Hugging Face's logo

MaksymHalych
/
polyphony-seed-vc

TensorBoard
Model card Files
xet
 Metrics Community
MaksymHalych commited on
Commit
eb4efe6
·
verified ·
1 Parent(s): 963d8ea

Upload folder using huggingface_hub

Browse files
Files changed (8) hide show
  1. .hydra/config.yaml +100 -0
  2. .hydra/hydra.yaml +155 -0
  3. .hydra/overrides.yaml +1 -0
  4. DiT_epoch_00052_step_09000.pth +3 -0
  5. DiT_epoch_00055_step_09500.pth +3 -0
  6. events.out.tfevents.1774949625.85bdaf5299a2.4433.0 +3 -0
  7. ft_model.pth +3 -0
  8. train.log +81 -0
.hydra/config.yaml ADDED
@@ -0,0 +1,100 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ pretrained_model: DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ft_ema.pth
2
+ log_interval: 10
3
+ preprocess_params:
4
+ sr: 44100
5
+ spect_params:
6
+ n_fft: 2048
7
+ win_length: 2048
8
+ hop_length: 512
9
+ n_mels: 128
10
+ fmin: 0
11
+ fmax: None
12
+ model_params:
13
+ dit_type: DiT
14
+ reg_loss_type: l1
15
+ timbre_shifter:
16
+ se_db_path: ./modules/openvoice/checkpoints_v2/converter/se_db.pt
17
+ ckpt_path: ./modules/openvoice/checkpoints_v2/converter
18
+ vocoder:
19
+ type: bigvgan
20
+ name: nvidia/bigvgan_v2_44khz_128band_512x
21
+ speech_tokenizer:
22
+ type: whisper
23
+ name: openai/whisper-small
24
+ style_encoder:
25
+ dim: 192
26
+ campplus_path: campplus_cn_common.bin
27
+ DAC:
28
+ encoder_dim: 64
29
+ encoder_rates:
30
+ - 2
31
+ - 5
32
+ - 5
33
+ - 6
34
+ decoder_dim: 1536
35
+ decoder_rates:
36
+ - 6
37
+ - 5
38
+ - 5
39
+ - 2
40
+ sr: 24000
41
+ length_regulator:
42
+ channels: 768
43
+ is_discrete: false
44
+ in_channels: 768
45
+ content_codebook_size: 2048
46
+ sampling_ratios:
47
+ - 1
48
+ - 1
49
+ - 1
50
+ - 1
51
+ vector_quantize: false
52
+ n_codebooks: 1
53
+ quantizer_dropout: 0.0
54
+ f0_condition: true
55
+ n_f0_bins: 256
56
+ DiT:
57
+ hidden_dim: 768
58
+ num_heads: 12
59
+ depth: 17
60
+ class_dropout_prob: 0.1
61
+ block_size: 8192
62
+ in_channels: 128
63
+ style_condition: true
64
+ final_layer_type: mlp
65
+ target: mel
66
+ content_dim: 768
67
+ content_codebook_size: 1024
68
+ content_type: discrete
69
+ f0_condition: true
70
+ n_f0_bins: 256
71
+ content_codebooks: 1
72
+ is_causal: false
73
+ long_skip_connection: false
74
+ zero_prompt_speech_token: false
75
+ time_as_token: false
76
+ style_as_token: false
77
+ uvit_skip_connection: true
78
+ add_resblock_in_transformer: false
79
+ wavenet:
80
+ hidden_dim: 768
81
+ num_layers: 8
82
+ kernel_size: 5
83
+ dilation_rate: 1
84
+ p_dropout: 0.2
85
+ style_condition: true
86
+ loss_params:
87
+ base_lr: 0.0001
88
+ lambda_mel: 45
89
+ lambda_kl: 1.0
90
+ run_name: my_run-hydra-nfdsnflksdjflkds
91
+ batch_size: 2
92
+ max_steps: 10000
93
+ max_epochs: 1000
94
+ save_every: 500
95
+ eval_every: 1
96
+ num_workers: 0
97
+ require_features: true
98
+ pretrained_ckpt: null
99
+ tensorboard_logging: true
100
+ mlflow_logging: false
.hydra/hydra.yaml ADDED
@@ -0,0 +1,155 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ hydra:
2
+ run:
3
+ dir: runs/${run_name}_${now:%Y-%m-%d_%H-%M-%S}
4
+ sweep:
5
+ dir: multirun/${now:%Y-%m-%d}/${now:%H-%M-%S}
6
+ subdir: ${hydra.job.num}
7
+ launcher:
8
+ _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
9
+ sweeper:
10
+ _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
11
+ max_batch_size: null
12
+ params: null
13
+ help:
14
+ app_name: ${hydra.job.name}
15
+ header: '${hydra.help.app_name} is powered by Hydra.
