Upload 5 files

args.json

@@ -0,0 +1,199 @@
+{
+    "base_config": "config/valle.json",
+    "dataset": [
+        "LibriTTS",
+    ],
+    "dataset_path": {
+        "LibriTTS": "/home/aiscuser/librilight",
+    },
+    "exp_name": "amphion_librilight_valle_nar_gpus_adamw_coswarmup_16k_2e-4_load_libritts_mi200_new_symbols_dict",
+    "log_dir": "/blob/v-yuancwang/librilight_valle",
+    "model": {
+        "add_prenet": false,
+        "decoder_dim": 1024,
+        "nar_scale_factor": 1,
+        "nhead": 16,
+        "norm_first": true,
+        "num_decoder_layers": 12,
+        "num_quantizers": 8,
+        "prefix_mode": 1,
+        "prepend_bos": false,
+        "scaling_xformers": false,
+        "share_embedding": true,
+    },
+    "model_type": "VALLE",
+    "preprocess": {
+        "acoustic_token_dir": "acoutic_tokens",
+        "acoustic_token_extractor": "Encodec",
+        "align_mel_duration": false,
+        "audio_dir": "audios",
+        "bits": 8,
+        "contentvec_dir": "contentvec",
+        "data_augment": false,
+        "dur_dir": "durs",
+        "duration_dir": "duration",
+        "emo2id": "emo2id.json",
+        "energy_dir": "energys",
+        "energy_extract_mode": "from_mel",
+        "energy_norm": false,
+        "energy_remove_outlier": false,
+        "extract_acoustic_token": true,
+        "extract_audio": false,
+        "extract_contentvec_feature": false,
+        "extract_duration": false,
+        "extract_energy": false,
+        "extract_label": false,
+        "extract_linear_spec": false,
+        "extract_mcep": false,
+        "extract_mel": false,
+        "extract_mert_feature": false,
+        "extract_phoneme": true,
+        "extract_pitch": false,
+        "extract_uv": false,
+        "extract_wenet_feature": false,
+        "extract_whisper_feature": false,
+        "file_lst": "file.lst",
+        "fmax": 12000,
+        "fmin": 0,
+        "hop_size": 120,
+        "lab_dir": "labs",
+        "label_dir": "labels",
+        "lexicon_path": "./text/lexicon/librispeech-lexicon.txt",
+        "linear_dir": "linears",
+        "max_duration": 30,
+        "max_token": 1600,
+        "mcep_dir": "mcep",
+        "mel_dir": "mels",
+        "mel_extract_mode": "",
+        "mel_min_max_norm": false,
+        "min_duration": 0.5,
+        "min_level_db": -115,
+        "min_token": 45,
+        "n_fft": 1024,
+        "n_mel": 80,
+        "num_silent_frames": 8,
+        "phone_dir": "phones",
+        "phone_energy_dir": "phone_energys",
+        "phone_pitch_dir": "phone_pitches",
+        "phone_seq_file": "phone_seq_file",
+        "pitch_dir": "pitches",
+        "pitch_extractor": "parselmouth",
+        "pitch_norm": false,
+        "pitch_remove_outlier": false,
+        "processed_dir": "/home/aiscuser/librilight",
+        "raw_data": "raw_data",
+        "ref_level_db": 20,
+        "sample_rate": 24000,
+        "sampling_rate": "24000",
+        "spk2id": "spk2id.json",
+        "symbols_dict": "symbols_from_libritts.dict",
+        "text_extractor": "espeak",
+        "train_file": "train_filter.json",
+        "trim_fft_size": 512,
+        "trim_hop_size": 128,
+        "trim_silence": false,
+        "trim_top_db": 30,
+        "trimmed_wav_dir": "trimmed_wavs",
+        "use_acoustic_token": true,
+        "use_audio": false,
+        "use_dur": false,
+        "use_emoid": false,
+        "use_frame_duration": false,
+        "use_frame_energy": false,
+        "use_frame_pitch": false,
+        "use_lab": false,
+        "use_label": false,
+        "use_linear": false,
+        "use_log_scale_energy": false,
+        "use_log_scale_pitch": false,
+        "use_mel": false,
+        "use_min_max_norm_mel": false,
+        "use_one_hot": false,
+        "use_phn_seq": false,
+        "use_phone": true,
+        "use_phone_duration": false,
+        "use_phone_energy": false,
+        "use_phone_pitch": false,
+        "use_spkid": false,
+        "use_text": false,
+        "use_uv": false,
+        "use_wav": false,
+        "use_wenet": false,
+        "utt2emo": "utt2emo",
+        "utt2spk": "utt2spk",
+        "uv_dir": "uvs",
+        "valid_file": "test_filter.json",
+        "wav_dir": "wavs",
+        "wenet_dir": "wenet",
+        "win_size": 480,
+    },
+    "supported_model_type": [
+        "GANVocoder",
+        "Fastspeech2",
+        "DiffSVC",
+        "Transformer",
+        "EDM",
+        "CD",
+    ],
+    "train": {
+        "adamw": {
+            "lr": 0.0004,
+        },
+        "base_lr": 0.0001,
+        "batch_size": 5,
+        "clip_norm": false,
+        "clip_value": true,
+        "dataloader": {
+            "num_worker": 28,
+            "pin_memory": true,
+        },
+        "ddp": false,
+        "gradient_accumulation_step": 1,
+        "keep_checkpoint_max": 5,
+        "keep_last": [
+            3,
+            -1,
+        ],
+        "load_pretrained_model": false,
+        "load_pretrained_model_path": "/blob/v-yuancwang/librilight_valle/libritts_ckpt/final_epoch-0100_step-0837900_loss-3.883116",
+        "max_epoch": 40,
+        "max_sentences": 8,
+        "max_steps": 1000000,
+        "max_tokens": 4800,
+        "multi_speaker_training": false,
+        "optimizer": "AdamW",
+        "optimizer_name": "AdamW",
+        "random_seed": 10086,
+        "reducelronplateau": {
+            "factor": 0.8,
+            "min_lr": 0.0001,
+            "patience": 10,
+        },
+        "reset_interval": 200,
+        "run_eval": [
+            false,
+            true,
+        ],
+        "sampler": {
+            "drop_last": true,
+            "holistic_shuffle": true,
+        },
+        "save_checkpoint_stride": [
+            1,
+            1,
+        ],
+        "save_checkpoints_steps": 10000,
+        "save_summary_steps": 500,
+        "scheduler": "ReduceLROnPlateau",
+        "scheduler_name": "Cosine",
+        "start_epoch": 1,
+        "total_training_steps": 50000,
+        "tracker": [
+            "tensorboard",
+        ],
+        "train_stage": 0,
+        "valid_interval": 1000,
+        "warmup_steps": 16000,
+    },
+    "use_custom_dataset": false,
+}

