Add model checkpoint

best_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cf932ed3bba5a553960e51053406660f48ffccea9e2d399a1e75f13a68eab2d
+size 1038567590

best_model_22458.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cf932ed3bba5a553960e51053406660f48ffccea9e2d399a1e75f13a68eab2d
+size 1038567590

checkpoint_20000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4910a666bdb5f46e370448673a491ee6acb681935ec9611c5c6090b152dd93b
+size 1038567590

checkpoint_25000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:111ecf522d83593fa3500550deaf8142c11160d5442c4a295e2c61c7c0df88cb
+size 1038567590

checkpoint_30000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c27b6cabed0d094cbb08edc20d36b6f331cf7ffc1703665cb7e96f37457df972
+size 1038567590

checkpoint_35000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e8b556152cb0ec34fd483306b402cf9ac1b581b638d41afe206fa2187da8e06
+size 1038567590

config.json

@@ -0,0 +1,256 @@
+{
+    "output_path": "/home/ubuntu/vits-train",
+    "logger_uri": null,
+    "run_name": "vits_nctb",
+    "project_name": null,
+    "run_description": "\ud83d\udc38Coqui trainer run.",
+    "print_step": 25,
+    "plot_step": 100,
+    "model_param_stats": false,
+    "wandb_entity": null,
+    "dashboard_logger": "tensorboard",
+    "save_on_interrupt": true,
+    "log_model_step": 5000,
+    "save_step": 5000,
+    "save_n_checkpoints": 5,
+    "save_checkpoints": true,
+    "save_all_best": false,
+    "save_best_after": 0,
+    "target_loss": null,
+    "print_eval": false,
+    "test_delay_epochs": -1,
+    "run_eval": true,
+    "run_eval_steps": null,
+    "distributed_backend": "nccl",
+    "distributed_url": "tcp://localhost:54321",
+    "mixed_precision": true,
+    "precision": "fp16",
+    "epochs": 1000,
+    "batch_size": 64,
+    "eval_batch_size": 8,
+    "grad_clip": [
+        1000,
+        1000
+    ],
+    "scheduler_after_epoch": true,
+    "lr": 0.001,
+    "optimizer": "AdamW",
+    "optimizer_params": {
+        "betas": [
+            0.8,
+            0.99
+        ],
+        "eps": 1e-09,
+        "weight_decay": 0.01
+    },
+    "lr_scheduler": null,
+    "lr_scheduler_params": {},
+    "use_grad_scaler": false,
+    "allow_tf32": false,
+    "cudnn_enable": true,
+    "cudnn_deterministic": false,
+    "cudnn_benchmark": true,
+    "training_seed": 54321,
+    "model": "vits",
+    "num_loader_workers": 4,
+    "num_eval_loader_workers": 4,
+    "use_noise_augment": false,
+    "audio": {
+        "fft_size": 1024,
+        "sample_rate": 16000,
+        "win_length": 1024,
+        "hop_length": 256,
+        "num_mels": 80,
+        "mel_fmin": 0,
+        "mel_fmax": null
+    },
+    "use_phonemes": true,
+    "phonemizer": "bn_phonemizer",
+    "phoneme_language": "bn",
+    "compute_input_seq_cache": true,
+    "text_cleaner": "phoneme_cleaners",
+    "enable_eos_bos_chars": false,
+    "test_sentences_file": "",
+    "phoneme_cache_path": "/home/ubuntu/vits-train/phoneme_cache",
+    "characters": {
+        "characters_class": "TTS.tts.utils.text.characters.IPAPhonemes",
+        "vocab_dict": null,
+        "pad": "<PAD>",
+        "eos": "<EOS>",
+        "bos": "<BOS>",
+        "blank": "<BLNK>",