16
+
17
+ '
18
+ footer: 'Powered by Hydra (https://hydra.cc)
19
+
20
+ Use --hydra-help to view Hydra specific help
21
+
22
+ '
23
+ template: '${hydra.help.header}
24
+
25
+ == Configuration groups ==
26
+
27
+ Compose your configuration from those groups (group=option)
28
+
29
+
30
+ $APP_CONFIG_GROUPS
31
+
32
+
33
+ == Config ==
34
+
35
+ Override anything in the config (foo.bar=value)
36
+
37
+
38
+ $CONFIG
39
+
40
+
41
+ ${hydra.help.footer}
42
+
43
+ '
44
+ hydra_help:
45
+ template: 'Hydra (${hydra.runtime.version})
46
+
47
+ See https://hydra.cc for more info.
48
+
49
+
50
+ == Flags ==
51
+
52
+ $FLAGS_HELP
53
+
54
+
55
+ == Configuration groups ==
56
+
57
+ Compose your configuration from those groups (For example, append hydra/job_logging=disabled
58
+ to command line)
59
+
60
+
61
+ $HYDRA_CONFIG_GROUPS
62
+
63
+
64
+ Use ''--cfg hydra'' to Show the Hydra config.
65
+
66
+ '
67
+ hydra_help: ???
68
+ hydra_logging:
69
+ version: 1
70
+ formatters:
71
+ simple:
72
+ format: '[%(asctime)s][HYDRA] %(message)s'
73
+ handlers:
74
+ console:
75
+ class: logging.StreamHandler
76
+ formatter: simple
77
+ stream: ext://sys.stdout
78
+ root:
79
+ level: INFO
80
+ handlers:
81
+ - console
82
+ loggers:
83
+ logging_example:
84
+ level: DEBUG
85
+ disable_existing_loggers: false
86
+ job_logging:
87
+ version: 1
88
+ formatters:
89
+ simple:
90
+ format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
91
+ handlers:
92
+ console:
93
+ class: logging.StreamHandler
94
+ formatter: simple
95
+ stream: ext://sys.stdout
96
+ file:
97
+ class: logging.FileHandler
98
+ formatter: simple
99
+ filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
100
+ root:
101
+ level: INFO
102
+ handlers:
103
+ - console
104
+ - file
105
+ disable_existing_loggers: false
106
+ env: {}
107
+ mode: RUN
108
+ searchpath: []
109
+ callbacks: {}
110
+ output_subdir: .hydra
111
+ overrides:
112
+ hydra:
113
+ - hydra.mode=RUN
114
+ task: []
115
+ job:
116
+ name: train
117
+ chdir: false
118
+ override_dirname: ''
119
+ id: ???
120
+ num: ???
121
+ config_name: config
122
+ env_set: {}
123
+ env_copy: []
124
+ config:
125
+ override_dirname:
126
+ kv_sep: '='
127
+ item_sep: ','
128
+ exclude_keys: []
129
+ runtime:
130
+ version: 1.3.2
131
+ version_base: '1.3'
132
+ cwd: /workspace/polyphony-seed-vc
133
+ config_sources:
134
+ - path: hydra.conf
135
+ schema: pkg
136
+ provider: hydra
137
+ - path: /workspace/polyphony-seed-vc/configs
138
+ schema: file
139
+ provider: main
140
+ - path: ''
141
+ schema: structured
142
+ provider: schema
143
+ output_dir: /workspace/polyphony-seed-vc/runs/my_run-hydra-nfdsnflksdjflkds_2026-03-31_09-33-42
144
+ choices:
145
+ model: uvit_whisper_44k
146
+ hydra/env: default
147
+ hydra/callbacks: null
148
+ hydra/job_logging: default
149
+ hydra/hydra_logging: default
150
+ hydra/hydra_help: default
151
+ hydra/help: default
152
+ hydra/sweeper: basic
153
+ hydra/launcher: basic
154
+ hydra/output: default
155
+ verbose: false
.hydra/overrides.yaml ADDED
@@ -0,0 +1 @@
 
 
1
+ []
DiT_epoch_00052_step_09000.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c45007f399556355beb653ebc46952947c71cfef26ff5f1402e2dcf1dcd916a
3
+ size 2246796075
DiT_epoch_00055_step_09500.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8cbccbbb94c152b16394fd2ef52f65e2179629734eb1341cd7fa4da5e30f6a98
3
+ size 2246796075
