Hugging Face's logo

LIT-TTS
/
voice_design

Safetensors
Model card Files
xet
 Community
suwon commited on
Commit
ff15069
·
verified ·
1 Parent(s): 54af85e

Upload folder using huggingface_hub

Browse files
Files changed (11) hide show
  1. .gitattributes +1 -0
  2. checkpoints/v9/chat_template.jinja +89 -0
  3. checkpoints/v9/config.json +134 -0
  4. checkpoints/v9/model.safetensors +3 -0
  5. checkpoints/v9/optimizer.bin +3 -0
  6. checkpoints/v9/random_states_0.pkl +3 -0
  7. checkpoints/v9/scheduler.bin +3 -0
  8. checkpoints/v9/tokenizer.json +3 -0
  9. checkpoints/v9/tokenizer_config.json +23 -0
  10. checkpoints/v9/train_config.json +227 -0
  11. checkpoints/v9/trainer_state.json +6 -0
.gitattributes CHANGED
@@ -41,3 +41,4 @@ checkpoints/v6/tokenizer.json filter=lfs diff=lfs merge=lfs -text
41
  checkpoints/v6.1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
42
  checkpoints/v7/tokenizer.json filter=lfs diff=lfs merge=lfs -text
43
  checkpoints/v8/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
41
  checkpoints/v6.1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
42
  checkpoints/v7/tokenizer.json filter=lfs diff=lfs merge=lfs -text
43
  checkpoints/v8/tokenizer.json filter=lfs diff=lfs merge=lfs -text
44
+ checkpoints/v9/tokenizer.json filter=lfs diff=lfs merge=lfs -text
checkpoints/v9/chat_template.jinja ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {{- messages[0].content + '\n\n' }}
5
+ {%- endif %}
6
+ {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
7
+ {%- for tool in tools %}
8
+ {{- "\n" }}
9
+ {{- tool | tojson }}
10
+ {%- endfor %}
11
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
12
+ {%- else %}
13
+ {%- if messages[0].role == 'system' %}
14
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
15
+ {%- endif %}
16
+ {%- endif %}
17
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
+ {%- for message in messages[::-1] %}
19
+ {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
+ {%- set ns.multi_step_tool = false %}
22
+ {%- set ns.last_query_index = index %}
23
+ {%- endif %}
24
+ {%- endfor %}
25
+ {%- for message in messages %}
26
+ {%- if message.content is string %}
27
+ {%- set content = message.content %}
28
+ {%- else %}
29
+ {%- set content = '' %}
30
+ {%- endif %}
31
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
32
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
33
+ {%- elif message.role == "assistant" %}
34
+ {%- set reasoning_content = '' %}
35
+ {%- if message.reasoning_content is string %}
36
+ {%- set reasoning_content = message.reasoning_content %}
37
+ {%- else %}
38
+ {%- if '</think>' in content %}
39
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
40
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
41
+ {%- endif %}
42
+ {%- endif %}
43
+ {%- if loop.index0 > ns.last_query_index %}
44
+ {%- if loop.last or (not loop.last and reasoning_content) %}
45
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
46
+ {%- else %}
47
+ {{- '<|im_start|>' + message.role + '\n' + content }}
48
+ {%- endif %}
49
+ {%- else %}
50
+ {{- '<|im_start|>' + message.role + '\n' + content }}
51
+ {%- endif %}
52
+ {%- if message.tool_calls %}
53
+ {%- for tool_call in message.tool_calls %}
54
+ {%- if (loop.first and content) or (not loop.first) %}
55
+ {{- '\n' }}
56
+ {%- endif %}
57
+ {%- if tool_call.function %}
58
+ {%- set tool_call = tool_call.function %}
