import os import torch from trainer import Trainer, TrainerArgs from TTS.bin.compute_embeddings import compute_embeddings from TTS.bin.resample import resample_files from TTS.config.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 CharactersConfig, Vits, VitsArgs, VitsAudioConfig, VitsDataset from TTS.utils.downloaders import download_libri_tts from torch.utils.data import DataLoader from TTS.utils.samplers import PerfectBatchSampler torch.set_num_threads(24) # pylint: disable=W0105 """ This recipe replicates the first experiment proposed in the CML-TTS paper (https://arxiv.org/abs/2306.10097). It uses the YourTTS model. YourTTS model is based on the VITS model however it uses external speaker embeddings extracted from a pre-trained speaker encoder and has small architecture changes. """ CURRENT_PATH = os.path.dirname(os.path.abspath(__file__)) # Name of the run for the Trainer RUN_NAME = "YourTTS-Syntacc-PT_continue" # Path where you want to save the models outputs (configs, checkpoints and tensorboard logs) OUT_PATH = os.path.join(os.path.dirname(os.path.abspath(__file__)), "runs") # "/raid/coqui/Checkpoints/original-YourTTS/" # If you want to do transfer learning and speedup your training you can set here the path to the CML-TTS available checkpoint that cam be downloaded here: https://drive.google.com/u/2/uc?id=1yDCSJ1pFZQTHhL09GMbOrdjcPULApa0p RESTORE_PATH = "/raid/datasets/MUPE/Experiments/runs/YourTTS-Syntacc-PT-January-28-2024_09+50AM-5f5841de1/best_model_88444.pth" # Download the checkpoint here: https://drive.google.com/u/2/uc?id=1yDCSJ1pFZQTHhL09GMbOrdjcPULApa0p # This paramter is useful to debug, it skips the training epochs and just do the evaluation and produce the test sentences SKIP_TRAIN_EPOCH = False # Set here the batch size to be used in training and evaluation BATCH_SIZE = 26 # Training Sampling rate and the target sampling rate for resampling the downloaded dataset (Note: If you change this you might need to redownload the dataset !!) # Note: If you add new datasets, please make sure that the dataset sampling rate and this parameter are matching, otherwise resample your audios SAMPLE_RATE = 16000 DASHBOARD_LOGGER="tensorboard" LOGGER_URI = None DASHBOARD_LOGGER = "clearml" LOGGER_URI = "s3://coqui-ai-models/TTS/Checkpoints/YourTTS/MUPE/" # Max audio length in seconds to be used in training (every audio bigger than it will be ignored) MAX_AUDIO_LEN_IN_SECONDS = float("inf") # Define here the datasets config brpb_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brpb.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brpb" ) brba_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brba.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brba" ) brportugal_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brportugal.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brportugal" ) brsp_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brsp.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brsp" ) brpe_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brpe.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brpe" ) brmg_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brmg.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brmg" ) brrj_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brrj.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brrj" ) brce_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brce.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brce" ) brrs_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brrs.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brrs" ) bralemanha_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_bralemanha.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="bralemanha" ) brgo_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brgo.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brgo" ) bral_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_bral.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="bral" ) brpr_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brpr.