor<fp32, [256]> noise_res_1_adain2_0_fc_bias = const()[name = tensor<string, []>("noise_res_1_adain2_0_fc_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(544576)))];
51
+ tensor<fp32, [256, 128]> noise_res_1_adain2_0_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [32768]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(545664))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(578496))), name = tensor<string, []>("noise_res_1_adain2_0_fc_weight_palettized"), shape = tensor<uint32, [2]>([256, 128])];
52
+ tensor<fp32, [128]> noise_res_1_convs2_0_bias = const()[name = tensor<string, []>("noise_res_1_convs2_0_bias"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(579584)))];
53
+ tensor<fp32, [256]> noise_res_1_adain1_1_fc_bias = const()[name = tensor<string, []>("noise_res_1_adain1_1_fc_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(580160)))];
54
+ tensor<fp32, [256, 128]> noise_res_1_adain1_1_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [32768]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(581248))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(614080))), name = tensor<string, []>("noise_res_1_adain1_1_fc_weight_palettized"), shape = tensor<uint32, [2]>([256, 128])];
55
+ tensor<fp32, [128]> noise_res_1_convs1_1_bias = const()[name = tensor<string, []>("noise_res_1_convs1_1_bias"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(615168)))];
56
+ tensor<fp32, [256]> noise_res_1_adain2_1_fc_bias = const()[name = tensor<string, []>("noise_res_1_adain2_1_fc_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(615744)))];
57
+ tensor<fp32, [256, 128]> noise_res_1_adain2_1_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [32768]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(616832))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(649664))), name = tensor<string, []>("noise_res_1_adain2_1_fc_weight_palettized"), shape = tensor<uint32, [2]>([256, 128])];
58
+ tensor<fp32, [128]> noise_res_1_convs2_1_bias = const()[name = tensor<string, []>("noise_res_1_convs2_1_bias"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(650752)))];
59
+ tensor<fp32, [256]> noise_res_1_adain1_2_fc_bias = const()[name = tensor<string, []>("noise_res_1_adain1_2_fc_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(651328)))];
60
+ tensor<fp32, [256, 128]> noise_res_1_adain1_2_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [32768]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(652416))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(685248))), name = tensor<string, []>("noise_res_1_adain1_2_fc_weight_palettized"), shape = tensor<uint32, [2]>([256, 128])];
61
+ tensor<fp32, [128]> noise_res_1_convs1_2_bias = const()[name = tensor<string, []>("noise_res_1_convs1_2_bias"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(686336)))];
62
+ tensor<fp32, [256]> noise_res_1_adain2_2_fc_bias = const()[name = tensor<string, []>("noise_res_1_adain2_2_fc_bias"), val = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(686912)))];
63
+ tensor<fp32, [256, 128]> noise_res_1_adain2_2_fc_weight_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [32768]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(688000))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(720832))), name = tensor<string, []>("noise_res_1_adain2_2_fc_weight_palettized"), shape = tensor<uint32, [2]>([256, 128])];
64
+ tensor<fp32, [128]> noise_res_1_convs2_2_bias = const()[name = tensor<string, []>("noise_res_1_convs2_2_bias"), val = tensor<fp32, [128]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(721920)))];
65
+ tensor<int32, [1]> input_1_axes_0 = const()[name = tensor<string, []>("input_1_axes_0"), val = tensor<int32, [1]>([1])];
66
+ tensor<fp32, [1, 1, ?]> input_1 = expand_dims(axes = input_1_axes_0, x = F0_curve)[name = tensor<string, []>("input_1")];
67
+ tensor<int32, [1]> expand_dims_0_axes_0 = const()[name = tensor<string, []>("expand_dims_0_axes_0"), val = tensor<int32, [1]>([3])];
68
+ tensor<fp32, [1, 1, ?, 1]> expand_dims_0 = expand_dims(axes = expand_dims_0_axes_0, x = input_1)[name = tensor<string, []>("expand_dims_0")];
69
+ tensor<int32, []> upsample_nearest_neighbor_0_scale_factor_height_0 = const()[name = tensor<string, []>("upsample_nearest_neighbor_0_scale_factor_height_0"), val = tensor<int32, []>(300)];
70
+ tensor<int32, []> upsample_nearest_neighbor_0_scale_factor_width_0 = const()[name = tensor<string, []>("upsample_nearest_neighbor_0_scale_factor_width_0"), val = tensor<int32, []>(1)];
71
+ tensor<fp32, [1, 1, ?, 1]> upsample_nearest_neighbor_0 = upsample_nearest_neighbor(scale_factor_height = upsample_nearest_neighbor_0_scale_factor_height_0, scale_factor_width = upsample_nearest_neighbor_0_scale_factor_width_0, x = expand_dims_0)[name = tensor<string, []>("upsample_nearest_neighbor_0")];
72
+ tensor<int32, [1]> var_26_axes_0 = const()[name = tensor<string, []>("op_26_axes_0"), val = tensor<int32, [1]>([3])];
73
+ tensor<fp32, [1, 1, ?]> var_26 = squeeze(axes = var_26_axes_0, x = upsample_nearest_neighbor_0)[name = tensor<string, []>("op_26")];
74
+ tensor<int32, []> var_30 = const()[name = tensor<string, []>("op_30"), val = tensor<int32, []>(1)];
75
+ tensor<fp32, [1, 9, 1]> const_26 = const()[name = tensor<string, []>("const_26"), val = tensor<fp32, [1, 9, 1]>([[[0x1p+0], [0x1p+1], [0x1.8p+1], [0x1p+2], [0x1.4p+2], [0x1.8p+2], [0x1.cp+2], [0x1p+3], [0x1.2p+3]]])];
76
+ tensor<fp32, [1, 9, ?]> fn = mul(x = var_26, y = const_26)[name = tensor<string, []>("fn")];
77
+ tensor<fp32, []> _inversed_rad_values_y_0 = const()[name = tensor<string, []>("_inversed_rad_values_y_0"), val = tensor<fp32, []>(0x1.5d867cp-15)];
78
+ tensor<fp32, [1, 9, ?]> _inversed_rad_values = mul(x = fn, y = _inversed_rad_values_y_0)[name = tensor<string, []>("_inversed_rad_values")];
79
+ tensor<int32, [1]> var_50 = const()[name = tensor<string, []>("op_50"), val = tensor<int32, [1]>([300])];
80
+ tensor<int32, [1]> var_51 = const()[name = tensor<string, []>("op_51"), val = tensor<int32, [1]>([300])];
81
+ tensor<string, []> rv_down_pad_type_0 = const()[name = tensor<string, []>("rv_down_pad_type_0"), val = tensor<string, []>("custom")];
82
+ tensor<int32, [2]> rv_down_pad_0 = const()[name = tensor<string, []>("rv_down_pad_0"), val = tensor<int32, [2]>([0, 0])];
83
+ tensor<bool, []> rv_down_exclude_padding_from_average_0 = const()[name = tensor<string, []>("rv_down_exclude_padding_from_average_0"), val = tensor<bool, []>(false)];
84
+ tensor<bool, []> rv_down_ceil_mode_0 = const()[name = tensor<string, []>("rv_down_ceil_mode_0"), val = tensor<bool, []>(false)];
85
+ tensor<fp32, [1, 9, ?]> rv_down = avg_pool(ceil_mode = rv_down_ceil_mode_0, exclude_padding_from_average = rv_down_exclude_padding_from_average_0, kernel_sizes = var_50, pad = rv_down_pad_0, pad_type = rv_down_pad_type_0, strides = var_51, x = _inversed_rad_values)[name = tensor<string, []>("rv_down")];
86
+ tensor<int32, [3]> rad_values_down_perm_0 = const()[name = tensor<string, []>("rad_values_down_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
87
+ tensor<bool, []> var_55_exclusive_0 = const()[name = tensor<string, []>("op_55_exclusive_0"), val = tensor<bool, []>(false)];
88
+ tensor<bool, []> var_55_reverse_0 = const()[name = tensor<string, []>("op_55_reverse_0"), val = tensor<bool, []>(false)];
89
+ tensor<fp32, [1, ?, 9]> rad_values_down = transpose(perm = rad_values_down_perm_0, x = rv_down)[name = tensor<string, []>("transpose_4")];
90
+ tensor<fp32, [1, ?, 9]> var_55 = cumsum(axis = var_30, exclusive = var_55_exclusive_0, reverse = var_55_reverse_0, x = rad_values_down)[name = tensor<string, []>("op_55")];
91
+ tensor<fp32, []> var_56 = const()[name = tensor<string, []>("op_56"), val = tensor<fp32, []>(0x1.921fb6p+2)];
92
+ tensor<fp32, [1, ?, 9]> phase_1 = mul(x = var_55, y = var_56)[name = tensor<string, []>("phase_1")];
93
+ tensor<int32, [3]> var_58_perm_0 = const()[name = tensor<string, []>("op_58_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
94
+ tensor<fp32, []> var_59_promoted = const()[name = tensor<string, []>("op_59_promoted"), val = tensor<fp32, []>(0x1.2cp+8)];
95
+ tensor<fp32, [1, 9, ?]> var_58 = transpose(perm = var_58_perm_0, x = phase_1)[name = tensor<string, []>("transpose_3")];
96
+ tensor<fp32, [1, 9, ?]> input_3 = mul(x = var_58, y = var_59_promoted)[name = tensor<string, []>("input_3")];
97
+ tensor<int32, [1]> expand_dims_1_axes_0 = const()[name = tensor<string, []>("expand_dims_1_axes_0"), val = tensor<int32, [1]>([3])];
98
+ tensor<fp32, [1, 9, ?, 1]> expand_dims_1 = expand_dims(axes = expand_dims_1_axes_0, x = input_3)[name = tensor<string, []>("expand_dims_1")];
99
+ tensor<int32, []> upsample_bilinear_0_scale_factor_height_0 = const()[name = tensor<string, []>("upsample_bilinear_0_scale_factor_height_0"), val = tensor<int32, []>(300)];
100
+ tensor<bool, []> upsample_bilinear_0_align_corners_0 = const()[name = tensor<string, []>("upsample_bilinear_0_align_corners_0"), val = tensor<bool, []>(false)];
101
+ tensor<int32, []> upsample_bilinear_0_scale_factor_width_0 = const()[name = tensor<string, []>("upsample_bilinear_0_scale_factor_width_0"), val = tensor<int32, []>(1)];
