Upload 9 files
Browse files- .gitattributes +1 -0
- Jelly Hoshiumi/D_2333333.pth +3 -0
- Jelly Hoshiumi/G_2333333.pth +3 -0
- Jelly Hoshiumi/Jelly Hoshiumi.index +3 -0
- Jelly Hoshiumi/Jelly Hoshiumi.pth +3 -0
- Jelly Hoshiumi/config.json +75 -0
- Jelly Hoshiumi/model_info.json +8 -0
- spin-v2/config.json +72 -0
- spin-v2/pytorch_model.bin +3 -0
- spin-v2/spin-v2.zip +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 |
+
Jelly[[:space:]]Hoshiumi/Jelly[[:space:]]Hoshiumi.index filter=lfs diff=lfs merge=lfs -text
|
Jelly Hoshiumi/D_2333333.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:227b519acf6923a808ab1b1f4c145604cbf901652277cd5bbdadf6e41a94f2e0
|
| 3 |
+
size 857123915
|
Jelly Hoshiumi/G_2333333.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f78a64379341b0aabae8c4ba22dbaa976708e233e3fadaf3cb5ea57f24e1c65
|
| 3 |
+
size 419538195
|
Jelly Hoshiumi/Jelly Hoshiumi.index
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f8d68fb5fad6c2420940e647ed0f088be8b6ac35d974d0858804968b6af51239
|
| 3 |
+
size 2319458819
|
Jelly Hoshiumi/Jelly Hoshiumi.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5016a97913748283fd14357061f9ee5cbd820f8a371e4a53fc806ca64055d33
|
| 3 |
+
size 52133177
|
Jelly Hoshiumi/config.json
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"train": {
|
| 3 |
+
"log_interval": 200,
|
| 4 |
+
"seed": 1234,
|
| 5 |
+
"learning_rate": 0.0001,
|
| 6 |
+
"betas": [
|
| 7 |
+
0.8,
|
| 8 |
+
0.99
|
| 9 |
+
],
|
| 10 |
+
"eps": 1e-09,
|
| 11 |
+
"lr_decay": 0.999875,
|
| 12 |
+
"segment_size": 17280,
|
| 13 |
+
"c_mel": 45,
|
| 14 |
+
"c_kl": 1.0
|
| 15 |
+
},
|
| 16 |
+
"data": {
|
| 17 |
+
"max_wav_value": 32768.0,
|
| 18 |
+
"sample_rate": 48000,
|
| 19 |
+
"filter_length": 2048,
|
| 20 |
+
"hop_length": 480,
|
| 21 |
+
"win_length": 2048,
|
| 22 |
+
"n_mel_channels": 128,
|
| 23 |
+
"mel_fmin": 0.0,
|
| 24 |
+
"mel_fmax": null
|
| 25 |
+
},
|
| 26 |
+
"model": {
|
| 27 |
+
"inter_channels": 192,
|
| 28 |
+
"hidden_channels": 192,
|
| 29 |
+
"filter_channels": 768,
|
| 30 |
+
"text_enc_hidden_dim": 768,
|
| 31 |
+
"n_heads": 2,
|
| 32 |
+
"n_layers": 6,
|
| 33 |
+
"kernel_size": 3,
|
| 34 |
+
"p_dropout": 0,
|
| 35 |
+
"resblock": "1",
|
| 36 |
+
"resblock_kernel_sizes": [
|
| 37 |
+
3,
|
| 38 |
+
7,
|
| 39 |
+
11
|
| 40 |
+
],
|
| 41 |
+
"resblock_dilation_sizes": [
|
| 42 |
+
[
|
| 43 |
+
1,
|
| 44 |
+
3,
|
| 45 |
+
5
|
| 46 |
+
],
|
| 47 |
+
[
|
| 48 |
+
1,
|
| 49 |
+
3,
|
| 50 |
+
5
|
| 51 |
+
],
|
| 52 |
+
[
|
| 53 |
+
1,
|
| 54 |
+
3,
|
| 55 |
+
5
|
| 56 |
+
]
|
| 57 |
+
],
|
| 58 |
+
"upsample_rates": [
|
| 59 |
+
12,
|
| 60 |
+
10,
|
| 61 |
+
2,
|
| 62 |
+
2
|
| 63 |
+
],
|
| 64 |
+
