hifigan.py

import os

from TTS.config.shared_configs import BaseAudioConfig
from TTS.trainer import Trainer, TrainingArgs
from TTS.utils.audio import AudioProcessor
from TTS.vocoder.configs import HifiganConfig
from TTS.vocoder.datasets.preprocess import load_wav_data
from TTS.vocoder.models.gan import GAN

output_path = "/content/drive/MyDrive/Emergent/train/hifigan"

audio_config = BaseAudioConfig(
    sample_rate=22050,
    do_trim_silence=False,
    trim_db=60.0,
    signal_norm=False,
    mel_fmin=0.0,
    mel_fmax=8000,
    spec_gain=1.0,
    log_func="np.log",
    ref_level_db=20,
    preemphasis=0.98,
)

config = HifiganConfig(
    audio=audio_config,
    batch_size=32,
    eval_batch_size=16,
    num_loader_workers=8,
    num_eval_loader_workers=4,
    run_eval=True,
    test_delay_epochs=10,
    epochs=10000,
    wd= 0.0,
    seq_len=8192,
    pad_short=2000,
    use_noise_augment=True,
    eval_split_size=10,
    print_step=25,
    print_eval=False,
    mixed_precision=False,
    lr_gen=1e-5,
    lr_disc=1e-5,
    feat_match_loss_weight=10,
    target_loss="avg_G_loss",
    use_cache=True,
    data_path="/content/drive/MyDrive/Emergent/recorded_dataset/dataset/newData/LJSpeech-1.1/wavs",
    output_path=output_path,
)

# init audio processor
ap = AudioProcessor(**config.audio.to_dict())

# load training samples
eval_samples, train_samples = load_wav_data(config.data_path, config.eval_split_size)

# init model
model = GAN(config)

# init the trainer and 🚀
trainer = Trainer(
    TrainingArgs(),
    config,
    output_path,
    model=model,
    train_samples=train_samples,
    eval_samples=eval_samples,
    training_assets={"audio_processor": ap},
)
trainer.fit()