102
+ tensor<fp32, [1, 9, ?, 1]> upsample_bilinear_0 = upsample_bilinear(align_corners = upsample_bilinear_0_align_corners_0, scale_factor_height = upsample_bilinear_0_scale_factor_height_0, scale_factor_width = upsample_bilinear_0_scale_factor_width_0, x = expand_dims_1)[name = tensor<string, []>("upsample_bilinear_0")];
103
+ tensor<int32, [1]> ph_up_axes_0 = const()[name = tensor<string, []>("ph_up_axes_0"), val = tensor<int32, [1]>([3])];
104
+ tensor<fp32, [1, 9, ?]> ph_up = squeeze(axes = ph_up_axes_0, x = upsample_bilinear_0)[name = tensor<string, []>("ph_up")];
105
+ tensor<int32, [3]> phase_3_perm_0 = const()[name = tensor<string, []>("phase_3_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
106
+ tensor<fp32, [1, ?, 9]> phase_3 = transpose(perm = phase_3_perm_0, x = ph_up)[name = tensor<string, []>("transpose_2")];
107
+ tensor<fp32, [1, ?, 9]> var_64 = sin(x = phase_3)[name = tensor<string, []>("op_64")];
108
+ tensor<fp32, []> var_65 = const()[name = tensor<string, []>("op_65"), val = tensor<fp32, []>(0x1.99999ap-4)];
109
+ tensor<fp32, [1, ?, 9]> sines = mul(x = var_64, y = var_65)[name = tensor<string, []>("sines")];
110
+ tensor<fp32, []> var_31_promoted = const()[name = tensor<string, []>("op_31_promoted"), val = tensor<fp32, []>(0x1.4p+3)];
111
+ tensor<int32, [3]> transpose_0_perm_0 = const()[name = tensor<string, []>("transpose_0_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
112
+ tensor<fp32, [1, ?, 1]> transpose_0 = transpose(perm = transpose_0_perm_0, x = var_26)[name = tensor<string, []>("transpose_1")];
113
+ tensor<bool, [1, ?, 1]> var_67 = greater(x = transpose_0, y = var_31_promoted)[name = tensor<string, []>("op_67")];
114
+ tensor<string, []> uv_dtype_0 = const()[name = tensor<string, []>("uv_dtype_0"), val = tensor<string, []>("fp32")];
115
+ tensor<fp32, []> var_69 = const()[name = tensor<string, []>("op_69"), val = tensor<fp32, []>(0x1.89374cp-9)];
116
+ tensor<fp32, [1, ?, 1]> uv = cast(dtype = uv_dtype_0, x = var_67)[name = tensor<string, []>("cast_5")];
117
+ tensor<fp32, [1, ?, 1]> var_70 = mul(x = uv, y = var_69)[name = tensor<string, []>("op_70")];
118
+ tensor<fp32, []> var_30_promoted = const()[name = tensor<string, []>("op_30_promoted"), val = tensor<fp32, []>(0x1p+0)];
119
+ tensor<fp32, [1, ?, 1]> var_71 = sub(x = var_30_promoted, y = uv)[name = tensor<string, []>("op_71")];
120
+ tensor<fp32, []> var_72 = const()[name = tensor<string, []>("op_72"), val = tensor<fp32, []>(0x1.99999ap-4)];
121
+ tensor<fp32, [1, ?, 1]> var_73 = mul(x = var_71, y = var_72)[name = tensor<string, []>("op_73")];
122
+ tensor<fp32, []> _inversed_75_y_0 = const()[name = tensor<string, []>("_inversed_75_y_0"), val = tensor<fp32, []>(0x1.555556p-2)];
123
+ tensor<fp32, [1, ?, 1]> _inversed_75 = mul(x = var_73, y = _inversed_75_y_0)[name = tensor<string, []>("_inversed_75")];
124
+ tensor<fp32, [1, ?, 1]> noise_amp = add(x = var_70, y = _inversed_75)[name = tensor<string, []>("noise_amp")];
125
+ tensor<fp32, []> var_77 = const()[name = tensor<string, []>("op_77"), val = tensor<fp32, []>(0x1.47ae14p-7)];
126
+ tensor<fp32, [1, ?, 1]> noise = mul(x = noise_amp, y = var_77)[name = tensor<string, []>("noise")];
127
+ tensor<fp32, [1, ?, 9]> var_79 = mul(x = sines, y = uv)[name = tensor<string, []>("op_79")];
128
+ tensor<fp32, [1, ?, 9]> input_5 = add(x = var_79, y = noise)[name = tensor<string, []>("input_5")];
129
+ tensor<fp32, [1, ?, 1]> input_7 = linear(bias = m_source_l_linear_bias, weight = m_source_l_linear_weight, x = input_5)[name = tensor<string, []>("linear_0")];
130
+ tensor<fp32, [1, ?, 1]> har_source = tanh(x = input_7)[name = tensor<string, []>("har_source")];
131
+ tensor<int32, [3]> var_90_perm_0 = const()[name = tensor<string, []>("op_90_perm_0"), val = tensor<int32, [3]>([0, 2, 1])];
132
+ tensor<int32, [1]> input_9_axes_0 = const()[name = tensor<string, []>("input_9_axes_0"), val = tensor<int32, [1]>([1])];
133
+ tensor<fp32, [1, 1, ?]> var_90 = transpose(perm = var_90_perm_0, x = har_source)[name = tensor<string, []>("transpose_0")];
134
+ tensor<fp32, [1, ?]> input_9 = squeeze(axes = input_9_axes_0, x = var_90)[name = tensor<string, []>("input_9")];
135
+ tensor<fp32, []> const_1 = const()[name = tensor<string, []>("const_1"), val = tensor<fp32, []>(0x0p+0)];
136
+ tensor<int32, [4]> waveform_pad_0 = const()[name = tensor<string, []>("waveform_pad_0"), val = tensor<int32, [4]>([0, 0, 10, 10])];
137
+ tensor<string, []> waveform_mode_0 = const()[name = tensor<string, []>("waveform_mode_0"), val = tensor<string, []>("replicate")];
138
+ tensor<fp32, [1, ?]> waveform = pad(constant_val = const_1, mode = waveform_mode_0, pad = waveform_pad_0, x = input_9)[name = tensor<string, []>("waveform")];
139
+ tensor<int32, [1]> input_11_axes_0 = const()[name = tensor<string, []>("input_11_axes_0"), val = tensor<int32, [1]>([1])];
140
+ tensor<fp32, [1, 1, ?]> input_11 = expand_dims(axes = input_11_axes_0, x = waveform)[name = tensor<string, []>("input_11")];
141
+ tensor<string, []> real_out_pad_type_0 = const()[name = tensor<string, []>("real_out_pad_type_0"), val = tensor<string, []>("valid")];
142
+ tensor<int32, [1]> real_out_strides_0 = const()[name = tensor<string, []>("real_out_strides_0"), val = tensor<int32, [1]>([5])];
143
+ tensor<int32, [2]> real_out_pad_0 = const()[name = tensor<string, []>("real_out_pad_0"), val = tensor<int32, [2]>([0, 0])];
144
+ tensor<int32, [1]> real_out_dilations_0 = const()[name = tensor<string, []>("real_out_dilations_0"), val = tensor<int32, [1]>([1])];
145
+ tensor<int32, []> real_out_groups_0 = const()[name = tensor<string, []>("real_out_groups_0"), val = tensor<int32, []>(1)];
146
+ tensor<fp32, [1, 11, ?]> real_out = conv(dilations = real_out_dilations_0, groups = real_out_groups_0, pad = real_out_pad_0, pad_type = real_out_pad_type_0, strides = real_out_strides_0, weight = stft_conv_real_weight, x = input_11)[name = tensor<string, []>("real_out")];
147
+ tensor<string, []> imag_out_pad_type_0 = const()[name = tensor<string, []>("imag_out_pad_type_0"), val = tensor<string, []>("valid")];
148
+ tensor<int32, [1]> imag_out_strides_0 = const()[name = tensor<string, []>("imag_out_strides_0"), val = tensor<int32, [1]>([5])];
149
+ tensor<int32, [2]> imag_out_pad_0 = const()[name = tensor<string, []>("imag_out_pad_0"), val = tensor<int32, [2]>([0, 0])];
150
+ tensor<int32, [1]> imag_out_dilations_0 = const()[name = tensor<string, []>("imag_out_dilations_0"), val = tensor<int32, [1]>([1])];
151
+ tensor<int32, []> imag_out_groups_0 = const()[name = tensor<string, []>("imag_out_groups_0"), val = tensor<int32, []>(1)];
152
+ tensor<fp32, [1, 11, ?]> imag_out = conv(dilations = imag_out_dilations_0, groups = imag_out_groups_0, pad = imag_out_pad_0, pad_type = imag_out_pad_type_0, strides = imag_out_strides_0, weight = stft_conv_imag_weight, x = input_11)[name = tensor<string, []>("imag_out")];
153
+ tensor<fp32, [1, 11, ?]> var_125 = abs(x = imag_out)[name = tensor<string, []>("op_125")];
154
+ tensor<fp32, []> var_126 = const()[name = tensor<string, []>("op_126"), val = tensor<fp32, []>(0x1.4f8b58p-17)];
155
+ tensor<bool, [1, 11, ?]> var_127 = less(x = var_125, y = var_126)[name = tensor<string, []>("op_127")];
156
+ tensor<fp32, [1, 11, ?]> var_133 = sub(x = imag_out, y = imag_out)[name = tensor<string, []>("sub_0")];
157
+ tensor<fp32, [1, 11, ?]> imag_clipped = select(a = var_133, b = imag_out, cond = var_127)[name = tensor<string, []>("imag_clipped")];
158
+ tensor<fp32, []> var_135_promoted = const()[name = tensor<string, []>("op_135_promoted"), val = tensor<fp32, []>(0x1p+1)];
159
+ tensor<fp32, [1, 11, ?]> var_136 = pow(x = real_out, y = var_135_promoted)[name = tensor<string, []>("op_136")];
160
+ tensor<fp32, []> var_137_promoted = const()[name = tensor<string, []>("op_137_promoted"), val = tensor<fp32, []>(0x1p+1)];
161
+ tensor<fp32, [1, 11, ?]> var_138 = pow(x = imag_clipped, y = var_137_promoted)[name = tensor<string, []>("op_138")];
162
+ tensor<fp32, [1, 11, ?]> var_140 = add(x = var_136, y = var_138)[name = tensor<string, []>("op_140")];
163
+ tensor<fp32, []> var_142 = const()[name = tensor<string, []>("op_142"), val = tensor<fp32, []>(0x1.6849b8p-47)];
164
+ tensor<fp32, [1, 11, ?]> var_143 = add(x = var_140, y = var_142)[name = tensor<string, []>("op_143")];
165
+ tensor<fp32, [1, 11, ?]> har_spec = sqrt(x = var_143)[name = tensor<string, []>("har_spec")];
166
+ tensor<fp32, []> less_0_y_0 = const()[name = tensor<string, []>("less_0_y_0"), val = tensor<fp32, []>(0x0p+0)];
167
+ tensor<bool, [1, 11, ?]> less_0 = less(x = imag_clipped, y = less_0_y_0)[name = tensor<string, []>("less_0")];
168
+ tensor<fp32, []> greater_0_y_0 = const()[name = tensor<string, []>("greater_0_y_0"), val = tensor<fp32, []>(0x0p+0)];
169
+ tensor<bool, [1, 11, ?]> greater_0 = greater(x = imag_clipped, y = greater_0_y_0)[name = tensor<string, []>("greater_0")];
170
+ tensor<fp32, []> less_1_y_0 = const()[name = tensor<string, []>("less_1_y_0"), val = tensor<fp32, []>(0x0p+0)];
171