"upsample_initial_channel": 512,
|
| 65 |
+
"upsample_kernel_sizes": [
|
| 66 |
+
24,
|
| 67 |
+
20,
|
| 68 |
+
4,
|
| 69 |
+
4
|
| 70 |
+
],
|
| 71 |
+
"use_spectral_norm": false,
|
| 72 |
+
"gin_channels": 256,
|
| 73 |
+
"spk_embed_dim": 109
|
| 74 |
+
}
|
| 75 |
+
}
|
Jelly Hoshiumi/model_info.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"total_dataset_duration": "05:20:57",
|
| 3 |
+
"total_seconds": 19257.195729166666,
|
| 4 |
+
"sample_rate": 48000,
|
| 5 |
+
"embedder_model": "/home/user/app/models/rvc/embedders/custom/spin-v2",
|
| 6 |
+
"custom_embedder_model_hash": "1dd3ebef7d",
|
| 7 |
+
"speakers_id": 1
|
| 8 |
+
}
|
spin-v2/config.json
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.1,
|
| 3 |
+
"apply_spec_augment": true,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"HubertModelWithFinalProj"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.1,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"classifier_proj_size": 256,
|
| 10 |
+
"conv_bias": false,
|
| 11 |
+
"conv_dim": [
|
| 12 |
+
512,
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512
|
| 19 |
+
],
|
| 20 |
+
"conv_kernel": [
|
| 21 |
+
10,
|
| 22 |
+
3,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
2,
|
| 27 |
+
2
|
| 28 |
+
],
|
| 29 |
+
"conv_pos_batch_norm": false,
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_norm": "group",
|
| 45 |
+
"feat_proj_dropout": 0.0,
|
| 46 |
+
"feat_proj_layer_norm": true,
|
| 47 |
+
"final_dropout": 0.1,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.1,
|
| 50 |
+
"hidden_size": 768,
|
| 51 |
+
"initializer_range": 0.02,
|
| 52 |
+
"intermediate_size": 3072,
|
| 53 |
+
"layer_norm_eps": 1e-05,
|
| 54 |
+
"layerdrop": 0.1,
|
| 55 |
+
"mask_feature_length": 10,
|
| 56 |
+
"mask_feature_min_masks": 0,
|
| 57 |
+
"mask_feature_prob": 0.0,
|
| 58 |
+
"mask_time_length": 10,
|
| 59 |
+
"mask_time_min_masks": 2,
|
| 60 |
+
"mask_time_prob": 0.05,
|
| 61 |
+
"model_type": "hubert",
|
| 62 |
+
"num_attention_heads": 12,
|
| 63 |
+
"num_conv_pos_embedding_groups": 16,
|
| 64 |
+
"num_conv_pos_embeddings": 128,
|
| 65 |
+
"num_feat_extract_layers": 7,
|
| 66 |
+
"num_hidden_layers": 12,
|
| 67 |
+
"pad_token_id": 0,
|
| 68 |
+
"torch_dtype": "float32",
|
| 69 |
+
"transformers_version": "4.55.0",
|
| 70 |
+
"use_weighted_layer_sum": false,
|
| 71 |
+
"vocab_size": 32
|
| 72 |
+
}
|
spin-v2/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a9ac0be326057b17607a988be497793817f8274e987cf691a1b61192510f823
|
| 3 |
+
size 378356791
|
spin-v2/spin-v2.zip
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:50791eeb635bd117aa56ac0521e64931d50c13d10789885fe0d441e0407e7013
|
| 3 |
+
size 299610944
|