+        "characters": "abcdefghijklmnopqrstuvwxyz0123456789+=/*\u221a\u09a4\u099f\u09eb\u09ad\u09bf\u0990\u098b\u0996\u098a\u09dc\u0987\u099c\u09ae\u098f\u09c7\u0998\u0999\u09b8\u09c0\u09dd\u09b9\u099e\u2018\u0988\u0995\u09a3\u09ec\u0981\u09d7\u09b6\u09a2\u09a0\u200c\u09e7\u09cd\u09e8\u09ee\u09a6\u09c3\u0994\u0997\u0993\u2014\u099b\u0989\u0982\u09ac\u09c8\u099d\u09be\u09af\u09ab\u200d\u099a\u09b0\u09b7\u0985\u09cc\u09ce\u09a5\u09a1\u09bc\u09ea\u09a7\u09e6\u09c1\u09c2\u09e9\u0986\u0983\u09aa\u09df\u2019'\u201d\u09a8\u09b2\u09cb_\u2026\u09f0",
+        "punctuations": "-\u2013:;!,|.?\u0965\u0964 \u201c",
+        "phonemes": null,
+        "is_unique": true,
+        "is_sorted": true
+    },
+    "add_blank": true,
+    "batch_group_size": 5,
+    "loss_masking": null,
+    "min_audio_len": 1,
+    "max_audio_len": Infinity,
+    "min_text_len": 1,
+    "max_text_len": 325,
+    "compute_f0": false,
+    "compute_energy": false,
+    "compute_linear_spec": true,
+    "precompute_num_workers": 0,
+    "start_by_longest": false,
+    "shuffle": false,
+    "drop_last": false,
+    "datasets": [
+        {
+            "formatter": "",
+            "dataset_name": "",
+            "path": "/home/ubuntu/nctb-cropped/",
+            "meta_file_train": "/home/ubuntu/nctb-cropped/metadata.txt",
+            "ignored_speakers": null,
+            "language": "bn",
+            "phonemizer": "",
+            "meta_file_val": "",
+            "meta_file_attn_mask": ""
+        }
+    ],
+    "test_sentences": [
+        "\u0986\u09ae\u09be\u09b0   \u09b8\u09cb\u09a8\u09be\u09b0 \u09ac\u09be\u0982\u09b2\u09be, \u0986\u09ae\u09bf \u09a4\u09cb\u09ae\u09be\u09df \u09ad\u09be\u09b2\u09cb\u09ac\u09be\u09b8\u09bf\u0964",
+        "\u099a\u09bf\u09b0\u09a6\u09bf\u09a8   \u09a4\u09cb\u09ae\u09be\u09b0 \u0986\u0995\u09be\u09b6, \u09a4\u09cb\u09ae\u09be\u09b0 \u09ac\u09be\u09a4\u09be\u09b8, \u0986\u09ae\u09be\u09b0 \u09aa\u09cd\u09b0\u09be\u09a3\u09c7 \u09ac\u09be\u099c\u09be\u09df \u09ac\u09be\u0981\u09b6\u09bf",
+        "\u0993 \u09ae\u09be,   \u09ab\u09be\u0997\u09c1\u09a8\u09c7 \u09a4\u09cb\u09b0 \u0986\u09ae\u09c7\u09b0 \u09ac\u09a8\u09c7 \u0998\u09cd\u09b0\u09be\u09a3\u09c7 \u09aa\u09be\u0997\u09b2 \u0995\u09b0\u09c7,\u09ae\u09b0\u09bf \u09b9\u09be\u09df, \u09b9\u09be\u09df \u09b0\u09c7\u0964"
+    ],
+    "eval_split_max_size": null,
+    "eval_split_size": 0.01,
+    "use_speaker_weighted_sampler": false,
+    "speaker_weighted_sampler_alpha": 1.0,
+    "use_language_weighted_sampler": false,
+    "language_weighted_sampler_alpha": 1.0,
+    "use_length_weighted_sampler": false,
+    "length_weighted_sampler_alpha": 1.0,
+    "model_args": {
+        "num_chars": 139,
+        "out_channels": 513,
+        "spec_segment_size": 32,
+        "hidden_channels": 192,
+        "hidden_channels_ffn_text_encoder": 768,
+        "num_heads_text_encoder": 2,
+        "num_layers_text_encoder": 6,
+        "kernel_size_text_encoder": 3,
+        "dropout_p_text_encoder": 0.1,
+        "dropout_p_duration_predictor": 0.5,
+        "kernel_size_posterior_encoder": 5,
+        "dilation_rate_posterior_encoder": 1,
+        "num_layers_posterior_encoder": 16,
+        "kernel_size_flow": 5,
+        "dilation_rate_flow": 1,
+        "num_layers_flow": 4,