+ tensor<bool, [1, 11, ?]> less_1 = less(x = real_out, y = less_1_y_0)[name = tensor<string, []>("less_1")];
172
+ tensor<fp32, []> equal_0_y_0 = const()[name = tensor<string, []>("equal_0_y_0"), val = tensor<fp32, []>(0x0p+0)];
173
+ tensor<bool, [1, 11, ?]> equal_0 = equal(x = real_out, y = equal_0_y_0)[name = tensor<string, []>("equal_0")];
174
+ tensor<bool, [1, 11, ?]> logical_and_0 = logical_and(x = greater_0, y = less_1)[name = tensor<string, []>("logical_and_0")];
175
+ tensor<bool, [1, 11, ?]> logical_and_1 = logical_and(x = less_0, y = less_1)[name = tensor<string, []>("logical_and_1")];
176
+ tensor<bool, [1, 11, ?]> logical_and_2 = logical_and(x = greater_0, y = equal_0)[name = tensor<string, []>("logical_and_2")];
177
+ tensor<bool, [1, 11, ?]> logical_and_3 = logical_and(x = less_0, y = equal_0)[name = tensor<string, []>("logical_and_3")];
178
+ tensor<string, []> cast_5_dtype_0 = const()[name = tensor<string, []>("cast_5_dtype_0"), val = tensor<string, []>("fp32")];
179
+ tensor<string, []> cast_6_dtype_0 = const()[name = tensor<string, []>("cast_6_dtype_0"), val = tensor<string, []>("fp32")];
180
+ tensor<string, []> cast_7_dtype_0 = const()[name = tensor<string, []>("cast_7_dtype_0"), val = tensor<string, []>("fp32")];
181
+ tensor<string, []> cast_8_dtype_0 = const()[name = tensor<string, []>("cast_8_dtype_0"), val = tensor<string, []>("fp32")];
182
+ tensor<fp32, []> mul_0_y_0 = const()[name = tensor<string, []>("mul_0_y_0"), val = tensor<fp32, []>(0x1.921fb6p+1)];
183
+ tensor<fp32, [1, 11, ?]> cast_5 = cast(dtype = cast_5_dtype_0, x = logical_and_0)[name = tensor<string, []>("cast_4")];
184
+ tensor<fp32, [1, 11, ?]> mul_0 = mul(x = cast_5, y = mul_0_y_0)[name = tensor<string, []>("mul_0")];
185
+ tensor<fp32, []> mul_1_y_0 = const()[name = tensor<string, []>("mul_1_y_0"), val = tensor<fp32, []>(0x1.921fb6p+1)];
186
+ tensor<fp32, [1, 11, ?]> cast_6 = cast(dtype = cast_6_dtype_0, x = logical_and_1)[name = tensor<string, []>("cast_3")];
187
+ tensor<fp32, [1, 11, ?]> mul_1 = mul(x = cast_6, y = mul_1_y_0)[name = tensor<string, []>("mul_1")];
188
+ tensor<fp32, []> sub_1_x_0 = const()[name = tensor<string, []>("sub_1_x_0"), val = tensor<fp32, []>(0x1p+0)];
189
+ tensor<fp32, [1, 11, ?]> cast_7 = cast(dtype = cast_7_dtype_0, x = logical_and_2)[name = tensor<string, []>("cast_2")];
190
+ tensor<fp32, [1, 11, ?]> sub_1 = sub(x = sub_1_x_0, y = cast_7)[name = tensor<string, []>("sub_1")];
191
+ tensor<fp32, []> mul_2_y_0 = const()[name = tensor<string, []>("mul_2_y_0"), val = tensor<fp32, []>(0x1.921fb6p+0)];
192
+ tensor<fp32, [1, 11, ?]> mul_2 = mul(x = cast_7, y = mul_2_y_0)[name = tensor<string, []>("mul_2")];
193
+ tensor<fp32, []> sub_2_x_0 = const()[name = tensor<string, []>("sub_2_x_0"), val = tensor<fp32, []>(0x1p+0)];
194
+ tensor<fp32, [1, 11, ?]> cast_8 = cast(dtype = cast_8_dtype_0, x = logical_and_3)[name = tensor<string, []>("cast_1")];
195
+ tensor<fp32, [1, 11, ?]> sub_2 = sub(x = sub_2_x_0, y = cast_8)[name = tensor<string, []>("sub_2")];
196
+ tensor<fp32, []> mul_3_y_0 = const()[name = tensor<string, []>("mul_3_y_0"), val = tensor<fp32, []>(-0x1.921fb6p+0)];
197
+ tensor<fp32, [1, 11, ?]> mul_3 = mul(x = cast_8, y = mul_3_y_0)[name = tensor<string, []>("mul_3")];
198
+ tensor<fp32, []> greater_1_y_0 = const()[name = tensor<string, []>("greater_1_y_0"), val = tensor<fp32, []>(-0x1.5798eep-27)];
199
+ tensor<bool, [1, 11, ?]> greater_1 = greater(x = real_out, y = greater_1_y_0)[name = tensor<string, []>("greater_1")];
200
+ tensor<fp32, []> less_2_y_0 = const()[name = tensor<string, []>("less_2_y_0"), val = tensor<fp32, []>(0x1.5798eep-27)];
201
+ tensor<bool, [1, 11, ?]> less_2 = less(x = real_out, y = less_2_y_0)[name = tensor<string, []>("less_2")];
202
+ tensor<bool, [1, 11, ?]> logical_and_4 = logical_and(x = greater_1, y = less_2)[name = tensor<string, []>("logical_and_4")];
203
+ tensor<string, []> cast_9_dtype_0 = const()[name = tensor<string, []>("cast_9_dtype_0"), val = tensor<string, []>("fp32")];
204
+ tensor<fp32, []> mul_4_y_0 = const()[name = tensor<string, []>("mul_4_y_0"), val = tensor<fp32, []>(0x1.5798eep-26)];
205
+ tensor<fp32, [1, 11, ?]> cast_9 = cast(dtype = cast_9_dtype_0, x = logical_and_4)[name = tensor<string, []>("cast_0")];
206
+ tensor<fp32, [1, 11, ?]> mul_4 = mul(x = cast_9, y = mul_4_y_0)[name = tensor<string, []>("mul_4")];
207
+ tensor<fp32, [1, 11, ?]> add_0 = add(x = real_out, y = mul_4)[name = tensor<string, []>("add_0")];
208
+ tensor<fp32, [1, 11, ?]> real_div_0 = real_div(x = imag_clipped, y = add_0)[name = tensor<string, []>("real_div_0")];
209
+ tensor<fp32, [1, 11, ?]> atan_0 = atan(x = real_div_0)[name = tensor<string, []>("atan_0")];
210
+ tensor<fp32, [1, 11, ?]> add_1 = add(x = atan_0, y = mul_0)[name = tensor<string, []>("add_1")];
211
+ tensor<fp32, [1, 11, ?]> sub_3 = sub(x = add_1, y = mul_1)[name = tensor<string, []>("sub_3")];
212
+ tensor<fp32, [1, 11, ?]> mul_5 = mul(x = sub_3, y = sub_1)[name = tensor<string, []>("mul_5")];
213
+ tensor<fp32, [1, 11, ?]> add_2 = add(x = mul_5, y = mul_2)[name = tensor<string, []>("add_2")];
214
+ tensor<fp32, [1, 11, ?]> mul_6 = mul(x = add_2, y = sub_2)[name = tensor<string, []>("mul_6")];
215
+ tensor<fp32, [1, 11, ?]> phase = add(x = mul_6, y = mul_3)[name = tensor<string, []>("phase")];
216
+ tensor<fp32, []> var_146_promoted = const()[name = tensor<string, []>("op_146_promoted"), val = tensor<fp32, []>(0x0p+0)];
217
+ tensor<bool, [1, 11, ?]> var_147 = equal(x = imag_clipped, y = var_146_promoted)[name = tensor<string, []>("op_147")];
218
+ tensor<bool, [1, 11, ?]> correction_mask = logical_and(x = var_147, y = less_1)[name = tensor<string, []>("correction_mask")];
219
+ tensor<fp32, []> var_157_value_0 = const()[name = tensor<string, []>("op_157_value_0"), val = tensor<fp32, []>(0x1.921fb6p+1)];
220
+ tensor<fp32, [1, 11, ?]> var_157 = fill_like(ref_tensor = phase, value = var_157_value_0)[name = tensor<string, []>("op_157")];
221
+ tensor<fp32, [1, 11, ?]> har_phase = select(a = var_157, b = phase, cond = correction_mask)[name = tensor<string, []>("har_phase")];
222
+ tensor<int32, []> var_160 = const()[name = tensor<string, []>("op_160"), val = tensor<int32, []>(1)];
223
+ tensor<bool, []> input_13_interleave_0 = const()[name = tensor<string, []>("input_13_interleave_0"), val = tensor<bool, []>(false)];
224
+ tensor<fp32, [1, 22, ?]> input_13 = concat(axis = var_160, interleave = input_13_interleave_0, values = (har_spec, har_phase))[name = tensor<string, []>("input_13")];
225
+ tensor<string, []> input_15_pad_type_0 = const()[name = tensor<string, []>("input_15_pad_type_0"), val = tensor<string, []>("custom")];
226
+ tensor<int32, [2]> input_15_pad_0 = const()[name = tensor<string, []>("input_15_pad_0"), val = tensor<int32, [2]>([3, 3])];
227
+ tensor<int32, [1]> input_15_strides_0 = const()[name = tensor<string, []>("input_15_strides_0"), val = tensor<int32, [1]>([6])];
228
+ tensor<int32, [1]> input_15_dilations_0 = const()[name = tensor<string, []>("input_15_dilations_0"), val = tensor<int32, [1]>([1])];
229
+ tensor<int32, []> input_15_groups_0 = const()[name = tensor<string, []>("input_15_groups_0"), val = tensor<int32, []>(1)];
230
+ tensor<fp32, [1, 256, ?]> input_15 = conv(bias = noise_convs_0_bias, dilations = input_15_dilations_0, groups = input_15_groups_0, pad = input_15_pad_0, pad_type = input_15_pad_type_0, strides = input_15_strides_0, weight = noise_convs_0_weight_palettized, x = input_13)[name = tensor<string, []>("input_15")];
231
+ tensor<fp32, []> var_183 = const()[name = tensor<string, []>("op_183"), val = tensor<fp32, []>(0x1.4f8b58p-17)];
232
+ tensor<fp32, [1, 512]> h_1 = linear(bias = noise_res_0_adain1_0_fc_bias, weight = noise_res_0_adain1_0_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_1")];
233
+ tensor<int32, [3]> var_266 = const()[name = tensor<string, []>("op_266"), val = tensor<int32, [3]>([1, 512, 1])];
234
+ tensor<fp32, [1, 512, 1]> h_3 = reshape(shape = var_266, x = h_1)[name = tensor<string, []>("h_3")];
235
+ tensor<int32, [2]> var_268_split_sizes_0 = const()[name = tensor<string, []>("op_268_split_sizes_0"), val = tensor<int32, [2]>([256, 256])];
236
+ tensor<int32, []> var_268_axis_0 = const()[name = tensor<string, []>("op_268_axis_0"), val = tensor<int32, []>(1)];
237
+ tensor<fp32, [1, 256, 1]> var_268_0, tensor<fp32, [1, 256, 1]> var_268_1 = split(axis = var_268_axis_0, split_sizes = var_268_split_sizes_0, x = h_3)[name = tensor<string, []>("op_268")];
238
+ tensor<fp32, []> var_270_promoted = const()[name = tensor<string, []>("op_270_promoted"), val = tensor<fp32, []>(0x1p+0)];
239
+ tensor<fp32, [1, 256, 1]> var_271 = add(x = var_268_0, y = var_270_promoted)[name = tensor<string, []>("op_271")];
240
+ tensor<fp32, [1, 256, ?]> var_274 = instance_norm(beta = noise_res_0_adain1_0_norm_bias, epsilon = var_183, gamma = noise_res_0_adain1_0_norm_weight, x = input_15)[name = tensor<string, []>("op_274")];
241
+ tensor<fp32, [1, 256, ?]> var_275 = mul(x = var_271, y = var_274)[name = tensor<string, []>("op_275")];
242
+ tensor<fp32, [1, 256, ?]> xt_1 = add(x = var_275, y = var_268_1)[name = tensor<string, []>("xt_1")];