59
+ {%- endif %}
60
+ {{- '<tool_call>\n{"name": "' }}
61
+ {{- tool_call.name }}
62
+ {{- '", "arguments": ' }}
63
+ {%- if tool_call.arguments is string %}
64
+ {{- tool_call.arguments }}
65
+ {%- else %}
66
+ {{- tool_call.arguments | tojson }}
67
+ {%- endif %}
68
+ {{- '}\n</tool_call>' }}
69
+ {%- endfor %}
70
+ {%- endif %}
71
+ {{- '<|im_end|>\n' }}
72
+ {%- elif message.role == "tool" %}
73
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
74
+ {{- '<|im_start|>user' }}
75
+ {%- endif %}
76
+ {{- '\n<tool_response>\n' }}
77
+ {{- content }}
78
+ {{- '\n</tool_response>' }}
79
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
80
+ {{- '<|im_end|>\n' }}
81
+ {%- endif %}
82
+ {%- endif %}
83
+ {%- endfor %}
84
+ {%- if add_generation_prompt %}
85
+ {{- '<|im_start|>assistant\n' }}
86
+ {%- if enable_thinking is defined and enable_thinking is false %}
87
+ {{- '<think>\n\n</think>\n\n' }}
88
+ {%- endif %}
89
+ {%- endif %}
checkpoints/v9/config.json ADDED
@@ -0,0 +1,134 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "OmniVoice"
4
+ ],
5
+ "attn_implementation": "flex_attention",
6
+ "audio_codebook_weights": [
7
+ 8,
8
+ 8,
9
+ 6,
10
+ 6,
11
+ 4,
12
+ 4,
13
+ 2,
14
+ 2
15
+ ],
16
+ "audio_mask_id": 1024,
17
+ "audio_vocab_size": 1025,
18
+ "bos_token_id": null,
19
+ "codebook_group_condition_scale": 1.0,
20
+ "codebook_group_state_mode": "teacher_forcing",
21
+ "codebook_group_teacher_forcing_ratio": 1.0,
22
+ "codebook_groups": [
23
+ [
24
+ 0,
25
+ 1
26
+ ],
27
+ [
28
+ 2,
29
+ 3,
30
+ 4,
31
+ 5
32
+ ],
33
+ [
34
+ 6,
35
+ 7
36
+ ]
37
+ ],
38
+ "codebook_head_type": "grouped",
39
+ "compile_flex_attention_mask": false,
40
+ "dtype": "float32",
41
+ "eos_token_id": 151645,
42
+ "llm_config": {
43
+ "_name_or_path": "",
44
+ "architectures": [
45
+ "Qwen3ForCausalLM"
46
+ ],
47
+ "attention_bias": false,
48
+ "attention_dropout": 0.0,
49
+ "bos_token_id": 151643,
50
+ "chunk_size_feed_forward": 0,
51
+ "dtype": "float32",
52
+ "eos_token_id": 151645,
53
+ "head_dim": 128,
54
+ "hidden_act": "silu",
55
+ "hidden_size": 1024,
56
+ "id2label": {
57
+ "0": "LABEL_0",
58
+ "1": "LABEL_1"
59
+ },
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 3072,
62
+ "is_encoder_decoder": false,
63
+ "label2id": {
64
+ "LABEL_0": 0,
65
+ "LABEL_1": 1
66
+ },
67
+ "layer_types": [
68
+ "full_attention",
69
+ "full_attention",
70
+ "full_attention",
71
+ "full_attention",
72
+ "full_attention",
73
+ "full_attention",
74
+ "full_attention",
75
+ "full_attention",
76
+ "full_attention",
77
+ "full_attention",
78
+ "full_attention",
79
+ "full_attention",
80
+ "full_attention",
81
+ "full_attention",
82
+ "full_attention",
83
+ "full_attention",
84
+ "full_attention",
85
+ "full_attention",
86
+ "full_attention",
87
+ "full_attention",
88
+ "full_attention",
89
+ "full_attention",
90
+ "full_attention",
91
+ "full_attention",
92
+ "full_attention",
93
+ "full_attention",
94
+ "full_attention",
95
+ "full_attention"
96
+ ],
97
+ "max_position_embeddings": 40960,
98
+ "max_window_layers": 28,
99
+ "model_type": "qwen3",
100
+ "num_attention_heads": 16,
101
+ "num_hidden_layers": 28,
102
+ "num_key_value_heads": 8,
103
+ "output_attentions": false,
104
+ "output_hidden_states": false,
105
+ "pad_token_id": null,
106
+ "problem_type": null,
107
+ "return_dict": true,
108
+ "rms_norm_eps": 1e-06,
109
