Thorsten-Voice/Youtube/train_vits_win.py
Thorsten Müller f13bcaf63e
Added Windows TTS training recipe
Added modified vits recipe for Thorsten-Voice model training using Windows
2023-03-05 16:19:50 +01:00

95 lines
3.0 KiB
Python

import os
from trainer import Trainer, TrainerArgs
from TTS.tts.configs.shared_configs import BaseDatasetConfig
from TTS.tts.configs.vits_config import VitsConfig
from TTS.tts.datasets import load_tts_samples
from TTS.tts.models.vits import Vits, VitsAudioConfig
from TTS.tts.utils.text.tokenizer import TTSTokenizer
from TTS.utils.audio import AudioProcessor
def main():
output_path = os.path.dirname(os.path.abspath(__file__))
#output_path = "c:\\temp\tts"
dataset_config = BaseDatasetConfig(
formatter="ljspeech", meta_file_train="metadata_small.csv", path="C:\\Users\\ThorstenVoice\\TTS-Training\\ThorstenVoice-Dataset_2022.10"
)
audio_config = VitsAudioConfig(
sample_rate=22050, win_length=1024, hop_length=256, num_mels=80, mel_fmin=0, mel_fmax=None
)
config = VitsConfig(
audio=audio_config,
run_name="vits_thorsten-voice",
batch_size=4,
eval_batch_size=4,
batch_group_size=5,
num_loader_workers=1,
num_eval_loader_workers=1,
run_eval=True,
test_delay_epochs=-1,
epochs=1000,
text_cleaner="phoneme_cleaners",
use_phonemes=True,
phoneme_language="de",
phoneme_cache_path=os.path.join(output_path, "phoneme_cache"),
compute_input_seq_cache=True,
print_step=25,
print_eval=True,
mixed_precision=False,
output_path=output_path,
datasets=[dataset_config],
cudnn_benchmark=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.",
],
)
# 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.
# config is updated with the default characters if not defined in 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 = Vits(config, ap, tokenizer, speaker_manager=None)
# init the trainer and 🚀
trainer = Trainer(
TrainerArgs(),
config,
output_path,
model=model,
train_samples=train_samples,
eval_samples=eval_samples,
)
trainer.fit()
print("Fertig!")
from multiprocessing import Process, freeze_support
if __name__ == '__main__':
freeze_support() # needed for Windows
main()