243
+ tensor<fp32, [1, 256, 1]> var_277 = const()[name = tensor<string, []>("op_277"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(722496)))];
244
+ tensor<fp32, [1, 256, ?]> var_280 = mul(x = noise_res_0_alpha1_0, y = xt_1)[name = tensor<string, []>("op_280")];
245
+ tensor<fp32, [1, 256, ?]> var_281 = sin(x = var_280)[name = tensor<string, []>("op_281")];
246
+ tensor<fp32, []> var_182_promoted = const()[name = tensor<string, []>("op_182_promoted"), val = tensor<fp32, []>(0x1p+1)];
247
+ tensor<fp32, [1, 256, ?]> var_282 = pow(x = var_281, y = var_182_promoted)[name = tensor<string, []>("op_282")];
248
+ tensor<fp32, [1, 256, ?]> var_283 = mul(x = var_277, y = var_282)[name = tensor<string, []>("op_283")];
249
+ tensor<fp32, [1, 256, ?]> input_17 = add(x = xt_1, y = var_283)[name = tensor<string, []>("input_17")];
250
+ tensor<fp32, [256, 256, 7]> weight_9_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [458752]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(723584))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1182400))), name = tensor<string, []>("weight_9_palettized"), shape = tensor<uint32, [3]>([256, 256, 7])];
251
+ tensor<string, []> input_19_pad_type_0 = const()[name = tensor<string, []>("input_19_pad_type_0"), val = tensor<string, []>("custom")];
252
+ tensor<int32, [2]> input_19_pad_0 = const()[name = tensor<string, []>("input_19_pad_0"), val = tensor<int32, [2]>([3, 3])];
253
+ tensor<int32, [1]> input_19_strides_0 = const()[name = tensor<string, []>("input_19_strides_0"), val = tensor<int32, [1]>([1])];
254
+ tensor<int32, [1]> input_19_dilations_0 = const()[name = tensor<string, []>("input_19_dilations_0"), val = tensor<int32, [1]>([1])];
255
+ tensor<int32, []> input_19_groups_0 = const()[name = tensor<string, []>("input_19_groups_0"), val = tensor<int32, []>(1)];
256
+ tensor<fp32, [1, 256, ?]> input_19 = conv(bias = noise_res_0_convs1_0_bias, dilations = input_19_dilations_0, groups = input_19_groups_0, pad = input_19_pad_0, pad_type = input_19_pad_type_0, strides = input_19_strides_0, weight = weight_9_palettized, x = input_17)[name = tensor<string, []>("input_19")];
257
+ tensor<fp32, [1, 512]> h_5 = linear(bias = noise_res_0_adain2_0_fc_bias, weight = noise_res_0_adain2_0_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_2")];
258
+ tensor<int32, [3]> var_299 = const()[name = tensor<string, []>("op_299"), val = tensor<int32, [3]>([1, 512, 1])];
259
+ tensor<fp32, [1, 512, 1]> h_7 = reshape(shape = var_299, x = h_5)[name = tensor<string, []>("h_7")];
260
+ tensor<int32, [2]> var_301_split_sizes_0 = const()[name = tensor<string, []>("op_301_split_sizes_0"), val = tensor<int32, [2]>([256, 256])];
261
+ tensor<int32, []> var_301_axis_0 = const()[name = tensor<string, []>("op_301_axis_0"), val = tensor<int32, []>(1)];
262
+ tensor<fp32, [1, 256, 1]> var_301_0, tensor<fp32, [1, 256, 1]> var_301_1 = split(axis = var_301_axis_0, split_sizes = var_301_split_sizes_0, x = h_7)[name = tensor<string, []>("op_301")];
263
+ tensor<fp32, []> var_303_promoted = const()[name = tensor<string, []>("op_303_promoted"), val = tensor<fp32, []>(0x1p+0)];
264
+ tensor<fp32, [1, 256, 1]> var_304 = add(x = var_301_0, y = var_303_promoted)[name = tensor<string, []>("op_304")];
265
+ tensor<fp32, [1, 256, ?]> var_307 = instance_norm(beta = noise_res_0_adain1_0_norm_bias, epsilon = var_183, gamma = noise_res_0_adain1_0_norm_weight, x = input_19)[name = tensor<string, []>("op_307")];
266
+ tensor<fp32, [1, 256, ?]> var_308 = mul(x = var_304, y = var_307)[name = tensor<string, []>("op_308")];
267
+ tensor<fp32, [1, 256, ?]> xt_3 = add(x = var_308, y = var_301_1)[name = tensor<string, []>("xt_3")];
268
+ tensor<fp32, [1, 256, 1]> var_310 = const()[name = tensor<string, []>("op_310"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1183488)))];
269
+ tensor<fp32, [1, 256, ?]> var_313 = mul(x = noise_res_0_alpha2_0, y = xt_3)[name = tensor<string, []>("op_313")];
270
+ tensor<fp32, [1, 256, ?]> var_314 = sin(x = var_313)[name = tensor<string, []>("op_314")];
271
+ tensor<fp32, []> var_182_promoted_1 = const()[name = tensor<string, []>("op_182_promoted_1"), val = tensor<fp32, []>(0x1p+1)];
272
+ tensor<fp32, [1, 256, ?]> var_315 = pow(x = var_314, y = var_182_promoted_1)[name = tensor<string, []>("op_315")];
273
+ tensor<fp32, [1, 256, ?]> var_316 = mul(x = var_310, y = var_315)[name = tensor<string, []>("op_316")];
274
+ tensor<fp32, [1, 256, ?]> input_21 = add(x = xt_3, y = var_316)[name = tensor<string, []>("input_21")];
275
+ tensor<fp32, [256, 256, 7]> weight_13_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [458752]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1184576))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1643392))), name = tensor<string, []>("weight_13_palettized"), shape = tensor<uint32, [3]>([256, 256, 7])];
276
+ tensor<string, []> xt_5_pad_type_0 = const()[name = tensor<string, []>("xt_5_pad_type_0"), val = tensor<string, []>("custom")];
277
+ tensor<int32, [2]> xt_5_pad_0 = const()[name = tensor<string, []>("xt_5_pad_0"), val = tensor<int32, [2]>([3, 3])];
278
+ tensor<int32, [1]> xt_5_strides_0 = const()[name = tensor<string, []>("xt_5_strides_0"), val = tensor<int32, [1]>([1])];
279
+ tensor<int32, [1]> xt_5_dilations_0 = const()[name = tensor<string, []>("xt_5_dilations_0"), val = tensor<int32, [1]>([1])];
280
+ tensor<int32, []> xt_5_groups_0 = const()[name = tensor<string, []>("xt_5_groups_0"), val = tensor<int32, []>(1)];
281
+ tensor<fp32, [1, 256, ?]> xt_5 = conv(bias = noise_res_0_convs2_0_bias, dilations = xt_5_dilations_0, groups = xt_5_groups_0, pad = xt_5_pad_0, pad_type = xt_5_pad_type_0, strides = xt_5_strides_0, weight = weight_13_palettized, x = input_21)[name = tensor<string, []>("xt_5")];
282
+ tensor<fp32, [1, 256, ?]> input_23 = add(x = xt_5, y = input_15)[name = tensor<string, []>("input_23")];
283
+ tensor<fp32, [1, 512]> h_9 = linear(bias = noise_res_0_adain1_1_fc_bias, weight = noise_res_0_adain1_1_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_3")];
284
+ tensor<int32, [3]> var_333 = const()[name = tensor<string, []>("op_333"), val = tensor<int32, [3]>([1, 512, 1])];
285
+ tensor<fp32, [1, 512, 1]> h_11 = reshape(shape = var_333, x = h_9)[name = tensor<string, []>("h_11")];
286
+ tensor<int32, [2]> var_335_split_sizes_0 = const()[name = tensor<string, []>("op_335_split_sizes_0"), val = tensor<int32, [2]>([256, 256])];
287
+ tensor<int32, []> var_335_axis_0 = const()[name = tensor<string, []>("op_335_axis_0"), val = tensor<int32, []>(1)];
288
+ tensor<fp32, [1, 256, 1]> var_335_0, tensor<fp32, [1, 256, 1]> var_335_1 = split(axis = var_335_axis_0, split_sizes = var_335_split_sizes_0, x = h_11)[name = tensor<string, []>("op_335")];
289
+ tensor<fp32, []> var_337_promoted = const()[name = tensor<string, []>("op_337_promoted"), val = tensor<fp32, []>(0x1p+0)];
290
+ tensor<fp32, [1, 256, 1]> var_338 = add(x = var_335_0, y = var_337_promoted)[name = tensor<string, []>("op_338")];
291
+ tensor<fp32, [1, 256, ?]> var_341 = instance_norm(beta = noise_res_0_adain1_0_norm_bias, epsilon = var_183, gamma = noise_res_0_adain1_0_norm_weight, x = input_23)[name = tensor<string, []>("op_341")];
292
+ tensor<fp32, [1, 256, ?]> var_342 = mul(x = var_338, y = var_341)[name = tensor<string, []>("op_342")];
293
+ tensor<fp32, [1, 256, ?]> xt_7 = add(x = var_342, y = var_335_1)[name = tensor<string, []>("xt_7")];
294
+ tensor<fp32, [1, 256, 1]> var_344 = const()[name = tensor<string, []>("op_344"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1644480)))];
295
+ tensor<fp32, [1, 256, ?]> var_347 = mul(x = noise_res_0_alpha1_1, y = xt_7)[name = tensor<string, []>("op_347")];
296
+ tensor<fp32, [1, 256, ?]> var_348 = sin(x = var_347)[name = tensor<string, []>("op_348")];
297
+ tensor<fp32, []> var_182_promoted_2 = const()[name = tensor<string, []>("op_182_promoted_2"), val = tensor<fp32, []>(0x1p+1)];
298
+ tensor<fp32, [1, 256, ?]> var_349 = pow(x = var_348, y = var_182_promoted_2)[name = tensor<string, []>("op_349")];
299
+ tensor<fp32, [1, 256, ?]> var_350 = mul(x = var_344, y = var_349)[name = tensor<string, []>("op_350")];
300
+ tensor<fp32, [1, 256, ?]> input_25 = add(x = xt_7, y = var_350)[name = tensor<string, []>("input_25")];
301
+ tensor<fp32, [256, 256, 7]> weight_17_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [458752]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(1645568))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2104384))), name = tensor<string, []>("weight_17_palettized"), shape = tensor<uint32, [3]>([256, 256, 7])];
302
+ tensor<string, []> input_27_pad_type_0 = const()[name = tensor<string, []>("input_27_pad_type_0"), val = tensor<string, []>("custom")];
303
+ tensor<int32, [2]> input_27_pad_0 = const()[name = tensor<string, []>("input_27_pad_0"), val = tensor<int32, [2]>([9, 9])];
304
+ tensor<int32, [1]> input_27_dilations_0 = const()[name = tensor<string, []>("input_27_dilations_0"), val = tensor<int32, [1]>([3])];
305
+ tensor<int32, [1]> input_27_strides_0 = const()[name = tensor<string, []>("input_27_strides_0"), val = tensor<int32, [1]>([1])];