+ "rope_parameters": {
110
+ "rope_theta": 1000000,
111
+ "rope_type": "default"
112
+ },
113
+ "sliding_window": null,
114
+ "tie_word_embeddings": true,
115
+ "use_cache": true,
116
+ "use_sliding_window": false,
117
+ "vocab_size": 151683
118
+ },
119
+ "loss_codebook_weighting_mode": "static",
120
+ "loss_entropy_regularization_weight": 0.0,
121
+ "loss_focal_gamma": 0.0,
122
+ "loss_label_smoothing": 0.01,
123
+ "loss_mask_ratio_codebook_weight_buckets": null,
124
+ "model_type": "omnivoice",
125
+ "num_audio_codebook": 8,
126
+ "pad_token_id": 151643,
127
+ "paired_ref_target_base_loss_weight": 1.0,
128
+ "paired_ref_target_normalize_loss_by_active_weights": true,
129
+ "paired_ref_target_self_loss": true,
130
+ "paired_ref_target_self_loss_weight": 0.2,
131
+ "paired_ref_use_boundary_tokens": true,
132
+ "transformers_version": "5.5.4",
133
+ "validate_audio_token_range": true
134
+ }
checkpoints/v9/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5887688011f6b5aae78db6afad450d6b7d2ee518bf934babf6bce7589ec81ad9
3
+ size 2458761872
checkpoints/v9/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fe47f4cfee4cd9a5586afe9832ddd8147677a38388047ed6b791f4a2da53ee5
3
+ size 4917720171
checkpoints/v9/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8319cd9fabdfaddcfe88f29b894435d9b531dbff300438df7cff835d02bf434
3
+ size 14757
checkpoints/v9/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1be7f1680dd8ccd57815aaa274dc1b47f15174c5d4b4bfc4764b6bc74a78f80
3
+ size 1465
checkpoints/v9/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c16473266e9ae04e079c2c3ff62e899d71426108398c078c7e375b7b0e535dc
3
+ size 11425357
checkpoints/v9/tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": null,
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<|im_end|>",
7
+ "errors": "replace",
8
+ "extra_special_tokens": [
9
+ "<|ref_text_start|>",
10
+ "<|ref_text_end|>",
11
+ "<|target_text_start|>",
12
+ "<|target_text_end|>",
13
+ "<|ref_audio_start|>",
14
+ "<|ref_audio_end|>",
15
+ "<|target_audio_start|>"
16
+ ],
17
+ "is_local": false,
18
+ "model_max_length": 131072,
19
+ "pad_token": "<|endoftext|>",
20
+ "split_special_tokens": false,
21
+ "tokenizer_class": "Qwen2Tokenizer",
22
+ "unk_token": null
23
+ }
checkpoints/v9/train_config.json ADDED
@@ -0,0 +1,227 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "output_dir": "/home/infidea/suwon/omni-voice-advance-plus-pair/exp/omnivoice_0.6B-PAIR-FT-hccd-TF-up",
3
+ "data_config": "/home/infidea/suwon/omni-voice-advance-plus-pair/examples/config/data_config_pair_devpack.json",
4
+ "llm_name_or_path": "Qwen/Qwen3-0.6B",
5
+ "tokenizer_name_or_path": null,
6
+ "expected_llm_model_type": null,
7
+ "expected_llm_hidden_size": null,
8
+ "expected_llm_intermediate_size": null,
9
+ "expected_llm_num_hidden_layers": null,
10
+ "expected_llm_num_attention_heads": null,
11
+ "expected_llm_num_key_value_heads": null,
12
+ "expected_llm_vocab_size": null,
13
+ "audio_vocab_size": 1025,
14
+ "audio_mask_id": 1024,
15
+ "num_audio_codebook": 8,
16
+ "audio_codebook_weights": [
17
+ 8,
18
+ 8,
19
+ 6,
20
+ 6,
21
+ 4,
22
+ 4,
23
+ 2,
24
+ 2
25
+ ],
26
+ "drop_cond_ratio": 0.1,
27
+ "force_uncond_ratio": 0.07,
28
+ "uncond_mask_ratio_range": [
29
+ 0.7,
30
+ 1.0
31
+ ],
32
+ "prompt_ratio_range": [
33
+ 0.05,
34
+ 0.3
35
+ ],
36
+ "mask_ratio_range": [
37
+ 0.3,
38
+ 0.9
39
+ ],
40