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brpr" ) bres_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_bres.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="bres" ) brpi_train_config = BaseDatasetConfig( formatter="coqui", dataset_name="mupe", meta_file_train="metadata_coqui_brpi.csv", path="/raid/datasets/MUPE/dataset/mupe/", language="brpi" ) # bres_train_config, brpi_train_config no files found DATASETS_CONFIG_LIST = [brpb_train_config,brba_train_config,brportugal_train_config,brsp_train_config,brpe_train_config,brmg_train_config,brrj_train_config,brce_train_config,brrs_train_config,bralemanha_train_config,brgo_train_config,bral_train_config,brpr_train_config] ### Extract speaker embeddings SPEAKER_ENCODER_CHECKPOINT_PATH = ( "https://github.com/coqui-ai/TTS/releases/download/speaker_encoder_model/model_se.pth.tar" ) SPEAKER_ENCODER_CONFIG_PATH = "https://github.com/coqui-ai/TTS/releases/download/speaker_encoder_model/config_se.json" D_VECTOR_FILES = [] # List of speaker embeddings/d-vectors to be used during the training # Iterates all the dataset configs checking if the speakers embeddings are already computated, if not compute it for dataset_conf in DATASETS_CONFIG_LIST: # Check if the embeddings weren't already computed, if not compute it embeddings_file = os.path.join(dataset_conf.path, f"H_ASP_speaker_embeddings_{dataset_conf.language}.pth") if not os.path.isfile(embeddings_file): print(f">>> Computing the speaker embeddings for the {dataset_conf.dataset_name} dataset") compute_embeddings( SPEAKER_ENCODER_CHECKPOINT_PATH, SPEAKER_ENCODER_CONFIG_PATH, embeddings_file, old_speakers_file=None, config_dataset_path=None, formatter_name=dataset_conf.formatter, dataset_name=dataset_conf.dataset_name, dataset_path=dataset_conf.path, meta_file_train=dataset_conf.meta_file_train, meta_file_val=dataset_conf.meta_file_val, disable_cuda=False, no_eval=False, ) D_VECTOR_FILES.append(embeddings_file) # Audio config used in training. audio_config = VitsAudioConfig( sample_rate=SAMPLE_RATE, hop_length=256, win_length=1024, fft_size=1024, mel_fmin=0.0, mel_fmax=None, num_mels=80, ) # Init VITSArgs setting the arguments that are needed for the YourTTS model model_args = VitsArgs( inference_noise_scale=0.33, inference_noise_scale_dp=0.33, spec_segment_size=62, hidden_channels=192, hidden_channels_ffn_text_encoder=768, num_heads_text_encoder=2, num_layers_text_encoder=10, kernel_size_text_encoder=3, dropout_p_text_encoder=0.1, d_vector_file=D_VECTOR_FILES, use_d_vector_file=True, d_vector_dim=512, speaker_encoder_model_path=SPEAKER_ENCODER_CHECKPOINT_PATH, speaker_encoder_config_path=SPEAKER_ENCODER_CONFIG_PATH, resblock_type_decoder="2", # In the paper, we accidentally trained the YourTTS using ResNet blocks type 2, if you like you can use the ResNet blocks type 1 like the VITS model # Useful parameters to enable the Speaker Consistency Loss (SCL) described in the paper use_speaker_encoder_as_loss=False, # Useful parameters to enable multilingual training use_language_embedding=False, embedded_language_dim=4, use_adaptive_weight_text_encoder=True, use_perfect_class_batch_sampler=True, perfect_class_batch_sampler_key="language" ) # General training config, here you can change the batch size and others useful parameters config = VitsConfig( output_path=OUT_PATH, model_args=model_args, run_name=RUN_NAME, project_name="SYNTACC", run_description=""" - YourTTS with SYNTACC text encoder """, 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, print_step=50, plot_step=100, log_model_step=1000, save_step=5000, save_n_checkpoints=2, save_checkpoints=True, # target_loss="loss_1", print_eval=False, use_phonemes=False, phonemizer="espeak", phoneme_language="en", compute_input_seq_cache=True, add_blank=True, text_cleaner="multilingual_cleaners", characters=CharactersConfig( characters_class="TTS.tts.models.vits.VitsCharacters", pad="_", eos="&", bos="*", blank=None, characters="ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\u00a1\u00a3\u00b7\u00b8\u00c0\u00c1\u00c2\u00c3\u00c4\u00c5\u00c7\u00c8\u00c9\u00ca\u00cb\u00cc\u00cd\u00ce\u00cf\u00d1\u00d2\u00d3\u00d4\u00d5\u00d6\u00d9\u00da\u00db\u00dc\u00df\u00e0\u00e1\u00e2\u00e3\u00e4\u00e5\u00e7\u00e8\u00e9\u00ea\u00eb\u00ec\u00ed\u00ee\u00ef\u00f1\u00f2\u00f3\u00f4\u00f5\u00f6\u00f9\u00fa\u00fb\u00fc\u0101\u0104\u0105\u0106\u0107\u010b\u0119\u0141\u0142\u0143\u0144\u0152\u0153\u015a\u015b\u0161\u0178\u0179\u017a\u017b\u017c\u020e\u04e7\u05c2\u1b20", punctuations="\u2014!'