events.out.tfevents.1774949625.85bdaf5299a2.4433.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5a20464a3e2876dec659afbed2698613464cd75fd12efd9831c375eee92ac68
3
+ size 51905
ft_model.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c03d59f289a60e95405f0a3ba839882fc60f7c61b323d77ade48d0bdc8cb0d1
3
+ size 785751672
train.log ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [2026-03-31 09:33:45,418][__main__][INFO] - Loaded HF checkpoint from ./checkpoints/models--Plachta--Seed-VC/snapshots/257283f9f41585055e8f858fba4fd044e5caed6e/DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ft_ema.pth
2
+ [2026-03-31 09:34:45,467][__main__][INFO] - Eval loss at epoch 1: 0.5503
3
+ [2026-03-31 09:35:43,500][__main__][INFO] - Eval loss at epoch 2: 0.5421
4
+ [2026-03-31 09:36:37,482][__main__][INFO] - Saving checkpoint
5
+ [2026-03-31 09:36:44,539][__main__][INFO] - Eval loss at epoch 3: 0.5184
6
+ [2026-03-31 09:37:42,581][__main__][INFO] - Eval loss at epoch 4: 0.5492
7
+ [2026-03-31 09:38:41,957][__main__][INFO] - Eval loss at epoch 5: 0.5689
8
+ [2026-03-31 09:39:32,846][__main__][INFO] - Saving checkpoint
9
+ [2026-03-31 09:39:43,004][__main__][INFO] - Eval loss at epoch 6: 0.5222
10
+ [2026-03-31 09:40:41,991][__main__][INFO] - Eval loss at epoch 7: 0.5251
11
+ [2026-03-31 09:41:41,652][__main__][INFO] - Eval loss at epoch 8: 0.5492
12
+ [2026-03-31 09:42:26,318][__main__][INFO] - Saving checkpoint
13
+ [2026-03-31 09:42:41,890][__main__][INFO] - Eval loss at epoch 9: 0.5331
14
+ [2026-03-31 09:43:41,666][__main__][INFO] - Eval loss at epoch 10: 0.5145
15
+ [2026-03-31 09:44:40,606][__main__][INFO] - Eval loss at epoch 11: 0.5239
16
+ [2026-03-31 09:45:20,529][__main__][INFO] - Saving checkpoint
17
+ [2026-03-31 09:45:39,905][__main__][INFO] - Eval loss at epoch 12: 0.5110
18
+ [2026-03-31 09:46:39,583][__main__][INFO] - Eval loss at epoch 13: 0.5125
19
+ [2026-03-31 09:47:39,332][__main__][INFO] - Eval loss at epoch 14: 0.5054
20
+ [2026-03-31 09:48:18,655][__main__][INFO] - Saving checkpoint
21
+ [2026-03-31 09:48:39,514][__main__][INFO] - Eval loss at epoch 15: 0.5371
22
+ [2026-03-31 09:49:37,754][__main__][INFO] - Eval loss at epoch 16: 0.4939
23
+ [2026-03-31 09:50:36,564][__main__][INFO] - Eval loss at epoch 17: 0.5318
24
+ [2026-03-31 09:51:12,058][__main__][INFO] - Saving checkpoint
25
+ [2026-03-31 09:51:35,973][__main__][INFO] - Eval loss at epoch 18: 0.5058
26
+ [2026-03-31 09:52:34,475][__main__][INFO] - Eval loss at epoch 19: 0.5134
27
+ [2026-03-31 09:53:34,121][__main__][INFO] - Eval loss at epoch 20: 0.5317
28
+ [2026-03-31 09:54:08,935][__main__][INFO] - Saving checkpoint
29
+ [2026-03-31 09:54:34,447][__main__][INFO] - Eval loss at epoch 21: 0.5369
30
+ [2026-03-31 09:55:33,298][__main__][INFO] - Eval loss at epoch 22: 0.5119
31
+ [2026-03-31 09:56:32,735][__main__][INFO] - Eval loss at epoch 23: 0.5454
32
+ [2026-03-31 09:57:03,544][__main__][INFO] - Saving checkpoint
33
+ [2026-03-31 09:57:33,559][__main__][INFO] - Eval loss at epoch 24: 0.5248
34
+ [2026-03-31 09:58:33,395][__main__][INFO] - Eval loss at epoch 25: 0.5219
35
+ [2026-03-31 09:59:32,889][__main__][INFO] - Eval loss at epoch 26: 0.5220
36
+ [2026-03-31 09:59:59,173][__main__][INFO] - Saving checkpoint
37
+ [2026-03-31 10:00:32,473][__main__][INFO] - Eval loss at epoch 27: 0.5058
38
+ [2026-03-31 10:01:31,127][__main__][INFO] - Eval loss at epoch 28: 0.5103