+        "resblock_type_decoder": "1",
+        "resblock_kernel_sizes_decoder": [
+            3,
+            7,
+            11
+        ],
+        "resblock_dilation_sizes_decoder": [
+            [
+                1,
+                3,
+                5
+            ],
+            [
+                1,
+                3,
+                5
+            ],
+            [
+                1,
+                3,
+                5
+            ]
+        ],
+        "upsample_rates_decoder": [
+            8,
+            8,
+            2,
+            2
+        ],
+        "upsample_initial_channel_decoder": 512,
+        "upsample_kernel_sizes_decoder": [
+            16,
+            16,
+            4,
+            4
+        ],
+        "periods_multi_period_discriminator": [
+            2,
+            3,
+            5,
+            7,
+            11
+        ],
+        "use_sdp": true,
+        "noise_scale": 1.0,
+        "inference_noise_scale": 0.667,
+        "length_scale": 1,
+        "noise_scale_dp": 1.0,
+        "inference_noise_scale_dp": 1.0,
+        "max_inference_len": null,
+        "init_discriminator": true,
+        "use_spectral_norm_disriminator": false,
+        "use_speaker_embedding": true,
+        "num_speakers": 126,
+        "speakers_file": "/home/ubuntu/vits-train/vits_nctb-June-25-2024_02+44AM-0000000/speakers.pth",
+        "d_vector_file": null,
+        "speaker_embedding_channels": 256,
+        "use_d_vector_file": false,
+        "d_vector_dim": 0,
+        "detach_dp_input": true,
+        "use_language_embedding": false,
+        "embedded_language_dim": 4,
+        "num_languages": 0,
+        "language_ids_file": null,
+        "use_speaker_encoder_as_loss": false,
+        "speaker_encoder_config_path": "",
+        "speaker_encoder_model_path": "",
+        "condition_dp_on_speaker": true,
+        "freeze_encoder": false,
+        "freeze_DP": false,
+        "freeze_PE": false,
+        "freeze_flow_decoder": false,
+        "freeze_waveform_decoder": false,
+        "encoder_sample_rate": null,
+        "interpolate_z": true,
+        "reinit_DP": false,
+        "reinit_text_encoder": false
+    },
+    "lr_gen": 0.0002,
+    "lr_disc": 0.0002,
+    "lr_scheduler_gen": "ExponentialLR",
+    "lr_scheduler_gen_params": {
+        "gamma": 0.999875,
+        "last_epoch": -1
+    },
+    "lr_scheduler_disc": "ExponentialLR",
+    "lr_scheduler_disc_params": {
+        "gamma": 0.999875,
+        "last_epoch": -1
+    },
+    "kl_loss_alpha": 1.0,
+    "disc_loss_alpha": 1.0,
+    "gen_loss_alpha": 1.0,
+    "feat_loss_alpha": 1.0,
+    "mel_loss_alpha": 45.0,
+    "dur_loss_alpha": 1.0,
+    "speaker_encoder_loss_alpha": 1.0,
+    "return_wav": true,
+    "use_weighted_sampler": false,
+    "weighted_sampler_attrs": {},
+    "weighted_sampler_multipliers": {},
+    "r": 1,
+    "num_speakers": 0,
+    "use_speaker_embedding": true,
+    "speakers_file": "/home/ubuntu/vits-train/vits_nctb-June-25-2024_02+44AM-0000000/speakers.pth",
+    "speaker_embedding_channels": 256,
+    "language_ids_file": null,
+    "use_language_embedding": false,
+    "use_d_vector_file": false,
+    "d_vector_file": null,
+    "d_vector_dim": 0
+}