(),-.:;?\u00bf ", phonemes="iy\u0268\u0289\u026fu\u026a\u028f\u028ae\u00f8\u0258\u0259\u0275\u0264o\u025b\u0153\u025c\u025e\u028c\u0254\u00e6\u0250a\u0276\u0251\u0252\u1d7b\u0298\u0253\u01c0\u0257\u01c3\u0284\u01c2\u0260\u01c1\u029bpbtd\u0288\u0256c\u025fk\u0261q\u0262\u0294\u0274\u014b\u0272\u0273n\u0271m\u0299r\u0280\u2c71\u027e\u027d\u0278\u03b2fv\u03b8\u00f0sz\u0283\u0292\u0282\u0290\u00e7\u029dx\u0263\u03c7\u0281\u0127\u0295h\u0266\u026c\u026e\u028b\u0279\u027bj\u0270l\u026d\u028e\u029f\u02c8\u02cc\u02d0\u02d1\u028dw\u0265\u029c\u02a2\u02a1\u0255\u0291\u027a\u0267\u025a\u02de\u026b'\u0303' ", is_unique=True, is_sorted=True, ), phoneme_cache_path=None, precompute_num_workers=12, start_by_longest=True, datasets=DATASETS_CONFIG_LIST, cudnn_benchmark=False, max_audio_len=SAMPLE_RATE * MAX_AUDIO_LEN_IN_SECONDS, mixed_precision=False, test_sentences=[ #GUSTAVO: apenas pessoas do treino ["Voc\u00ea ter\u00e1 a vista do topo da montanha que voc\u00ea escalar.", "EDILEINE_FONSECA", None, "brsp"], ["Quem semeia ventos, colhe tempestades.", "JOSE_PAULO_DE_ARAUJO", None, "brpb"], ["O olho do dono \u00e9 que engorda o gado.", "VITOR_RAFAEL_OLIVEIRA_ALVES", None, "brba"], ["\u00c1gua mole em pedra dura, tanto bate at\u00e9 que fura.", "MARIA_AURORA_FELIX", None, "brportugal"], ["Quem espera sempre alcan\u00e7a.", "ANTONIO_DE_AMORIM_COSTA", None, "brpe"], ["Cada macaco no seu galho.", "ALCIDES_DE_LIMA", None, "brmg"], ["Em terra de cego, quem tem um olho \u00e9 rei.", "ALUISIO_SOARES_DE_SOUSA", None, "brrj"], ["A ocasi\u00e3o faz o ladr\u00e3o.", "FRANCISCO_JOSE_MOREIRA_MOTA", None, "brce"], ["De gr\u00e3o em gr\u00e3o, a galinha enche o papo.", "EVALDO_ANDRADA_CORREA", None, "brrs"], ["Mais vale um p\u00c1ssaro na m\u00e3o do que dois voando.", "DORIS_ALEXANDER", None, "bralemanha"], ["Quem n\u00e3o arrisca, n\u00e3o petisca.", "DONALDO_LUIZ_DE_ALMEIDA", None, "brgo"], ["A uni\u00e3o faz a for\u00e7a.", "GERONCIO_HENRIQUE_NETO", None, "bral"], ["Em boca fechada n\u00e3o entra mosquito.", "MALU_NATEL_FREIRE_WEBER", None, "brpr"], # ["Quem n\u00e3o tem dinheiro, n\u00e3o tem v\u00edcios.", "INES_VIEIRA_BOGEA", None, "bres"], # ["Quando voc\u00ea n\u00e3o corre nenhum risco, voc\u00ea arrisca tudo.", "MARIA_ASSUNCAO_SOUSA", None, "brpi"] ], # Enable the weighted sampler use_weighted_sampler=True, # Ensures that all speakers are seen in the training batch equally no matter how many samples each speaker has # weighted_sampler_attrs={"language": 1.0, "speaker_name": 1.0}, weighted_sampler_attrs={"language": 1.0}, weighted_sampler_multipliers={ # "speaker_name": { # you can force the batching scheme to give a higher weight to a certain speaker and then this speaker will appears more frequently on the batch. # It will speedup the speaker adaptation process. Considering the CML train dataset and "new_speaker" as the speaker name of the speaker that you want to adapt. # The line above will make the balancer consider the "new_speaker" as 106 speakers so 1/4 of the number of speakers present on CML dataset. # 'new_speaker': 106, # (CML tot. train speaker)/4 = (424/4) = 106 # } }, # It defines the Speaker Consistency Loss (SCL) α to 9 like the YourTTS paper speaker_encoder_loss_alpha=9.0, ) # Load all the datasets samples and split traning and evaluation sets train_samples, eval_samples = load_tts_samples( config.datasets, eval_split=True, eval_split_max_size=config.eval_split_max_size, eval_split_size=config.eval_split_size, ) # Init the model model = Vits.init_from_config(config) # Init the trainer and 🚀 trainer = Trainer( TrainerArgs(restore_path=RESTORE_PATH, skip_train_epoch=SKIP_TRAIN_EPOCH, start_with_eval=True), config, output_path=OUT_PATH, model=model, train_samples=train_samples, eval_samples=eval_samples, ) trainer.fit()