1. 유튜브에 있는 영상으로 데이터셋 만들기 : OK
cd D:\AI\vits-webui\EZVitsDataset
d:
conda activate dataset
del D:\AI\vits-webui\EZVitsDataset\output\*.*
del D:\AI\vits-webui\EZVitsDataset\download\*.*
del D:\AI\vits-webui\EZVitsDataset\filelist\*.*
main.py 내용 수정 : 다운받고 싶은 유튜브 주소나 로컬 주소
python main.py


결과
D:\AI\vits-webui\EZVitsDataset\filelist\
 + filelist.txt
 + train.txt
 + val.txt
 + config.json

D:\AI\vits-webui\EZVitsDataset\output\
 + U7wRzyG4ke0.0001.wav
 + U7wRzyG4ke0.0002.wav
 + ....(생략).....



2. 전처리 : OK

2-1. 요건 사전 설치
conda create --name vits python=3.8
conda activate vits
pip3 install torch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 --index-url https://download.pytorch.org/whl/cu117
pip install -r requirements.txt

cd monotonic_align
mkdir monotonic_align
python setup.py build_ext --inplace
cd ..

pip install numpy==1.22   # 이버전 설치하니까 됨.



(D:\AI\vits-webui\venv) D:\AI\vits>python preprocess.py --text_index 1 --filelists EZVitsDataset/filelist/train.txt EZVitsDataset/filelist/val.txt --text_cleaners "korean_cleaners"
START: EZVitsDataset/filelist/train.txt
100%|████████████████████████████████████████████████████████████████████████████████| 91/91 [00:00<00:00, 3033.12it/s]
START: EZVitsDataset/filelist/val.txt
100%|████████████████████████████████████████████████████████████████████████████████| 22/22 [00:00<00:00, 3668.10it/s]

결과
D:\AI\vits-webui\EZVitsDataset\filelist\
 + filelist.txt
 + train.txt
 + val.txt
 + train_cleaned.txt
 + val_cleaned.txt

3. config.json 내용 기입
{
  "train": {
    "log_interval": 100,
    "eval_interval": 100,
    "seed": 1234,
    "epochs": 100,
    "learning_rate": 0.0002,
    "betas": [
      0.8,
      0.99
    ],
    "eps": 1e-09,
    "batch_size": 16,
    "fp16_run": true,
    "lr_decay": 0.999875,
    "segment_size": 8192,
    "init_lr_ratio": 1,
    "warmup_epochs": 0,
    "c_mel": 45,
    "c_kl": 1.0
  },
  "data": {
    "training_files": "EZVitsDataset\\filelist\\train_cleaned.txt",
    "validation_files": "EZVitsDataset\\filelist\\val_cleaned.txt",
    "text_cleaners": [
      "korean_cleaners"
    ],
    "max_wav_value": 32768.0,
    "sampling_rate": 22050,
    "filter_length": 1024,
    "hop_length": 256,
    "win_length": 1024,
    "n_mel_channels": 80,
    "mel_fmin": 0.0,
    "mel_fmax": null,
    "add_blank": true,
    "n_speakers": 0,
    "cleaned_text": true
  },
  "model": {
    "inter_channels": 192,
    "hidden_channels": 192,
    "filter_channels": 768,
    "n_heads": 2,
    "n_layers": 6,
    "kernel_size": 3,
    "p_dropout": 0.1,
    "resblock": "1",
    "resblock_kernel_sizes": [
      3,
      7,
      11
    ],
    "resblock_dilation_sizes": [
      [
        1,
        3,
        5
      ],
      [
        1,
        3,
        5
      ],
      [
        1,
        3,
        5
      ]
    ],
    "upsample_rates": [
      8,
      8,
      2,
      2
    ],
    "upsample_initial_channel": 512,
    "upsample_kernel_sizes": [
      16,
      16,
      4,
      4
    ],
    "n_layers_q": 3,
    "use_spectral_norm": false,
    "gin_channels": 256
  },
  "speakers": [
    "0"
  ],
  "symbols": [
    "_",
    ",",
    ".",
    "!",
    "?",
    "\u2026",
    "~",
    "\u3131",
    "\u3134",
    "\u3137",
    "\u3139",
    "\u3141",
    "\u3142",
    "\u3145",
    "\u3147",
    "\u3148",
    "\u314a",
    "\u314b",
    "\u314c",
    "\u314d",
    "\u314e",
    "\u3132",
    "\u3138",
    "\u3143",
    "\u3146",
    "\u3149",
    "\u314f",
    "\u3153",
    "\u3157",
    "\u315c",
    "\u3161",
    "\u3163",
    "\u3150",
    "\u3154",
    " "
  ]
}


4. 학습
python train.py -c EZVitsDataset\filelist\config.json -m EZVitsDataset\filelist