306
+ tensor<int32, []> input_27_groups_0 = const()[name = tensor<string, []>("input_27_groups_0"), val = tensor<int32, []>(1)];
307
+ tensor<fp32, [1, 256, ?]> input_27 = conv(bias = noise_res_0_convs1_1_bias, dilations = input_27_dilations_0, groups = input_27_groups_0, pad = input_27_pad_0, pad_type = input_27_pad_type_0, strides = input_27_strides_0, weight = weight_17_palettized, x = input_25)[name = tensor<string, []>("input_27")];
308
+ tensor<fp32, [1, 512]> h_13 = linear(bias = noise_res_0_adain2_1_fc_bias, weight = noise_res_0_adain2_1_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_4")];
309
+ tensor<int32, [3]> var_366 = const()[name = tensor<string, []>("op_366"), val = tensor<int32, [3]>([1, 512, 1])];
310
+ tensor<fp32, [1, 512, 1]> h_15 = reshape(shape = var_366, x = h_13)[name = tensor<string, []>("h_15")];
311
+ tensor<int32, [2]> var_368_split_sizes_0 = const()[name = tensor<string, []>("op_368_split_sizes_0"), val = tensor<int32, [2]>([256, 256])];
312
+ tensor<int32, []> var_368_axis_0 = const()[name = tensor<string, []>("op_368_axis_0"), val = tensor<int32, []>(1)];
313
+ tensor<fp32, [1, 256, 1]> var_368_0, tensor<fp32, [1, 256, 1]> var_368_1 = split(axis = var_368_axis_0, split_sizes = var_368_split_sizes_0, x = h_15)[name = tensor<string, []>("op_368")];
314
+ tensor<fp32, []> var_370_promoted = const()[name = tensor<string, []>("op_370_promoted"), val = tensor<fp32, []>(0x1p+0)];
315
+ tensor<fp32, [1, 256, 1]> var_371 = add(x = var_368_0, y = var_370_promoted)[name = tensor<string, []>("op_371")];
316
+ tensor<fp32, [1, 256, ?]> var_374 = instance_norm(beta = noise_res_0_adain1_0_norm_bias, epsilon = var_183, gamma = noise_res_0_adain1_0_norm_weight, x = input_27)[name = tensor<string, []>("op_374")];
317
+ tensor<fp32, [1, 256, ?]> var_375 = mul(x = var_371, y = var_374)[name = tensor<string, []>("op_375")];
318
+ tensor<fp32, [1, 256, ?]> xt_9 = add(x = var_375, y = var_368_1)[name = tensor<string, []>("xt_9")];
319
+ tensor<fp32, [1, 256, 1]> var_377 = const()[name = tensor<string, []>("op_377"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2105472)))];
320
+ tensor<fp32, [1, 256, ?]> var_380 = mul(x = noise_res_0_alpha2_1, y = xt_9)[name = tensor<string, []>("op_380")];
321
+ tensor<fp32, [1, 256, ?]> var_381 = sin(x = var_380)[name = tensor<string, []>("op_381")];
322
+ tensor<fp32, []> var_182_promoted_3 = const()[name = tensor<string, []>("op_182_promoted_3"), val = tensor<fp32, []>(0x1p+1)];
323
+ tensor<fp32, [1, 256, ?]> var_382 = pow(x = var_381, y = var_182_promoted_3)[name = tensor<string, []>("op_382")];
324
+ tensor<fp32, [1, 256, ?]> var_383 = mul(x = var_377, y = var_382)[name = tensor<string, []>("op_383")];
325
+ tensor<fp32, [1, 256, ?]> input_29 = add(x = xt_9, y = var_383)[name = tensor<string, []>("input_29")];
326
+ tensor<fp32, [256, 256, 7]> weight_21_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [458752]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2106560))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2565376))), name = tensor<string, []>("weight_21_palettized"), shape = tensor<uint32, [3]>([256, 256, 7])];
327
+ tensor<string, []> xt_11_pad_type_0 = const()[name = tensor<string, []>("xt_11_pad_type_0"), val = tensor<string, []>("custom")];
328
+ tensor<int32, [2]> xt_11_pad_0 = const()[name = tensor<string, []>("xt_11_pad_0"), val = tensor<int32, [2]>([3, 3])];
329
+ tensor<int32, [1]> xt_11_strides_0 = const()[name = tensor<string, []>("xt_11_strides_0"), val = tensor<int32, [1]>([1])];
330
+ tensor<int32, [1]> xt_11_dilations_0 = const()[name = tensor<string, []>("xt_11_dilations_0"), val = tensor<int32, [1]>([1])];
331
+ tensor<int32, []> xt_11_groups_0 = const()[name = tensor<string, []>("xt_11_groups_0"), val = tensor<int32, []>(1)];
332
+ tensor<fp32, [1, 256, ?]> xt_11 = conv(bias = noise_res_0_convs2_1_bias, dilations = xt_11_dilations_0, groups = xt_11_groups_0, pad = xt_11_pad_0, pad_type = xt_11_pad_type_0, strides = xt_11_strides_0, weight = weight_21_palettized, x = input_29)[name = tensor<string, []>("xt_11")];
333
+ tensor<fp32, [1, 256, ?]> input_31 = add(x = xt_11, y = input_23)[name = tensor<string, []>("input_31")];
334
+ tensor<fp32, [1, 512]> h_17 = linear(bias = noise_res_0_adain1_2_fc_bias, weight = noise_res_0_adain1_2_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_5")];
335
+ tensor<int32, [3]> var_400 = const()[name = tensor<string, []>("op_400"), val = tensor<int32, [3]>([1, 512, 1])];
336
+ tensor<fp32, [1, 512, 1]> h_19 = reshape(shape = var_400, x = h_17)[name = tensor<string, []>("h_19")];
337
+ tensor<int32, [2]> var_402_split_sizes_0 = const()[name = tensor<string, []>("op_402_split_sizes_0"), val = tensor<int32, [2]>([256, 256])];
338
+ tensor<int32, []> var_402_axis_0 = const()[name = tensor<string, []>("op_402_axis_0"), val = tensor<int32, []>(1)];
339
+ tensor<fp32, [1, 256, 1]> var_402_0, tensor<fp32, [1, 256, 1]> var_402_1 = split(axis = var_402_axis_0, split_sizes = var_402_split_sizes_0, x = h_19)[name = tensor<string, []>("op_402")];
340
+ tensor<fp32, []> var_404_promoted = const()[name = tensor<string, []>("op_404_promoted"), val = tensor<fp32, []>(0x1p+0)];
341
+ tensor<fp32, [1, 256, 1]> var_405 = add(x = var_402_0, y = var_404_promoted)[name = tensor<string, []>("op_405")];
342
+ tensor<fp32, [1, 256, ?]> var_408 = instance_norm(beta = noise_res_0_adain1_0_norm_bias, epsilon = var_183, gamma = noise_res_0_adain1_0_norm_weight, x = input_31)[name = tensor<string, []>("op_408")];
343
+ tensor<fp32, [1, 256, ?]> var_409 = mul(x = var_405, y = var_408)[name = tensor<string, []>("op_409")];
344
+ tensor<fp32, [1, 256, ?]> xt_13 = add(x = var_409, y = var_402_1)[name = tensor<string, []>("xt_13")];
345
+ tensor<fp32, [1, 256, 1]> var_411 = const()[name = tensor<string, []>("op_411"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2566464)))];
346
+ tensor<fp32, [1, 256, ?]> var_414 = mul(x = noise_res_0_alpha1_2, y = xt_13)[name = tensor<string, []>("op_414")];
347
+ tensor<fp32, [1, 256, ?]> var_415 = sin(x = var_414)[name = tensor<string, []>("op_415")];
348
+ tensor<fp32, []> var_182_promoted_4 = const()[name = tensor<string, []>("op_182_promoted_4"), val = tensor<fp32, []>(0x1p+1)];
349
+ tensor<fp32, [1, 256, ?]> var_416 = pow(x = var_415, y = var_182_promoted_4)[name = tensor<string, []>("op_416")];
350
+ tensor<fp32, [1, 256, ?]> var_417 = mul(x = var_411, y = var_416)[name = tensor<string, []>("op_417")];
351
+ tensor<fp32, [1, 256, ?]> input_33 = add(x = xt_13, y = var_417)[name = tensor<string, []>("input_33")];
352
+ tensor<fp32, [256, 256, 7]> weight_25_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [458752]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(2567552))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3026368))), name = tensor<string, []>("weight_25_palettized"), shape = tensor<uint32, [3]>([256, 256, 7])];
353
+ tensor<string, []> input_35_pad_type_0 = const()[name = tensor<string, []>("input_35_pad_type_0"), val = tensor<string, []>("custom")];
354
+ tensor<int32, [2]> input_35_pad_0 = const()[name = tensor<string, []>("input_35_pad_0"), val = tensor<int32, [2]>([15, 15])];
355
+ tensor<int32, [1]> input_35_dilations_0 = const()[name = tensor<string, []>("input_35_dilations_0"), val = tensor<int32, [1]>([5])];
356
+ tensor<int32, [1]> input_35_strides_0 = const()[name = tensor<string, []>("input_35_strides_0"), val = tensor<int32, [1]>([1])];
357
+ tensor<int32, []> input_35_groups_0 = const()[name = tensor<string, []>("input_35_groups_0"), val = tensor<int32, []>(1)];
358
+ tensor<fp32, [1, 256, ?]> input_35 = conv(bias = noise_res_0_convs1_2_bias, dilations = input_35_dilations_0, groups = input_35_groups_0, pad = input_35_pad_0, pad_type = input_35_pad_type_0, strides = input_35_strides_0, weight = weight_25_palettized, x = input_33)[name = tensor<string, []>("input_35")];
359
+ tensor<fp32, [1, 512]> h_21 = linear(bias = noise_res_0_adain2_2_fc_bias, weight = noise_res_0_adain2_2_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_6")];
360
+ tensor<int32, [3]> var_433 = const()[name = tensor<string, []>("op_433"), val = tensor<int32, [3]>([1, 512, 1])];
361
+ tensor<fp32, [1, 512, 1]> h_23 = reshape(shape = var_433, x = h_21)[name = tensor<string, []>("h_23")];
362
+ tensor<int32, [2]> var_435_split_sizes_0 = const()[name = tensor<string, []>("op_435_split_sizes_0"), val = tensor<int32, [2]>([256, 256])];
363
+ tensor<int32, []> var_435_axis_0 = const()[name = tensor<string, []>("op_435_axis_0"), val = tensor<int32, []>(1)];
364
+ tensor<fp32, [1, 256, 1]> var_435_0, tensor<fp32, [1, 256, 1]> var_435_1 = split(axis = var_435_axis_0, split_sizes = var_435_split_sizes_0, x = h_23)[name = tensor<string, []>("op_435")];
365
+ tensor<fp32, []> var_437_promoted = const()[name = tensor<string, []>("op_437_promoted"), val = tensor<fp32, []>(0x1p+0)];
366
+ tensor<fp32, [1, 256, 1]> var_438 = add(x = var_435_0, y = var_437_promoted)[name = tensor<string, []>("op_438")];
367
