train_xtts_pt.py

import os

from trainer import Trainer, TrainerArgs

from TTS.config.shared_configs import BaseDatasetConfig
from TTS.tts.datasets import load_tts_samples
from TTS.tts.layers.xtts.trainer.gpt_trainer import GPTArgs, GPTTrainer, GPTTrainerConfig, XttsAudioConfig
from TTS.utils.manage import ModelManager

# Logging parameters
RUN_NAME = "GPT_XTTS_v2.0_pt"
PROJECT_NAME = "XTTS_trainer"
DASHBOARD_LOGGER = "tensorboard"
LOGGER_URI = None

# Set here the path that the checkpoints will be saved. Default: ./run/training/
OUT_PATH = os.path.join(os.path.dirname(os.path.abspath(__file__)), "checkpoints_xtts")

# Training Parameters
OPTIMIZER_WD_ONLY_ON_WEIGHTS = True  # for multi-gpu training please make it False
START_WITH_EVAL = False  # if True it will star with evaluation
BATCH_SIZE = 23  # set here the batch size
GRAD_ACUMM_STEPS = 84  # set here the grad accumulation steps
# Note: we recommend that BATCH_SIZE * GRAD_ACUMM_STEPS need to be at least 252 for more efficient training. You can increase/decrease BATCH_SIZE but then set GRAD_ACUMM_STEPS accordingly.

# Define here the dataset that you want to use for the fine-tuning on.

config_dataset_brspeech = BaseDatasetConfig(
    formatter="brspeech", 
    path="/root/DATASETS/BRSpeech_CML_TTS_v14012024_24khz/",
    meta_file_train="metadata.csv",
    language="pt",
)
# Add here the configs of the datasets
DATASETS_CONFIG_LIST = [
	config_dataset_brspeech,
]
# Define the path where XTTS v2.0.1 files will be downloaded
CHECKPOINTS_OUT_PATH = os.path.join(OUT_PATH, "XTTS_v2.0_original_model_files/")
os.makedirs(CHECKPOINTS_OUT_PATH, exist_ok=True)


# DVAE files
DVAE_CHECKPOINT_LINK = "https://coqui.gateway.scarf.sh/hf-coqui/XTTS-v2/main/dvae.pth"
MEL_NORM_LINK = "https://coqui.gateway.scarf.sh/hf-coqui/XTTS-v2/main/mel_stats.pth"

# Set the path to the downloaded files
DVAE_CHECKPOINT = os.path.join(CHECKPOINTS_OUT_PATH, os.path.basename(DVAE_CHECKPOINT_LINK))
MEL_NORM_FILE = os.path.join(CHECKPOINTS_OUT_PATH, os.path.basename(MEL_NORM_LINK))

# download DVAE files if needed
if not os.path.isfile(DVAE_CHECKPOINT) or not os.path.isfile(MEL_NORM_FILE):
    print(" > Downloading DVAE files!")
    ModelManager._download_model_files([MEL_NORM_LINK, DVAE_CHECKPOINT_LINK], CHECKPOINTS_OUT_PATH, progress_bar=True)


# Download XTTS v2.0 checkpoint if needed
TOKENIZER_FILE_LINK = "https://coqui.gateway.scarf.sh/hf-coqui/XTTS-v2/main/vocab.json"
XTTS_CHECKPOINT_LINK = "https://coqui.gateway.scarf.sh/hf-coqui/XTTS-v2/main/model.pth"

# XTTS transfer learning parameters: You we need to provide the paths of XTTS model checkpoint that you want to do the fine tuning.
TOKENIZER_FILE = os.path.join(CHECKPOINTS_OUT_PATH, os.path.basename(TOKENIZER_FILE_LINK))  # vocab.json file
#XTTS_CHECKPOINT = os.path.join(CHECKPOINTS_OUT_PATH, os.path.basename(XTTS_CHECKPOINT_LINK))  # model.pth file
#XTTS_CHECKPOINT = "/root/TTS-XTTS-decoder/checkpoints_xtts/GPT_XTTS_v2.0_pt-May-13-2024_12+37AM-3fef64e9/checkpoint_217163.pth"
XTTS_CHECKPOINT ="/root/TTS-XTTS-decoder/checkpoints_xtts/GPT_XTTS_v2.0_pt-May-17-2024_01+34AM-3fef64e9/checkpoint_39030.pth"
# download XTTS v2.0 files if needed
'''
if not os.path.isfile(TOKENIZER_FILE) or not os.path.isfile(XTTS_CHECKPOINT):
    print(" > Downloading XTTS v2.0 files!")
    ModelManager._download_model_files(
        [TOKENIZER_FILE_LINK, XTTS_CHECKPOINT_LINK], CHECKPOINTS_OUT_PATH, progress_bar=True
    )
'''

