Upload 3 files
Browse files- .gitattributes +1 -0
- LibriTTS960_50Hz.json +63 -0
- LibriTTS960_50Hz.safetensors +3 -0
- focalcodec.png +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
focalcodec.png filter=lfs diff=lfs merge=lfs -text
|
LibriTTS960_50Hz.json
ADDED
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"encoder_name": "WavLM",
|
3 |
+
"encoder_config": {
|
4 |
+
"hidden_dims": [512, 512, 512, 512, 512, 512, 512],
|
5 |
+
"kernel_sizes": [10, 3, 3, 3, 3, 2, 2],
|
6 |
+
"strides": [5, 2, 2, 2, 2, 2, 2],
|
7 |
+
"num_layers": 6,
|
8 |
+
"dim": 1024,
|
9 |
+
"ffn_dim": 4096,
|
10 |
+
"num_heads": 16,
|
11 |
+
"num_buckets": 320,
|
12 |
+
"max_distance": 800,
|
13 |
+
"dropout": 0.0,
|
14 |
+
"conv_pos": 128,
|
15 |
+
"conv_pos_groups": 16
|
16 |
+
},
|
17 |
+
"compressor_name": "FocalEncoder",
|
18 |
+
"compressor_config": {
|
19 |
+
"input_dim": 1024,
|
20 |
+
"output_dim": 13,
|
21 |
+
"hidden_dims": [1024, 512, 256],
|
22 |
+
"downscale_factors": [1, 1, 1],
|
23 |
+
"focal_window": 7,
|
24 |
+
"focal_level": 2,
|
25 |
+
"focal_factor": 2,
|
26 |
+
"dropout": 0.0,
|
27 |
+
"use_post_norm": false,
|
28 |
+
"use_layerscale": false,
|
29 |
+
"layerscale_init": 0.0001,
|
30 |
+
"normalize_modulator": false
|
31 |
+
},
|
32 |
+
"quantizer_name": "BinarySphericalQuantizer",
|
33 |
+
"quantizer_config": {
|
34 |
+
"codebook_size": 8192
|
35 |
+
},
|
36 |
+
"decompressor_name": "FocalDecoder",
|
37 |
+
"decompressor_config": {
|
38 |
+
"input_dim": 13,
|
39 |
+
"output_dim": 1024,
|
40 |
+
"hidden_dims": [256, 512, 1024],
|
41 |
+
"upscale_factors": [1, 1, 1],
|
42 |
+
"focal_window": 7,
|
43 |
+
"focal_level": 2,
|
44 |
+
"focal_factor": 2,
|
45 |
+
"dropout": 0.0,
|
46 |
+
"use_post_norm": false,
|
47 |
+
"use_layerscale": false,
|
48 |
+
"layerscale_init": 0.0001,
|
49 |
+
"normalize_modulator": false
|
50 |
+
},
|
51 |
+
"decoder_name": "Vocos",
|
52 |
+
"decoder_config": {
|
53 |
+
"input_channels": 1024,
|
54 |
+
"num_layers": 8,
|
55 |
+
"dim": 512,
|
56 |
+
"ffn_dim": 1536,
|
57 |
+
"kernel_size": 7,
|
58 |
+
"padding": 3,
|
59 |
+
"layerscale_init": null,
|
60 |
+
"n_fft": 1024,
|
61 |
+
"hop_length": 320
|
62 |
+
}
|
63 |
+
}
|
LibriTTS960_50Hz.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2700a916ca8d1c11a899995ef8d451ee53a481486650c4b78cd96feff9ac77f0
|
3 |
+
size 568554604
|
focalcodec.png
ADDED
![]() |
Git LFS Details
|