connor-henderson
commited on
Commit
·
ef328ad
1
Parent(s):
ba406a3
Upload model
Browse files- config.json +23 -17
- pytorch_model.bin +2 -2
config.json
CHANGED
@@ -2,28 +2,34 @@
|
|
2 |
"architectures": [
|
3 |
"FastSpeech2ConformerModel"
|
4 |
],
|
5 |
-
"
|
6 |
-
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
|
|
|
|
|
14 |
"duration_predictor_channels": 256,
|
15 |
"duration_predictor_dropout_rate": 0.2,
|
16 |
"duration_predictor_kernel_size": 3,
|
17 |
"duration_predictor_layers": 2,
|
18 |
-
"
|
19 |
-
|
20 |
-
|
21 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
"encoder_layers": 4,
|
23 |
-
"encoder_linear_units": 1536,
|
24 |
-
"encoder_normalize_before": false,
|
25 |
"encoder_num_attention_heads": 2,
|
26 |
-
"encoder_positional_dropout_rate": 0.2,
|
27 |
"energy_embed_dropout": 0.0,
|
28 |
"energy_embed_kernel_size": 1,
|
29 |
"energy_predictor_channels": 256,
|
@@ -55,7 +61,7 @@
|
|
55 |
"stop_gradient_from_energy_predictor": false,
|
56 |
"stop_gradient_from_pitch_predictor": true,
|
57 |
"torch_dtype": "float32",
|
58 |
-
"transformers_version": "4.
|
59 |
"use_cnn_in_conformer": true,
|
60 |
"use_macaron_style_in_conformer": true,
|
61 |
"use_masking": true,
|
|
|
2 |
"architectures": [
|
3 |
"FastSpeech2ConformerModel"
|
4 |
],
|
5 |
+
"decoder_config": {
|
6 |
+
"attention_dropout_rate": 0.2,
|
7 |
+
"concat_after": false,
|
8 |
+
"dropout_rate": 0.2,
|
9 |
+
"kernel_size": 31,
|
10 |
+
"layers": 4,
|
11 |
+
"linear_units": 1536,
|
12 |
+
"normalize_before": false,
|
13 |
+
"num_attention_heads": 2,
|
14 |
+
"positional_dropout_rate": 0.2
|
15 |
+
},
|
16 |
"duration_predictor_channels": 256,
|
17 |
"duration_predictor_dropout_rate": 0.2,
|
18 |
"duration_predictor_kernel_size": 3,
|
19 |
"duration_predictor_layers": 2,
|
20 |
+
"encoder_config": {
|
21 |
+
"attention_dropout_rate": 0.2,
|
22 |
+
"concat_after": false,
|
23 |
+
"dropout_rate": 0.2,
|
24 |
+
"kernel_size": 7,
|
25 |
+
"layers": 4,
|
26 |
+
"linear_units": 1536,
|
27 |
+
"normalize_before": false,
|
28 |
+
"num_attention_heads": 2,
|
29 |
+
"positional_dropout_rate": 0.2
|
30 |
+
},
|
31 |
"encoder_layers": 4,
|
|
|
|
|
32 |
"encoder_num_attention_heads": 2,
|
|
|
33 |
"energy_embed_dropout": 0.0,
|
34 |
"energy_embed_kernel_size": 1,
|
35 |
"energy_predictor_channels": 256,
|
|
|
61 |
"stop_gradient_from_energy_predictor": false,
|
62 |
"stop_gradient_from_pitch_predictor": true,
|
63 |
"torch_dtype": "float32",
|
64 |
+
"transformers_version": "4.33.0.dev0",
|
65 |
"use_cnn_in_conformer": true,
|
66 |
"use_macaron_style_in_conformer": true,
|
67 |
"use_masking": true,
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fa4096f1c4ce019198a7ec8054a7828916bc94b6c6cb968d7723d5a4b808b7a8
|
3 |
+
size 281215241
|