+ "min_masked_audio_tokens": 8,
41
+ "language_ratio": 0.0,
42
+ "use_pinyin_ratio": 0.0,
43
+ "instruct_ratio": 1.0,
44
+ "only_instruct_ratio": 0.3,
45
+ "log_codebook_losses": true,
46
+ "loss_label_smoothing": 0.01,
47
+ "loss_focal_gamma": 0.0,
48
+ "loss_entropy_regularization_weight": 0.0,
49
+ "loss_codebook_weighting_mode": "static",
50
+ "loss_mask_ratio_codebook_weight_buckets": null,
51
+ "use_paired_reference_training": true,
52
+ "paired_ref_full_mask": true,
53
+ "paired_ref_drop_cond_ratio": 0.0,
54
+ "paired_ref_branch_weights": {
55
+ "full": 0.55,
56
+ "partial": 0.3,
57
+ "span": 0.1,
58
+ "ref_perturbed": 0.05,
59
+ "no_ref": 0.0
60
+ },
61
+ "paired_ref_partial_mask_ratio_range": [
62
+ 0.25,
63
+ 0.75
64
+ ],
65
+ "paired_ref_span_mask_ratio_range": [
66
+ 0.1,
67
+ 0.35
68
+ ],
69
+ "paired_ref_perturbed_target_full_ratio": 0.5,
70
+ "paired_ref_perturb_crop_ratio_range": [
71
+ 0.6,
72
+ 1.0
73
+ ],
74
+ "paired_ref_perturb_token_drop_ratio": 0.02,
75
+ "paired_ref_perturb_token_replace_ratio": 0.01,
76
+ "paired_ref_use_boundary_tokens": true,
77
+ "paired_ref_target_self_loss": true,
78
+ "paired_ref_target_base_loss_weight": 1.0,
79
+ "paired_ref_target_self_loss_weight": 0.2,
80
+ "paired_ref_target_normalize_loss_by_active_weights": true,
81
+ "paired_ref_target_self_loss_mask_ratio_range": [
82
+ 0.3,
83
+ 0.9
84
+ ],
85
+ "paired_ref_target_self_loss_prompt_ratio_range": [
86
+ 0.05,
87
+ 0.25
88
+ ],
89
+ "paired_ref_fallback_to_prompt": true,
90
+ "paired_ref_fallback_prompt_ratio_range": [
91
+ 0.05,
92
+ 0.3
93
+ ],
94
+ "paired_ref_fallback_min_ref_tokens": 1,
95
+ "paired_ref_fallback_min_target_tokens": 1,
96
+ "require_ref_audio_tokens": false,
97
+ "codebook_head_type": "grouped",
98
+ "codebook_groups": [
99
+ [
100
+ 0,
101
+ 1
102
+ ],
103
+ [
104
+ 2,
105
+ 3,
106
+ 4,
107
+ 5
108
+ ],
109
+ [
110
+ 6,
111
+ 7
112
+ ]
113
+ ],
114
+ "codebook_group_names": [
115
+ "coarse",
116
+ "mid",
117
+ "fine"
118
+ ],
119
+ "codebook_group_state_mode": "teacher_forcing",
120
+ "codebook_group_teacher_forcing_ratio": 1.0,
121
+ "codebook_group_condition_scale": 1.0,
122
+ "codebook_group_tf_schedule_enabled": true,
123
+ "codebook_group_tf_schedule_decay_start": 100000,
124
+ "codebook_group_tf_schedule_decay_end": 200000,
125
+ "codebook_group_tf_schedule_start_ratio": 1.0,
126
+ "codebook_group_tf_schedule_end_ratio": 0.0,
127
+ "codebook_group_tf_schedule_decay_type": "inverse_sigmoid",
128
+ "codebook_group_tf_schedule_sigmoid_k": 8.0,
129
+ "codebook_group_tf_eval_force_expected": true,
130
+ "resume_from_checkpoint": "/home/infidea/suwon/omni-voice-advance-plus-pair/exp/omnivoice_0.6B-PAIR-FT-hccd-TF-up/checkpoint-72000",
131
+ "init_from_checkpoint": "k2-fsa/OmniVoice",
132
+ "allow_partial_resume": false,
133
+ "learning_rate": 2e-05,
134
+ "weight_decay": 0.01,
135
+ "max_grad_norm": 1.0,
136
+ "steps": 500000,
137
+ "seed": 42,
138
+ "lr_scheduler_type": "cosine",
139
+ "warmup_type": "steps",
140
+ "warmup_ratio": 0.03,
141
+ "warmup_steps": 10000,
142
+ "lr_scheduler_total_steps": 500000,
143
+ "lr_min_ratio": 0.0,
144
+ "lr_resync_on_resume": true,
145
+ "lr_rebase_on_resume": true,
146
+ "batch_tokens": 16384,
147
+ "gradient_accumulation_steps": 4,
148
+ "num_workers": 2,
149
+ "mixed_precision": "bf16",
150
+ "allow_tf32": true,
151
+ "require_cuda": true,
152
+ "use_deepspeed": false,
153