events.out.tfevents.1719283482.ip-172-31-47-220.253748.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce272de8ebef3c36b93a7e934e0c7ad00cb9ef9402d176d1ad8125e93d763697
+size 52008599

speakers.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a51f7805073ce561b112b7097cdb1fe1bc8382c92c34637f50a15a2e644f8a32
+size 2528

train_vits.py

@@ -0,0 +1,150 @@
+import os
+
+from trainer import Trainer, TrainerArgs
+
+from TTS.tts.configs.shared_configs import BaseDatasetConfig, CharactersConfig
+from TTS.tts.configs.vits_config import VitsConfig
+from TTS.tts.datasets import load_tts_samples
+from TTS.tts.models.vits import Vits, VitsArgs, VitsAudioConfig
+from TTS.tts.utils.speakers import SpeakerManager
+from TTS.tts.utils.text.tokenizer import TTSTokenizer
+from TTS.utils.audio import AudioProcessor
+
+output_path = os.path.dirname(os.path.abspath(__file__))
+# dataset_config = BaseDatasetConfig(
+#     formatter="vctk", meta_file_train="", language="en-us", path=os.path.join(output_path, "../VCTK/")
+# )
+
+CONTINUE_PATH=None
+RESTORE_PATH=None
+START_WITH_EVAL=True
+GRAD_ACUMM_STEPS=1
+
+meta_file = '/home/ubuntu/nctb-cropped/metadata.txt'
+root_path = '/home/ubuntu/nctb-cropped/'
+
+def formatter(root_path, meta_file, **kwargs):  # pylint: disable=unused-argument
+    """Normalizes the LJSpeech meta data file to TTS format
+    https://keithito.com/LJ-Speech-Dataset/"""
+    txt_file = meta_file
+    items = []
+    with open(txt_file, "r", encoding="utf-8") as ttf:
+        for line in ttf:
+            cols = line.split("|")
+            wav_file = os.path.join(root_path,'audio', cols[0])
+            speaker_name = cols[0].split('_')[-1].split('.')[0]
+            try:
+                text = cols[1]
+            except:
+                print("not found")
+
+            items.append({"text": text, "audio_file": wav_file, "speaker_name": speaker_name, "root_path": root_path})
+    return items
+
+
+dataset_config = BaseDatasetConfig(
+     meta_file_train=meta_file, path=os.path.join(root_path, ""), language="bn"
+)
+
+
+characters_config = CharactersConfig(
+    pad = '<PAD>',
+    eos = '<EOS>', #'<EOS>', #'।',
+    bos = '<BOS>',# None,
+    blank = '<BLNK>',
+    phonemes = None,
+    characters =  "abcdefghijklmnopqrstuvwxyz0123456789+=/*√তট৫ভিঐঋখঊড়ইজমএেঘঙসীঢ়হঞ‘ঈকণ৬ঁৗশঢঠ\u200c১্২৮দৃঔগও—ছউংবৈঝাযফ\u200dচরষঅৌৎথড়৪ধ০ুূ৩আঃপয়’'”নলো_…ৰ",
+    punctuations = "-–:;!,|.?॥। “",
+)
+
+
+audio_config = VitsAudioConfig(
+    sample_rate=16000, win_length=1024, hop_length=256, num_mels=80, mel_fmin=0, mel_fmax=None
+)
+
+vitsArgs = VitsArgs(
+    use_speaker_embedding=True,
+)
+
+config = VitsConfig(
+    model_args=vitsArgs,
+    audio=audio_config,
+    run_name="vits_nctb",
+    batch_size=64,
+    eval_batch_size=8,
+    batch_group_size=5,
+    num_loader_workers=4,
+    num_eval_loader_workers=4,
+    run_eval=True,
+    test_delay_epochs=-1,
+    epochs=1000,
+    # text_cleaner="english_cleaners",
+    text_cleaner='phoneme_cleaners',
+    use_phonemes=True,
+    phoneme_language="bn",
+    phoneme_cache_path=os.path.join(output_path, "phoneme_cache"),
+    compute_input_seq_cache=True,
+    print_step=25,
+    print_eval=False,
+    mixed_precision=True,
+    max_text_len=325,  # change this if you have a larger VRAM than 16GB
+    output_path=output_path,
+    datasets=[dataset_config],
+    characters=characters_config,
+    save_step=5000,
+    cudnn_benchmark=True,
+    test_sentences = [
+        'আমার   সোনার বাংলা, আমি তোমায় ভালোবাসি।',
+        'চিরদিন   তোমার আকাশ, তোমার বাতাস, আমার প্রাণে বাজায় বাঁশি', 
+        'ও মা,   ফাগুনে তোর আমের বনে ঘ্রাণে পাগল করে,মরি হায়, হায় রে।'
+    ]
+)
+
+# 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,
+    formatter=formatter,
+    eval_split=True,
+    eval_split_max_size=config.eval_split_max_size,
+    eval_split_size=config.eval_split_size,
+)
+
+# init speaker manager for multi-speaker training
+# it maps speaker-id to speaker-name in the model and data-loader
+speaker_manager = SpeakerManager()
+speaker_manager.set_ids_from_data(train_samples + eval_samples, parse_key="speaker_name")
+config.model_args.num_speakers = speaker_manager.num_speakers
+
+# init model
+model = Vits(config, ap, tokenizer, speaker_manager)
+
+# init the trainer and 🚀
+trainer = Trainer(
+    TrainerArgs(
+        continue_path=CONTINUE_PATH,
+        restore_path=RESTORE_PATH,
+        skip_train_epoch=False,
+        start_with_eval=START_WITH_EVAL,
+        grad_accum_steps=GRAD_ACUMM_STEPS,
+    ),
+    config,
+    output_path,
+    model=model,
+    train_samples=train_samples,
+    eval_samples=eval_samples,
+)
+trainer.fit()