+ tensor<fp32, [1, 256, ?]> var_441 = instance_norm(beta = noise_res_0_adain1_0_norm_bias, epsilon = var_183, gamma = noise_res_0_adain1_0_norm_weight, x = input_35)[name = tensor<string, []>("op_441")];
368
+ tensor<fp32, [1, 256, ?]> var_442 = mul(x = var_438, y = var_441)[name = tensor<string, []>("op_442")];
369
+ tensor<fp32, [1, 256, ?]> xt_15 = add(x = var_442, y = var_435_1)[name = tensor<string, []>("xt_15")];
370
+ tensor<fp32, [1, 256, 1]> var_444 = const()[name = tensor<string, []>("op_444"), val = tensor<fp32, [1, 256, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3027456)))];
371
+ tensor<fp32, [1, 256, ?]> var_447 = mul(x = noise_res_0_alpha2_2, y = xt_15)[name = tensor<string, []>("op_447")];
372
+ tensor<fp32, [1, 256, ?]> var_448 = sin(x = var_447)[name = tensor<string, []>("op_448")];
373
+ tensor<fp32, []> var_182_promoted_5 = const()[name = tensor<string, []>("op_182_promoted_5"), val = tensor<fp32, []>(0x1p+1)];
374
+ tensor<fp32, [1, 256, ?]> var_449 = pow(x = var_448, y = var_182_promoted_5)[name = tensor<string, []>("op_449")];
375
+ tensor<fp32, [1, 256, ?]> var_450 = mul(x = var_444, y = var_449)[name = tensor<string, []>("op_450")];
376
+ tensor<fp32, [1, 256, ?]> input_37 = add(x = xt_15, y = var_450)[name = tensor<string, []>("input_37")];
377
+ tensor<fp32, [256, 256, 7]> weight_29_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [458752]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3028544))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3487360))), name = tensor<string, []>("weight_29_palettized"), shape = tensor<uint32, [3]>([256, 256, 7])];
378
+ tensor<string, []> xt_17_pad_type_0 = const()[name = tensor<string, []>("xt_17_pad_type_0"), val = tensor<string, []>("custom")];
379
+ tensor<int32, [2]> xt_17_pad_0 = const()[name = tensor<string, []>("xt_17_pad_0"), val = tensor<int32, [2]>([3, 3])];
380
+ tensor<int32, [1]> xt_17_strides_0 = const()[name = tensor<string, []>("xt_17_strides_0"), val = tensor<int32, [1]>([1])];
381
+ tensor<int32, [1]> xt_17_dilations_0 = const()[name = tensor<string, []>("xt_17_dilations_0"), val = tensor<int32, [1]>([1])];
382
+ tensor<int32, []> xt_17_groups_0 = const()[name = tensor<string, []>("xt_17_groups_0"), val = tensor<int32, []>(1)];
383
+ tensor<fp32, [1, 256, ?]> xt_17 = conv(bias = noise_res_0_convs2_2_bias, dilations = xt_17_dilations_0, groups = xt_17_groups_0, pad = xt_17_pad_0, pad_type = xt_17_pad_type_0, strides = xt_17_strides_0, weight = weight_29_palettized, x = input_37)[name = tensor<string, []>("xt_17")];
384
+ tensor<fp32, [1, 256, ?]> x_source_0 = add(x = xt_17, y = input_31)[name = tensor<string, []>("op_459")];
385
+ tensor<string, []> input_39_pad_type_0 = const()[name = tensor<string, []>("input_39_pad_type_0"), val = tensor<string, []>("valid")];
386
+ tensor<int32, [1]> input_39_strides_0 = const()[name = tensor<string, []>("input_39_strides_0"), val = tensor<int32, [1]>([1])];
387
+ tensor<int32, [2]> input_39_pad_0 = const()[name = tensor<string, []>("input_39_pad_0"), val = tensor<int32, [2]>([0, 0])];
388
+ tensor<int32, [1]> input_39_dilations_0 = const()[name = tensor<string, []>("input_39_dilations_0"), val = tensor<int32, [1]>([1])];
389
+ tensor<int32, []> input_39_groups_0 = const()[name = tensor<string, []>("input_39_groups_0"), val = tensor<int32, []>(1)];
390
+ tensor<fp32, [1, 128, ?]> input_39 = conv(bias = noise_convs_1_bias, dilations = input_39_dilations_0, groups = input_39_groups_0, pad = input_39_pad_0, pad_type = input_39_pad_type_0, strides = input_39_strides_0, weight = noise_convs_1_weight_palettized, x = input_13)[name = tensor<string, []>("input_39")];
391
+ tensor<fp32, []> var_479 = const()[name = tensor<string, []>("op_479"), val = tensor<fp32, []>(0x1.4f8b58p-17)];
392
+ tensor<fp32, [1, 256]> h_25 = linear(bias = noise_res_1_adain1_0_fc_bias, weight = noise_res_1_adain1_0_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_7")];
393
+ tensor<int32, [3]> var_562 = const()[name = tensor<string, []>("op_562"), val = tensor<int32, [3]>([1, 256, 1])];
394
+ tensor<fp32, [1, 256, 1]> h_27 = reshape(shape = var_562, x = h_25)[name = tensor<string, []>("h_27")];
395
+ tensor<int32, [2]> var_564_split_sizes_0 = const()[name = tensor<string, []>("op_564_split_sizes_0"), val = tensor<int32, [2]>([128, 128])];
396
+ tensor<int32, []> var_564_axis_0 = const()[name = tensor<string, []>("op_564_axis_0"), val = tensor<int32, []>(1)];
397
+ tensor<fp32, [1, 128, 1]> var_564_0, tensor<fp32, [1, 128, 1]> var_564_1 = split(axis = var_564_axis_0, split_sizes = var_564_split_sizes_0, x = h_27)[name = tensor<string, []>("op_564")];
398
+ tensor<fp32, []> var_566_promoted = const()[name = tensor<string, []>("op_566_promoted"), val = tensor<fp32, []>(0x1p+0)];
399
+ tensor<fp32, [1, 128, 1]> var_567 = add(x = var_564_0, y = var_566_promoted)[name = tensor<string, []>("op_567")];
400
+ tensor<fp32, [1, 128, ?]> var_570 = instance_norm(beta = noise_res_1_adain1_0_norm_bias, epsilon = var_479, gamma = noise_res_1_adain1_0_norm_weight, x = input_39)[name = tensor<string, []>("op_570")];
401
+ tensor<fp32, [1, 128, ?]> var_571 = mul(x = var_567, y = var_570)[name = tensor<string, []>("op_571")];
402
+ tensor<fp32, [1, 128, ?]> xt_19 = add(x = var_571, y = var_564_1)[name = tensor<string, []>("xt_19")];
403
+ tensor<fp32, [1, 128, 1]> var_573 = const()[name = tensor<string, []>("op_573"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3488448)))];
404
+ tensor<fp32, [1, 128, ?]> var_576 = mul(x = noise_res_1_alpha1_0, y = xt_19)[name = tensor<string, []>("op_576")];
405
+ tensor<fp32, [1, 128, ?]> var_577 = sin(x = var_576)[name = tensor<string, []>("op_577")];
406
+ tensor<fp32, []> var_478_promoted = const()[name = tensor<string, []>("op_478_promoted"), val = tensor<fp32, []>(0x1p+1)];
407
+ tensor<fp32, [1, 128, ?]> var_578 = pow(x = var_577, y = var_478_promoted)[name = tensor<string, []>("op_578")];
408
+ tensor<fp32, [1, 128, ?]> var_579 = mul(x = var_573, y = var_578)[name = tensor<string, []>("op_579")];
409
+ tensor<fp32, [1, 128, ?]> input_41 = add(x = xt_19, y = var_579)[name = tensor<string, []>("input_41")];
410
+ tensor<fp32, [128, 128, 11]> weight_35_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [180224]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3489024))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3669312))), name = tensor<string, []>("weight_35_palettized"), shape = tensor<uint32, [3]>([128, 128, 11])];
411
+ tensor<string, []> input_43_pad_type_0 = const()[name = tensor<string, []>("input_43_pad_type_0"), val = tensor<string, []>("custom")];
412
+ tensor<int32, [2]> input_43_pad_0 = const()[name = tensor<string, []>("input_43_pad_0"), val = tensor<int32, [2]>([5, 5])];
413
+ tensor<int32, [1]> input_43_strides_0 = const()[name = tensor<string, []>("input_43_strides_0"), val = tensor<int32, [1]>([1])];
414
+ tensor<int32, [1]> input_43_dilations_0 = const()[name = tensor<string, []>("input_43_dilations_0"), val = tensor<int32, [1]>([1])];
415
+ tensor<int32, []> input_43_groups_0 = const()[name = tensor<string, []>("input_43_groups_0"), val = tensor<int32, []>(1)];
416
+ tensor<fp32, [1, 128, ?]> input_43 = conv(bias = noise_res_1_convs1_0_bias, dilations = input_43_dilations_0, groups = input_43_groups_0, pad = input_43_pad_0, pad_type = input_43_pad_type_0, strides = input_43_strides_0, weight = weight_35_palettized, x = input_41)[name = tensor<string, []>("input_43")];
417
+ tensor<fp32, [1, 256]> h_29 = linear(bias = noise_res_1_adain2_0_fc_bias, weight = noise_res_1_adain2_0_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_8")];
418
+ tensor<int32, [3]> var_595 = const()[name = tensor<string, []>("op_595"), val = tensor<int32, [3]>([1, 256, 1])];
419
+ tensor<fp32, [1, 256, 1]> h_31 = reshape(shape = var_595, x = h_29)[name = tensor<string, []>("h_31")];
420
+ tensor<int32, [2]> var_597_split_sizes_0 = const()[name = tensor<string, []>("op_597_split_sizes_0"), val = tensor<int32, [2]>([128, 128])];
421
+ tensor<int32, []> var_597_axis_0 = const()[name = tensor<string, []>("op_597_axis_0"), val = tensor<int32, []>(1)];
422
+ tensor<fp32, [1, 128, 1]> var_597_0, tensor<fp32, [1, 128, 1]> var_597_1 = split(axis = var_597_axis_0, split_sizes = var_597_split_sizes_0, x = h_31)[name = tensor<string, []>("op_597")];
423
+ tensor<fp32, []> var_599_promoted = const()[name = tensor<string, []>("op_599_promoted"), val = tensor<fp32, []>(0x1p+0)];
424
+ tensor<fp32, [1, 128, 1]> var_600 = add(x = var_597_0, y = var_599_promoted)[name = tensor<string, []>("op_600")];
425
+ tensor<fp32, [1, 128, ?]> var_603 = instance_norm(beta = noise_res_1_adain1_0_norm_bias, epsilon = var_479, gamma = noise_res_1_adain1_0_norm_weight, x = input_43)[name = tensor<string, []>("op_603")];
426
+ tensor<fp32, [1, 128, ?]> var_604 = mul(x = var_600, y = var_603)[name = tensor<string, []>("op_604")];
427
+ tensor<fp32, [1, 128, ?]> xt_21 = add(x = var_604, y = var_597_1)[name = tensor<string, []>("xt_21")];
428
+ tensor<fp32, [1, 128, 1]> var_606 = const()[name = tensor<string, []>("op_606"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3670400)))];
429