# Training sentences generations
SPEAKER_REFERENCE = [
    "/root/DATASETS/BRSpeech_CML_TTS_v14012024_24khz/train/audio/12249/12765/12249_12765_000007-0003.wav"  # speaker reference to be used in training test sentences
]
LANGUAGE = config_dataset_brspeech.language


def main():
    # init args and config
    model_args = GPTArgs(
        max_conditioning_length=132300,  # 6 secs
        min_conditioning_length=66150,  # 3 secs
        debug_loading_failures=False,
        max_wav_length=255995,  # ~11.6 seconds
        max_text_length=200,
        mel_norm_file=MEL_NORM_FILE,
        dvae_checkpoint=DVAE_CHECKPOINT,
        xtts_checkpoint=XTTS_CHECKPOINT,  # checkpoint path of the model that you want to fine-tune
        tokenizer_file=TOKENIZER_FILE,
        gpt_num_audio_tokens=1026,
        gpt_start_audio_token=1024,
        gpt_stop_audio_token=1025,
        gpt_use_masking_gt_prompt_approach=True,
        gpt_use_perceiver_resampler=True,
    )
    # define audio config
    audio_config = XttsAudioConfig(sample_rate=24000, dvae_sample_rate=24000, output_sample_rate=24000)
    # training parameters config
    config = GPTTrainerConfig(
        output_path=OUT_PATH,
        model_args=model_args,
        run_name=RUN_NAME,
        project_name=PROJECT_NAME,
        run_description="""
            GPT XTTS training
            """,
        dashboard_logger=DASHBOARD_LOGGER,
        logger_uri=LOGGER_URI,
        audio=audio_config,
        batch_size=BATCH_SIZE,
        batch_group_size=48,
        eval_batch_size=BATCH_SIZE,
        num_loader_workers=8,
        eval_split_max_size=256,
        eval_split_size=0.05,
        print_step=50,
        plot_step=100,
        log_model_step=1000,
        save_step=10000,
        save_n_checkpoints=3,
        save_checkpoints=True,
        # target_loss="loss",
        print_eval=True,
        run_eval_steps=10000,
        # Optimizer values like tortoise, pytorch implementation with modifications to not apply WD to non-weight parameters.
        optimizer="AdamW",
        optimizer_wd_only_on_weights=OPTIMIZER_WD_ONLY_ON_WEIGHTS,
        optimizer_params={"betas": [0.9, 0.96], "eps": 1e-8, "weight_decay": 1e-2},
        lr=5e-06,  # learning rate
        lr_scheduler="MultiStepLR",
        # it was adjusted accordly for the new step scheme
        lr_scheduler_params={"milestones": [50000 * 18, 150000 * 18, 300000 * 18], "gamma": 0.5, "last_epoch": -1},
        test_sentences=[
            {
                "text": "Ouviram do ipiranga às margens plácidas de um povo heróico o brado retumbante.",
                "speaker_wav": SPEAKER_REFERENCE,
                "language": LANGUAGE,
            },
            {
                "text": "Minha terra tem palmeiras onde canta o sabiá, as aves que aqui gorjeiam não gorjeiam como lá.",
                "speaker_wav": SPEAKER_REFERENCE,
                "language": LANGUAGE,
            },
            {
                "text": "Ó que saudades que tenho da aurora da minha vida, da minha infância querida, Que os anos não trazem mais.",
                "speaker_wav": SPEAKER_REFERENCE,
                "language": LANGUAGE,
            },
            {
                "text": "No princípio Deus criou o céu e a terra, entretanto a terra era sem forma e vazia.",
                "speaker_wav": SPEAKER_REFERENCE,
                "language": LANGUAGE,
            },
            {
                "text": "Amor é fogo que arde sem se ver é ferida que dói e não se sente é um contentamento descontente é dor que desatina sem doer.",
                "speaker_wav": SPEAKER_REFERENCE,
                "language": LANGUAGE,
            },
            {
                "text": "E agora José? A festa acabou, a luz apagou, o povo sumiu, a noite esfriou, e agora José?",
                "speaker_wav": SPEAKER_REFERENCE,
                "language": LANGUAGE,
            },
            {
                "text": "Vou-me embora pra Pasárgada, Lá sou amigo do rei, Lá tenho a mulher que eu quero, Na cama que escolherei!",
                "speaker_wav": SPEAKER_REFERENCE,
                "language": LANGUAGE,
            },
            {
                "text": "É pau, é pedra, é o fim do caminho. É um resto de toco, é um pouco sozinho.",
                "speaker_wav": SPEAKER_REFERENCE,
                "language": LANGUAGE,
            },
            {
                "text": "No meio do caminho tinha uma pedra; Tinha uma pedra no meio do caminho.",
                "speaker_wav": SPEAKER_REFERENCE,
                "language": LANGUAGE,
            },
            {
                "text": "Brasil, mostra tua cara; quero ver quem paga pra gente ficar assim.",
                "speaker_wav": SPEAKER_REFERENCE,
                "language": LANGUAGE,
            }
        ],
    )

    # init the model from config
    model = GPTTrainer.init_from_config(config)

    # load training samples
    train_samples, eval_samples = load_tts_samples(
        DATASETS_CONFIG_LIST,
        eval_split=True,
        eval_split_max_size=config.eval_split_max_size,
        eval_split_size=config.eval_split_size,
    )

    # init the trainer and 🚀
    trainer = Trainer(
        TrainerArgs(
            restore_path=None,  # xtts checkpoint is restored via xtts_checkpoint key so no need of restore it using Trainer restore_path parameter
            skip_train_epoch=False,
            start_with_eval=START_WITH_EVAL,
            grad_accum_steps=GRAD_ACUMM_STEPS,
        ),
        config,
        output_path=OUT_PATH,
        model=model,
        train_samples=train_samples,
        eval_samples=eval_samples,
    )
    trainer.fit()


if __name__ == "__main__":
    main()