symbols.dict

@@ -0,0 +1,138 @@
+<eps> 0
+! 1
+" 2
+, 3
+. 4
+: 5
+; 6
+? 7
+_ 8
+aɪ 9
+aɪw 10
+aɪə 11
+aɪɚ 12
+aʊ 13
+b 14
+d 15
+dh 16
+dt 17
+dʒ 18
+eɪ 19
+f 20
+fɛ 21
+h 22
+i 23
+iə 24
+iː 25
+iːd 26
+j 27
+k 28
+l 29
+m 30
+n 31
+nʲ 32
+oʊ 33
+oː 34
+oːɹ 35
+p 36
+r 37
+s 38
+t 39
+tw 40
+tʃ 41
+uː 42
+uːj 43
+v 44
+w 45
+x 46
+z 47
+æ 48
+ç 49
+ð 50
+ŋ 51
+ɐ 52
+ɑ 53
+ɑː 54
+ɑːɹ 55
+ɔ 56
+ɔɪ 57
+ɔː 58
+ɔːɹ 59
+ɔːɹt 60
+ə 61
+əl 62
+ən 63
+ɚ 64
+ɛ 65
+ɛɹ 66
+ɜː 67
+ɡ 68
+ɡʲ 69
+ɪ 70
+ɪɹ 71
+ɬ 72
+ɹ 73
+ɾ 74
+ʃ 75
+ʃm 76
+ʊ 77
+ʊɹ 78
+ʌ 79
+ʒ 80
+ʔ 81
+̃ 82
+̩ 83
+θ 84
+ᵻ 85
+— 86
+aɪʊɹ 87
+… 88
+aɪʊ 89
+ɒ 90
+a 91
+əʊ 92
+ʊə 93
+eə 94
+ɫ 95
+» 96
+ɛː 97
+æː 98
+1 99
+ɪː 100
+( 101
+ko 102
+) 103
+u 104
+enus 105
+« 106
+es 107
+e 108
+ɲ 109
+ɣ 110
+β 111
+o 112
+tɕ 113
+ɯ 114
+ʁ 115
+iːː 116
+aʊə 117
+ta 118
+ɭ 119
+q 120
+ɐː 121
+ph 122
+kh 123
+dʑ 124
+aː 125
+¡ 126
+¿ 127
+hy 128
+ʰχ 129
+fr 130
+ø 131
+hi 132
+ʈ 133
+ʉ 134
+ʲ 135
+ɖ 136
+oːː 137