+ tensor<fp32, [1, 128, ?]> var_609 = mul(x = noise_res_1_alpha2_0, y = xt_21)[name = tensor<string, []>("op_609")];
430
+ tensor<fp32, [1, 128, ?]> var_610 = sin(x = var_609)[name = tensor<string, []>("op_610")];
431
+ tensor<fp32, []> var_478_promoted_1 = const()[name = tensor<string, []>("op_478_promoted_1"), val = tensor<fp32, []>(0x1p+1)];
432
+ tensor<fp32, [1, 128, ?]> var_611 = pow(x = var_610, y = var_478_promoted_1)[name = tensor<string, []>("op_611")];
433
+ tensor<fp32, [1, 128, ?]> var_612 = mul(x = var_606, y = var_611)[name = tensor<string, []>("op_612")];
434
+ tensor<fp32, [1, 128, ?]> input_45 = add(x = xt_21, y = var_612)[name = tensor<string, []>("input_45")];
435
+ tensor<fp32, [128, 128, 11]> weight_39_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [180224]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3670976))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3851264))), name = tensor<string, []>("weight_39_palettized"), shape = tensor<uint32, [3]>([128, 128, 11])];
436
+ tensor<string, []> xt_23_pad_type_0 = const()[name = tensor<string, []>("xt_23_pad_type_0"), val = tensor<string, []>("custom")];
437
+ tensor<int32, [2]> xt_23_pad_0 = const()[name = tensor<string, []>("xt_23_pad_0"), val = tensor<int32, [2]>([5, 5])];
438
+ tensor<int32, [1]> xt_23_strides_0 = const()[name = tensor<string, []>("xt_23_strides_0"), val = tensor<int32, [1]>([1])];
439
+ tensor<int32, [1]> xt_23_dilations_0 = const()[name = tensor<string, []>("xt_23_dilations_0"), val = tensor<int32, [1]>([1])];
440
+ tensor<int32, []> xt_23_groups_0 = const()[name = tensor<string, []>("xt_23_groups_0"), val = tensor<int32, []>(1)];
441
+ tensor<fp32, [1, 128, ?]> xt_23 = conv(bias = noise_res_1_convs2_0_bias, dilations = xt_23_dilations_0, groups = xt_23_groups_0, pad = xt_23_pad_0, pad_type = xt_23_pad_type_0, strides = xt_23_strides_0, weight = weight_39_palettized, x = input_45)[name = tensor<string, []>("xt_23")];
442
+ tensor<fp32, [1, 128, ?]> input_47 = add(x = xt_23, y = input_39)[name = tensor<string, []>("input_47")];
443
+ tensor<fp32, [1, 256]> h_33 = linear(bias = noise_res_1_adain1_1_fc_bias, weight = noise_res_1_adain1_1_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_9")];
444
+ tensor<int32, [3]> var_629 = const()[name = tensor<string, []>("op_629"), val = tensor<int32, [3]>([1, 256, 1])];
445
+ tensor<fp32, [1, 256, 1]> h_35 = reshape(shape = var_629, x = h_33)[name = tensor<string, []>("h_35")];
446
+ tensor<int32, [2]> var_631_split_sizes_0 = const()[name = tensor<string, []>("op_631_split_sizes_0"), val = tensor<int32, [2]>([128, 128])];
447
+ tensor<int32, []> var_631_axis_0 = const()[name = tensor<string, []>("op_631_axis_0"), val = tensor<int32, []>(1)];
448
+ tensor<fp32, [1, 128, 1]> var_631_0, tensor<fp32, [1, 128, 1]> var_631_1 = split(axis = var_631_axis_0, split_sizes = var_631_split_sizes_0, x = h_35)[name = tensor<string, []>("op_631")];
449
+ tensor<fp32, []> var_633_promoted = const()[name = tensor<string, []>("op_633_promoted"), val = tensor<fp32, []>(0x1p+0)];
450
+ tensor<fp32, [1, 128, 1]> var_634 = add(x = var_631_0, y = var_633_promoted)[name = tensor<string, []>("op_634")];
451
+ tensor<fp32, [1, 128, ?]> var_637 = instance_norm(beta = noise_res_1_adain1_0_norm_bias, epsilon = var_479, gamma = noise_res_1_adain1_0_norm_weight, x = input_47)[name = tensor<string, []>("op_637")];
452
+ tensor<fp32, [1, 128, ?]> var_638 = mul(x = var_634, y = var_637)[name = tensor<string, []>("op_638")];
453
+ tensor<fp32, [1, 128, ?]> xt_25 = add(x = var_638, y = var_631_1)[name = tensor<string, []>("xt_25")];
454
+ tensor<fp32, [1, 128, 1]> var_640 = const()[name = tensor<string, []>("op_640"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3852352)))];
455
+ tensor<fp32, [1, 128, ?]> var_643 = mul(x = noise_res_1_alpha1_1, y = xt_25)[name = tensor<string, []>("op_643")];
456
+ tensor<fp32, [1, 128, ?]> var_644 = sin(x = var_643)[name = tensor<string, []>("op_644")];
457
+ tensor<fp32, []> var_478_promoted_2 = const()[name = tensor<string, []>("op_478_promoted_2"), val = tensor<fp32, []>(0x1p+1)];
458
+ tensor<fp32, [1, 128, ?]> var_645 = pow(x = var_644, y = var_478_promoted_2)[name = tensor<string, []>("op_645")];
459
+ tensor<fp32, [1, 128, ?]> var_646 = mul(x = var_640, y = var_645)[name = tensor<string, []>("op_646")];
460
+ tensor<fp32, [1, 128, ?]> input_49 = add(x = xt_25, y = var_646)[name = tensor<string, []>("input_49")];
461
+ tensor<fp32, [128, 128, 11]> weight_43_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [180224]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3852928))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4033216))), name = tensor<string, []>("weight_43_palettized"), shape = tensor<uint32, [3]>([128, 128, 11])];
462
+ tensor<string, []> input_51_pad_type_0 = const()[name = tensor<string, []>("input_51_pad_type_0"), val = tensor<string, []>("custom")];
463
+ tensor<int32, [2]> input_51_pad_0 = const()[name = tensor<string, []>("input_51_pad_0"), val = tensor<int32, [2]>([15, 15])];
464
+ tensor<int32, [1]> input_51_dilations_0 = const()[name = tensor<string, []>("input_51_dilations_0"), val = tensor<int32, [1]>([3])];
465
+ tensor<int32, [1]> input_51_strides_0 = const()[name = tensor<string, []>("input_51_strides_0"), val = tensor<int32, [1]>([1])];
466
+ tensor<int32, []> input_51_groups_0 = const()[name = tensor<string, []>("input_51_groups_0"), val = tensor<int32, []>(1)];
467
+ tensor<fp32, [1, 128, ?]> input_51 = conv(bias = noise_res_1_convs1_1_bias, dilations = input_51_dilations_0, groups = input_51_groups_0, pad = input_51_pad_0, pad_type = input_51_pad_type_0, strides = input_51_strides_0, weight = weight_43_palettized, x = input_49)[name = tensor<string, []>("input_51")];
468
+ tensor<fp32, [1, 256]> h_37 = linear(bias = noise_res_1_adain2_1_fc_bias, weight = noise_res_1_adain2_1_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_10")];
469
+ tensor<int32, [3]> var_662 = const()[name = tensor<string, []>("op_662"), val = tensor<int32, [3]>([1, 256, 1])];
470
+ tensor<fp32, [1, 256, 1]> h_39 = reshape(shape = var_662, x = h_37)[name = tensor<string, []>("h_39")];
471
+ tensor<int32, [2]> var_664_split_sizes_0 = const()[name = tensor<string, []>("op_664_split_sizes_0"), val = tensor<int32, [2]>([128, 128])];
472
+ tensor<int32, []> var_664_axis_0 = const()[name = tensor<string, []>("op_664_axis_0"), val = tensor<int32, []>(1)];
473
+ tensor<fp32, [1, 128, 1]> var_664_0, tensor<fp32, [1, 128, 1]> var_664_1 = split(axis = var_664_axis_0, split_sizes = var_664_split_sizes_0, x = h_39)[name = tensor<string, []>("op_664")];
474
+ tensor<fp32, []> var_666_promoted = const()[name = tensor<string, []>("op_666_promoted"), val = tensor<fp32, []>(0x1p+0)];
475
+ tensor<fp32, [1, 128, 1]> var_667 = add(x = var_664_0, y = var_666_promoted)[name = tensor<string, []>("op_667")];
476
+ tensor<fp32, [1, 128, ?]> var_670 = instance_norm(beta = noise_res_1_adain1_0_norm_bias, epsilon = var_479, gamma = noise_res_1_adain1_0_norm_weight, x = input_51)[name = tensor<string, []>("op_670")];
477
+ tensor<fp32, [1, 128, ?]> var_671 = mul(x = var_667, y = var_670)[name = tensor<string, []>("op_671")];
478
+ tensor<fp32, [1, 128, ?]> xt_27 = add(x = var_671, y = var_664_1)[name = tensor<string, []>("xt_27")];
479
+ tensor<fp32, [1, 128, 1]> var_673 = const()[name = tensor<string, []>("op_673"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4034304)))];
480
+ tensor<fp32, [1, 128, ?]> var_676 = mul(x = noise_res_1_alpha2_1, y = xt_27)[name = tensor<string, []>("op_676")];
481
+ tensor<fp32, [1, 128, ?]> var_677 = sin(x = var_676)[name = tensor<string, []>("op_677")];
482
+ tensor<fp32, []> var_478_promoted_3 = const()[name = tensor<string, []>("op_478_promoted_3"), val = tensor<fp32, []>(0x1p+1)];
483
+ tensor<fp32, [1, 128, ?]> var_678 = pow(x = var_677, y = var_478_promoted_3)[name = tensor<string, []>("op_678")];
484
+ tensor<fp32, [1, 128, ?]> var_679 = mul(x = var_673, y = var_678)[name = tensor<string, []>("op_679")];
485
+ tensor<fp32, [1, 128, ?]> input_53 = add(x = xt_27, y = var_679)[name = tensor<string, []>("input_53")];
486
+ tensor<fp32, [128, 128, 11]> weight_47_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [180224]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4034880))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4215168))), name = tensor<string, []>("weight_47_palettized"), shape = tensor<uint32, [3]>([128, 128, 11])];
487
+ tensor<string, []> xt_29_pad_type_0 = const()[name = tensor<string, []>("xt_29_pad_type_0"), val = tensor<string, []>("custom")];
488
+ tensor<int32, [2]> xt_29_pad_0 = const()[name = tensor<string, []>("xt_29_pad_0"), val = tensor<int32, [2]>([5, 5])];
489
+ tensor<int32, [1]> xt_29_strides_0 = const()[name = tensor<string, []>("xt_29_strides_0"), val = tensor<int32, [1]>([1])];
490
+ tensor<int32, [1]> xt_29_dilations_0 = const()[name = tensor<string, []>("xt_29_dilations_0"), val = tensor<int32, [1]>([1])];
491
+ tensor<int32, []> xt_29_groups_0 = const()[name = tensor<string, []>("xt_29_groups_0"), val = tensor<int32, []>(1)];