(vits) D:\AI\vits>python train.py -c EZVitsDataset\filelist\config.json -m EZVitsDataset\filelist
INFO:filelist:{'train': {'log_interval': 100, 'eval_interval': 100, 'seed': 1234, 'epochs': 100, 'learning_rate': 0.0002, 'betas': [0.8, 0.99], 'eps': 1e-09, 'batch_size': 16, 'fp16_run': True, 'lr_decay': 0.999875, 'segment_size': 8192, 'init_lr_ratio': 1, 'warmup_epochs': 0, 'c_mel': 45, 'c_kl': 1.0}, 'data': {'training_files': 'EZVitsDataset\\filelist\\train_cleaned.txt', 'validation_files': 'EZVitsDataset\\filelist\\val_cleaned.txt', 'text_cleaners': ['korean_cleaners'], 'max_wav_value': 32768.0, 'sampling_rate': 22050, 'filter_length': 1024, 'hop_length': 256, 'win_length': 1024, 'n_mel_channels': 80, 'mel_fmin': 0.0, 'mel_fmax': None, 'add_blank': True, 'n_speakers': 0, 'cleaned_text': True}, 'model': {'inter_channels': 192, 'hidden_channels': 192, 'filter_channels': 768, 'n_heads': 2, 'n_layers': 6, 'kernel_size': 3, 'p_dropout': 0.1, 'resblock': '1', 'resblock_kernel_sizes': [3, 7, 11], 'resblock_dilation_sizes': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'upsample_rates': [8, 8, 2, 2], 'upsample_initial_channel': 512, 'upsample_kernel_sizes': [16, 16, 4, 4], 'n_layers_q': 3, 'use_spectral_norm': False, 'gin_channels': 256}, 'speakers': ['0'], 'symbols': ['_', ',', '.', '!', '?', '…', '~', 'ㄱ', 'ㄴ', 'ㄷ', 'ㄹ', 'ㅁ', 'ㅂ', 'ㅅ', 'ㅇ', 'ㅈ', 'ㅊ', 'ㅋ', 'ㅌ', 'ㅍ', 'ㅎ', 'ㄲ', 'ㄸ', 'ㅃ', 'ㅆ', 'ㅉ', 'ㅏ', 'ㅓ', 'ㅗ', 'ㅜ', 'ㅡ', 'ㅣ', 'ㅐ', 'ㅔ', ' '], 'model_dir': 'checkpoints\\EZVitsDataset\\filelist'}
INFO:torch.distributed.distributed_c10d:Added key: store_based_barrier_key:1 to store for rank: 0
INFO:torch.distributed.distributed_c10d:Rank 0: Completed store-based barrier for key:store_based_barrier_key:1 with 1 nodes.
  0%|                                                                                            | 0/6 [00:27<?, ?it/s]
Traceback (most recent call last):
  File "train.py", line 302, in <module>
    main()
  File "train.py", line 56, in main
    mp.spawn(run, nprocs=n_gpus, args=(n_gpus, hps,))
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\torch\multiprocessing\spawn.py", line 240, in spawn
    return start_processes(fn, args, nprocs, join, daemon, start_method='spawn')
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\torch\multiprocessing\spawn.py", line 198, in start_processes
    while not context.join():
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\torch\multiprocessing\spawn.py", line 160, in join
    raise ProcessRaisedException(msg, error_index, failed_process.pid)
torch.multiprocessing.spawn.ProcessRaisedException:

-- Process 0 terminated with the following error:
Traceback (most recent call last):
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\torch\multiprocessing\spawn.py", line 69, in _wrap
    fn(i, *args)
  File "D:\AI\vits\train.py", line 123, in run
    train_and_evaluate(rank, epoch, hps, [net_g, net_d], [optim_g, optim_d], [scheduler_g, scheduler_d], scaler, [train_loader, eval_loader], logger, [writer, writer_eval])
  File "D:\AI\vits\train.py", line 143, in train_and_evaluate
    for batch_idx, (x, x_lengths, spec, spec_lengths, y, y_lengths) in enumerate(tqdm(train_loader)):
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\tqdm\std.py", line 1181, in __iter__
    for obj in iterable:
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\torch\utils\data\dataloader.py", line 628, in __next__
    data = self._next_data()
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\torch\utils\data\dataloader.py", line 1333, in _next_data
    return self._process_data(data)
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\torch\utils\data\dataloader.py", line 1359, in _process_data
    data.reraise()
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\torch\_utils.py", line 543, in reraise
    raise exception
IndexError: Caught IndexError in DataLoader worker process 0.
Original Traceback (most recent call last):
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\torch\utils\data\_utils\worker.py", line 302, in _worker_loop
    data = fetcher.fetch(index)
  File "C:\Users\hwyoo\miniconda3\envs\vits\lib\site-packages\torch\utils\data\_utils\fetch.py", line 58, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File