39
+ [2026-03-31 10:02:29,941][__main__][INFO] - Eval loss at epoch 29: 0.5056
40
+ [2026-03-31 10:02:54,663][__main__][INFO] - Saving checkpoint
41
+ [2026-03-31 10:03:30,831][__main__][INFO] - Eval loss at epoch 30: 0.5360
42
+ [2026-03-31 10:04:29,601][__main__][INFO] - Eval loss at epoch 31: 0.4924
43
+ [2026-03-31 10:05:29,342][__main__][INFO] - Eval loss at epoch 32: 0.4983
44
+ [2026-03-31 10:05:48,616][__main__][INFO] - Saving checkpoint
45
+ [2026-03-31 10:06:30,850][__main__][INFO] - Eval loss at epoch 33: 0.4768
46
+ [2026-03-31 10:07:30,413][__main__][INFO] - Eval loss at epoch 34: 0.4945
47
+ [2026-03-31 10:08:29,116][__main__][INFO] - Eval loss at epoch 35: 0.4879
48
+ [2026-03-31 10:08:46,756][__main__][INFO] - Saving checkpoint
49
+ [2026-03-31 10:09:29,204][__main__][INFO] - Eval loss at epoch 36: 0.5048
50
+ [2026-03-31 10:10:28,796][__main__][INFO] - Eval loss at epoch 37: 0.5370
51
+ [2026-03-31 10:11:28,681][__main__][INFO] - Eval loss at epoch 38: 0.4983
52
+ [2026-03-31 10:11:41,783][__main__][INFO] - Saving checkpoint
53
+ [2026-03-31 10:12:29,540][__main__][INFO] - Eval loss at epoch 39: 0.4753
54
+ [2026-03-31 10:13:29,204][__main__][INFO] - Eval loss at epoch 40: 0.5071
55
+ [2026-03-31 10:14:27,339][__main__][INFO] - Eval loss at epoch 41: 0.4829
56
+ [2026-03-31 10:14:38,158][__main__][INFO] - Saving checkpoint
57
+ [2026-03-31 10:15:27,468][__main__][INFO] - Eval loss at epoch 42: 0.4967
58
+ [2026-03-31 10:16:26,999][__main__][INFO] - Eval loss at epoch 43: 0.4897
59
+ [2026-03-31 10:17:25,940][__main__][INFO] - Eval loss at epoch 44: 0.4786
60
+ [2026-03-31 10:17:32,914][__main__][INFO] - Saving checkpoint
61
+ [2026-03-31 10:18:25,264][__main__][INFO] - Eval loss at epoch 45: 0.4875
62
+ [2026-03-31 10:19:23,252][__main__][INFO] - Eval loss at epoch 46: 0.4990
63
+ [2026-03-31 10:20:23,505][__main__][INFO] - Eval loss at epoch 47: 0.5173
64
+ [2026-03-31 10:20:26,632][__main__][INFO] - Saving checkpoint
65
+ [2026-03-31 10:21:21,067][__main__][INFO] - Eval loss at epoch 48: 0.4705
66
+ [2026-03-31 10:22:20,641][__main__][INFO] - Eval loss at epoch 49: 0.4964
67
+ [2026-03-31 10:23:16,196][__main__][INFO] - Saving checkpoint
68
+ [2026-03-31 10:23:20,179][__main__][INFO] - Eval loss at epoch 50: 0.4836
69
+ [2026-03-31 10:24:19,462][__main__][INFO] - Eval loss at epoch 51: 0.5069
70
+ [2026-03-31 10:25:18,946][__main__][INFO] - Eval loss at epoch 52: 0.4988
71
+ [2026-03-31 10:26:12,043][__main__][INFO] - Saving checkpoint
72
+ [2026-03-31 10:26:19,038][__main__][INFO] - Eval loss at epoch 53: 0.4600
73
+ [2026-03-31 10:27:17,638][__main__][INFO] - Eval loss at epoch 54: 0.4965
74
+ [2026-03-31 10:28:17,309][__main__][INFO] - Eval loss at epoch 55: 0.4579
75
+ [2026-03-31 10:29:05,862][__main__][INFO] - Saving checkpoint
76
+ [2026-03-31 10:29:16,698][__main__][INFO] - Eval loss at epoch 56: 0.5023
77
+ [2026-03-31 10:30:15,627][__main__][INFO] - Eval loss at epoch 57: 0.5086
78
+ [2026-03-31 10:31:13,689][__main__][INFO] - Eval loss at epoch 58: 0.5095
79
+ [2026-03-31 10:32:03,283][__main__][INFO] - Eval loss at epoch 59: 0.4990
80
+ [2026-03-31 10:32:03,283][__main__][INFO] - Saving final model
81
+ [2026-03-31 10:32:03,744][__main__][INFO] - Final model saved at /workspace/polyphony-seed-vc/runs/my_run-hydra-nfdsnflksdjflkds_2026-03-31_09-33-42/ft_model.pth