TTS/recipes/thorsten_DE/align_tts/train_aligntts.py

import os

from trainer import Trainer, TrainerArgs

from TTS.tts.configs.align_tts_config import AlignTTSConfig
from TTS.tts.configs.shared_configs import BaseDatasetConfig
from TTS.tts.datasets import load_tts_samples
from TTS.tts.models.align_tts import AlignTTS
from TTS.tts.utils.text.tokenizer import TTSTokenizer
from TTS.utils.audio import AudioProcessor
from TTS.utils.downloaders import download_thorsten_de

output_path = os.path.dirname(os.path.abspath(__file__))

# init configs
dataset_config = BaseDatasetConfig(
    formatter="thorsten", meta_file_train="metadata.csv", path=os.path.join(output_path, "../thorsten-de/")
)

# download dataset if not already present
if not os.path.exists(dataset_config.path):
    print("Downloading dataset")
    download_thorsten_de(os.path.split(os.path.abspath(dataset_config.path))[0])

config = AlignTTSConfig(
    batch_size=32,
    eval_batch_size=16,
    num_loader_workers=4,
    num_eval_loader_workers=4,
    run_eval=True,
    test_delay_epochs=-1,
    epochs=1000,
    text_cleaner="phoneme_cleaners",
    use_phonemes=False,
    phoneme_language="de",
    phoneme_cache_path=os.path.join(output_path, "phoneme_cache"),
    print_step=25,
    print_eval=True,
    mixed_precision=False,
    test_sentences=[
        "Es hat mich viel Zeit gekostet ein Stimme zu entwickeln, jetzt wo ich sie habe werde ich nicht mehr schweigen.",
        "Sei eine Stimme, kein Echo.",
        "Es tut mir Leid David. Das kann ich leider nicht machen.",
        "Dieser Kuchen ist großartig. Er ist so lecker und feucht.",
        "Vor dem 22. November 1963.",
    ],
    output_path=output_path,
    datasets=[dataset_config],
)

# INITIALIZE THE AUDIO PROCESSOR
# Audio processor is used for feature extraction and audio I/O.
# It mainly serves to the dataloader and the training loggers.
ap = AudioProcessor.init_from_config(config)

# INITIALIZE THE TOKENIZER
# Tokenizer is used to convert text to sequences of token IDs.
# If characters are not defined in the config, default characters are passed to the config
tokenizer, config = TTSTokenizer.init_from_config(config)

# LOAD DATA SAMPLES
# Each sample is a list of ```[text, audio_file_path, speaker_name]```
# You can define your custom sample loader returning the list of samples.
# Or define your custom formatter and pass it to the `load_tts_samples`.
# Check `TTS.tts.datasets.load_tts_samples` for more details.
train_samples, eval_samples = load_tts_samples(
    dataset_config,
    eval_split=True,
    eval_split_max_size=config.eval_split_max_size,
    eval_split_size=config.eval_split_size,
)

# init model
model = AlignTTS(config, ap, tokenizer)

# INITIALIZE THE TRAINER
# Trainer provides a generic API to train all the 🐸TTS models with all its perks like mixed-precision training,
# distributed training, etc.
trainer = Trainer(
    TrainerArgs(), config, output_path, model=model, train_samples=train_samples, eval_samples=eval_samples
)

# AND... 3,2,1... 🚀
trainer.fit()