+ "deepspeed_config": null,
154
+ "compile_flex_attention_mask": false,
155
+ "validate_audio_token_range": true,
156
+ "skip_bad_batches": true,
157
+ "max_consecutive_batch_skips": 50,
158
+ "loss_spike_skip_factor": 0.0,
159
+ "loss_spike_warmup_steps": 100,
160
+ "attn_implementation": "flex_attention",
161
+ "logging_steps": 100,
162
+ "eval_steps": 2000,
163
+ "save_steps": 2000,
164
+ "keep_last_n_checkpoints": 3,
165
+ "use_wandb": true,
166
+ "wandb_project": "omnivoice_0.6B-FT-hccd-Teacher-Forcing-UP",
167
+ "wandb_entity": null,
168
+ "wandb_run_name": null,
169
+ "wandb_group": "hccd_pair",
170
+ "wandb_tags": [
171
+ "paired-zero-shot",
172
+ "hccd",
173
+ "grouped-head"
174
+ ],
175
+ "wandb_mode": null,
176
+ "inference_logging_steps": 1000,
177
+ "inference_logging_text": null,
178
+ "inference_logging_language": null,
179
+ "inference_logging_ref_audio": null,
180
+ "inference_logging_ref_text": null,
181
+ "inference_logging_num_step": 32,
182
+ "inference_logging_guidance_scale": 2.0,
183
+ "inference_logging_speed": 1.0,
184
+ "inference_logging_duration": null,
185
+ "inference_logging_t_shift": 0.1,
186
+ "inference_logging_layer_penalty_factor": 5.0,
187
+ "inference_logging_position_temperature": 5.0,
188
+ "inference_logging_class_temperature": 0.0,
189
+ "inference_logging_cfg_confidence_decouple": true,
190
+ "inference_logging_denoise": true,
191
+ "inference_audio_tokenizer_path": null,
192
+ "inference_logging_jsonl_dir": [
193
+ "/home/infidea/tts-data/suwon/OmniVoice_data/zs_data/ch_data/txts",
194
+ "/home/infidea/tts-data/suwon/OmniVoice_data/zs_data/en_data/txts",
195
+ "/home/infidea/tts-data/suwon/OmniVoice_data/zs_data/ja_data/txts",
196
+ "/home/infidea/tts-data/suwon/OmniVoice_data/zs_data/ko_data/txts",
197
+ "/home/infidea/tts-data/suwon/OmniVoice_data/zs_data/pre_en_data/txts",
198
+ "/home/infidea/tts-data/suwon/OmniVoice_data/zs_data/pre_ja_data/txts",
199
+ "/home/infidea/tts-data/suwon/OmniVoice_data/zs_data/pre_ko_data/txts",
200
+ "/home/infidea/tts-data/suwon/OmniVoice_data/zs_data/pre_vi_data/txts"
201
+ ],
202
+ "inference_logging_voice_design_jsonl_dir": [
203
+ "/home/infidea/tts-data/suwon/OmniVoice_data/voice_design_enhance/ch/txts",
204
+ "/home/infidea/tts-data/suwon/OmniVoice_data/voice_design_enhance/en/txts",
205
+ "/home/infidea/tts-data/suwon/OmniVoice_data/voice_design_enhance/ja/txts",
206
+ "/home/infidea/tts-data/suwon/OmniVoice_data/voice_design_enhance/ko/txts",
207
+ "/home/infidea/tts-data/suwon/OmniVoice_data/voice_design_enhance/vi/txts",
208
+ "/home/infidea/tts-data/suwon/OmniVoice_data/voice_design_enhance/ru/txts"
209
+ ],
210
+ "inference_logging_save_eval_artifacts": true,
211
+ "inference_logging_eval_dir": null,
212
+ "inference_logging_dev_pack": null,
213
+ "inference_logging_audio_per_language": 1,
214
+ "inference_eval_metrics": [
215
+ "cer",
216
+ "sim",
217
+ "utmos"
218
+ ],
219
+ "inference_eval_models_dir": "/home/infidea/suwon/omni-voice-advance-plus-pair/examples/download/tts_eval_models",
220
+ "best_metric": "quality",
221
+ "best_metric_ema_beta": 0.7,
222
+ "best_quality_weights": {
223
+ "sim": 1.0,
224
+ "cer": 1.0,
225
+ "mos": 1.0
226
+ }
227
+ }
checkpoints/v9/trainer_state.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "step": 94000,
3
+ "epoch": 0,
4
+ "metric_ema": 0.06377743590076118,
5
+ "best_metric": "quality"
6
+ }