492
+ tensor<fp32, [1, 128, ?]> xt_29 = conv(bias = noise_res_1_convs2_1_bias, dilations = xt_29_dilations_0, groups = xt_29_groups_0, pad = xt_29_pad_0, pad_type = xt_29_pad_type_0, strides = xt_29_strides_0, weight = weight_47_palettized, x = input_53)[name = tensor<string, []>("xt_29")];
493
+ tensor<fp32, [1, 128, ?]> input_55 = add(x = xt_29, y = input_47)[name = tensor<string, []>("input_55")];
494
+ tensor<fp32, [1, 256]> h_41 = linear(bias = noise_res_1_adain1_2_fc_bias, weight = noise_res_1_adain1_2_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_11")];
495
+ tensor<int32, [3]> var_696 = const()[name = tensor<string, []>("op_696"), val = tensor<int32, [3]>([1, 256, 1])];
496
+ tensor<fp32, [1, 256, 1]> h_43 = reshape(shape = var_696, x = h_41)[name = tensor<string, []>("h_43")];
497
+ tensor<int32, [2]> var_698_split_sizes_0 = const()[name = tensor<string, []>("op_698_split_sizes_0"), val = tensor<int32, [2]>([128, 128])];
498
+ tensor<int32, []> var_698_axis_0 = const()[name = tensor<string, []>("op_698_axis_0"), val = tensor<int32, []>(1)];
499
+ tensor<fp32, [1, 128, 1]> var_698_0, tensor<fp32, [1, 128, 1]> var_698_1 = split(axis = var_698_axis_0, split_sizes = var_698_split_sizes_0, x = h_43)[name = tensor<string, []>("op_698")];
500
+ tensor<fp32, []> var_700_promoted = const()[name = tensor<string, []>("op_700_promoted"), val = tensor<fp32, []>(0x1p+0)];
501
+ tensor<fp32, [1, 128, 1]> var_701 = add(x = var_698_0, y = var_700_promoted)[name = tensor<string, []>("op_701")];
502
+ tensor<fp32, [1, 128, ?]> var_704 = instance_norm(beta = noise_res_1_adain1_0_norm_bias, epsilon = var_479, gamma = noise_res_1_adain1_0_norm_weight, x = input_55)[name = tensor<string, []>("op_704")];
503
+ tensor<fp32, [1, 128, ?]> var_705 = mul(x = var_701, y = var_704)[name = tensor<string, []>("op_705")];
504
+ tensor<fp32, [1, 128, ?]> xt_31 = add(x = var_705, y = var_698_1)[name = tensor<string, []>("xt_31")];
505
+ tensor<fp32, [1, 128, 1]> var_707 = const()[name = tensor<string, []>("op_707"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4216256)))];
506
+ tensor<fp32, [1, 128, ?]> var_710 = mul(x = noise_res_1_alpha1_2, y = xt_31)[name = tensor<string, []>("op_710")];
507
+ tensor<fp32, [1, 128, ?]> var_711 = sin(x = var_710)[name = tensor<string, []>("op_711")];
508
+ tensor<fp32, []> var_478_promoted_4 = const()[name = tensor<string, []>("op_478_promoted_4"), val = tensor<fp32, []>(0x1p+1)];
509
+ tensor<fp32, [1, 128, ?]> var_712 = pow(x = var_711, y = var_478_promoted_4)[name = tensor<string, []>("op_712")];
510
+ tensor<fp32, [1, 128, ?]> var_713 = mul(x = var_707, y = var_712)[name = tensor<string, []>("op_713")];
511
+ tensor<fp32, [1, 128, ?]> input_57 = add(x = xt_31, y = var_713)[name = tensor<string, []>("input_57")];
512
+ tensor<fp32, [128, 128, 11]> weight_51_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [180224]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4216832))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4397120))), name = tensor<string, []>("weight_51_palettized"), shape = tensor<uint32, [3]>([128, 128, 11])];
513
+ tensor<string, []> input_59_pad_type_0 = const()[name = tensor<string, []>("input_59_pad_type_0"), val = tensor<string, []>("custom")];
514
+ tensor<int32, [2]> input_59_pad_0 = const()[name = tensor<string, []>("input_59_pad_0"), val = tensor<int32, [2]>([25, 25])];
515
+ tensor<int32, [1]> input_59_dilations_0 = const()[name = tensor<string, []>("input_59_dilations_0"), val = tensor<int32, [1]>([5])];
516
+ tensor<int32, [1]> input_59_strides_0 = const()[name = tensor<string, []>("input_59_strides_0"), val = tensor<int32, [1]>([1])];
517
+ tensor<int32, []> input_59_groups_0 = const()[name = tensor<string, []>("input_59_groups_0"), val = tensor<int32, []>(1)];
518
+ tensor<fp32, [1, 128, ?]> input_59 = conv(bias = noise_res_1_convs1_2_bias, dilations = input_59_dilations_0, groups = input_59_groups_0, pad = input_59_pad_0, pad_type = input_59_pad_type_0, strides = input_59_strides_0, weight = weight_51_palettized, x = input_57)[name = tensor<string, []>("input_59")];
519
+ tensor<fp32, [1, 256]> h_45 = linear(bias = noise_res_1_adain2_2_fc_bias, weight = noise_res_1_adain2_2_fc_weight_palettized, x = style_timbre)[name = tensor<string, []>("linear_12")];
520
+ tensor<int32, [3]> var_729 = const()[name = tensor<string, []>("op_729"), val = tensor<int32, [3]>([1, 256, 1])];
521
+ tensor<fp32, [1, 256, 1]> h = reshape(shape = var_729, x = h_45)[name = tensor<string, []>("h")];
522
+ tensor<int32, [2]> var_731_split_sizes_0 = const()[name = tensor<string, []>("op_731_split_sizes_0"), val = tensor<int32, [2]>([128, 128])];
523
+ tensor<int32, []> var_731_axis_0 = const()[name = tensor<string, []>("op_731_axis_0"), val = tensor<int32, []>(1)];
524
+ tensor<fp32, [1, 128, 1]> var_731_0, tensor<fp32, [1, 128, 1]> var_731_1 = split(axis = var_731_axis_0, split_sizes = var_731_split_sizes_0, x = h)[name = tensor<string, []>("op_731")];
525
+ tensor<fp32, []> var_733_promoted = const()[name = tensor<string, []>("op_733_promoted"), val = tensor<fp32, []>(0x1p+0)];
526
+ tensor<fp32, [1, 128, 1]> var_734 = add(x = var_731_0, y = var_733_promoted)[name = tensor<string, []>("op_734")];
527
+ tensor<fp32, [1, 128, ?]> var_737 = instance_norm(beta = noise_res_1_adain1_0_norm_bias, epsilon = var_479, gamma = noise_res_1_adain1_0_norm_weight, x = input_59)[name = tensor<string, []>("op_737")];
528
+ tensor<fp32, [1, 128, ?]> var_738 = mul(x = var_734, y = var_737)[name = tensor<string, []>("op_738")];
529
+ tensor<fp32, [1, 128, ?]> xt_33 = add(x = var_738, y = var_731_1)[name = tensor<string, []>("xt_33")];
530
+ tensor<fp32, [1, 128, 1]> var_740 = const()[name = tensor<string, []>("op_740"), val = tensor<fp32, [1, 128, 1]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4398208)))];
531
+ tensor<fp32, [1, 128, ?]> var_743 = mul(x = noise_res_1_alpha2_2, y = xt_33)[name = tensor<string, []>("op_743")];
532
+ tensor<fp32, [1, 128, ?]> var_744 = sin(x = var_743)[name = tensor<string, []>("op_744")];
533
+ tensor<fp32, []> var_478_promoted_5 = const()[name = tensor<string, []>("op_478_promoted_5"), val = tensor<fp32, []>(0x1p+1)];
534
+ tensor<fp32, [1, 128, ?]> var_745 = pow(x = var_744, y = var_478_promoted_5)[name = tensor<string, []>("op_745")];
535
+ tensor<fp32, [1, 128, ?]> var_746 = mul(x = var_740, y = var_745)[name = tensor<string, []>("op_746")];
536
+ tensor<fp32, [1, 128, ?]> input = add(x = xt_33, y = var_746)[name = tensor<string, []>("input")];
537
+ tensor<fp32, [128, 128, 11]> weight_55_palettized = constexpr_lut_to_dense()[indices = tensor<uint8, [180224]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4398784))), lut = tensor<fp32, [256]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(4579072))), name = tensor<string, []>("weight_55_palettized"), shape = tensor<uint32, [3]>([128, 128, 11])];
538
+ tensor<string, []> xt_pad_type_0 = const()[name = tensor<string, []>("xt_pad_type_0"), val = tensor<string, []>("custom")];
539
+ tensor<int32, [2]> xt_pad_0 = const()[name = tensor<string, []>("xt_pad_0"), val = tensor<int32, [2]>([5, 5])];
540
+ tensor<int32, [1]> xt_strides_0 = const()[name = tensor<string, []>("xt_strides_0"), val = tensor<int32, [1]>([1])];
541
+ tensor<int32, [1]> xt_dilations_0 = const()[name = tensor<string, []>("xt_dilations_0"), val = tensor<int32, [1]>([1])];
542
+ tensor<int32, []> xt_groups_0 = const()[name = tensor<string, []>("xt_groups_0"), val = tensor<int32, []>(1)];
543
+ tensor<fp32, [1, 128, ?]> xt = conv(bias = noise_res_1_convs2_2_bias, dilations = xt_dilations_0, groups = xt_groups_0, pad = xt_pad_0, pad_type = xt_pad_type_0, strides = xt_strides_0, weight = weight_55_palettized, x = input)[name = tensor<string, []>("xt")];
544
+ tensor<fp32, [1, 128, ?]> x_source_1 = add(x = xt, y = input_55)[name = tensor<string, []>("op_755")];
545
+ } -> (x_source_0, x_source_1);
546
+ }
ANE/KokoroNoise_v2.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1102fc2d31dfcfe3de3978a4c78b65202ff8b0a4d55a9304213bd4e8bda66bc2
3
+ size 4580160
ANE/KokoroNoise_v2.mlpackage/Data/com.apple.CoreML/model.mlmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:894168c950938e0b776f010b4278b1aae65aa1ce3dd8bdfa4c4170ca80362d0d
3
+ size 76250
ANE/KokoroNoise_v2.mlpackage/Data/com.apple.CoreML/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1102fc2d31dfcfe3de3978a4c78b65202ff8b0a4d55a9304213bd4e8bda66bc2
3
+ size 4580160
ANE/KokoroNoise_v2.mlpackage/Manifest.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "fileFormatVersion": "1.0.0",
3
+ "itemInfoEntries": {
4
+ "03508E9C-C90A-4EE6-86F1-4481C0530C10": {
5
+ "author": "com.apple.CoreML",
6
+ "description": "CoreML Model Specification",
7
+ "name": "model.mlmodel",
8
+ "path": "com.apple.CoreML/model.mlmodel"
9
+ },
10
+ "237EFEC1-9F1E-45FF-8053-0573D8EBD28A": {
11
+ "author": "com.apple.CoreML",
12
+ "description": "CoreML Model Weights",
13
+ "name": "weights",
14
+ "path": "com.apple.CoreML/weights"
15
+ }
16
+ },
17
+ "rootModelIdentifier": "03508E9C-C90A-4EE6-86F1-4481C0530C10"
18
+ }