initial commit

.gitignore

@@ -0,0 +1 @@
+__pycache__

README.md

@@ -1,8 +1,8 @@
 ---
-title: VietnameseVITS
-emoji: 📚
-colorFrom: red
-colorTo: red
+title: MyOwnTexttospeech
+emoji: 🐢
+colorFrom: pink
+colorTo: green
 sdk: gradio
 sdk_version: 4.4.1
 app_file: app.py

app.py

@@ -0,0 +1,31 @@
+import gradio as gr
+import numpy as np
+from utils import load_model, normalize_text
+
+
+vits_model = load_model()
+vits_model.tts('Alo')
+
+def text_to_speech(text):
+    text = normalize_text(text)
+    audio = vits_model.tts(text)
+
+    audio = np.array(audio)
+    return 16000,audio
+
+gr.Interface(
+    fn=text_to_speech,
+    inputs="text",
+    outputs="audio",
+
+    examples=[
+        "Trăm năm trong cõi người ta, chữ tài chữ mệnh khéo là ghét nhau.",
+        "Đoạn trường tân thanh, thường được biết đến với cái tên đơn giản là Truyện Kiều, là một truyện thơ của đại thi hào Nguyễn Du",
+        "Lục Vân Tiên quê ở huyện Đông Thành, khôi ngô tuấn tú, tài kiêm văn võ. Nghe tin triều đình mở khoa thi, Vân Tiên từ giã thầy xuống núi đua tài.",
+        "Lê Quý Đôn, tên thuở nhỏ là Lê Danh Phương, là vị quan thời Lê trung hưng, cũng là nhà thơ và được mệnh danh là nhà bác học lớn của Việt Nam trong thời phong kiến",
+        "Tất cả mọi người đều sinh ra có quyền bình đẳng. Tạo hóa cho họ những quyền không ai có thể xâm phạm được; trong những quyền ấy, có quyền được sống, quyền tự do và quyền mưu cầu hạnh phúc.",
+    ],
+    theme="default",
+).launch(debug=False)
+
+

requirements.txt

@@ -0,0 +1,5 @@
+gradio==4.4.1
+numpy==1.24.3
+regex==2023.10.3
+TTS==0.17.8
+

utils.py

@@ -0,0 +1,94 @@
+from TTS.api import TTS
+import unicodedata
+import regex
+
+
+
+
+num_re = regex.compile(r"([0-9.,]*[0-9])")
+digits = ["không", "một", "hai", "ba", "bốn", "năm", "sáu", "bảy", "tám", "chín"]
+
+def read_number(num: str) -> str:
+    if len(num) == 1:
+        return digits[int(num)]
+    elif len(num) == 2 and num.isdigit():
+        n = int(num)
+        end = digits[n % 10]
+        if n == 10:
+            return "mười"
+        if n % 10 == 5:
+            end = "lăm"
+        if n % 10 == 0:
+            return digits[n // 10] + " mươi"
+        elif n < 20:
+            return "mười " + end
+        else:
+            if n % 10 == 1:
+                end = "mốt"
+            return digits[n // 10] + " mươi " + end
+    elif len(num) == 3 and num.isdigit():
+        n = int(num)
+        if n % 100 == 0:
+            return digits[n // 100] + " trăm"
+        elif num[1] == "0":
+            return digits[n // 100] + " trăm lẻ " + digits[n % 100]
+        else:
+            return digits[n // 100] + " trăm " + read_number(num[1:])
+    elif len(num) >= 4 and len(num) <= 6 and num.isdigit():
+        n = int(num)
+        n1 = n // 1000
+        return read_number(str(n1)) + " ngàn " + read_number(num[-3:])
+    elif "," in num:
+        n1, n2 = num.split(",")
+        return read_number(n1) + " phẩy " + read_number(n2)
+    elif "." in num:
+        parts = num.split(".")
+        if len(parts) == 2:
+            if parts[1] == "000":
+                return read_number(parts[0]) + " ngàn"
+            elif parts[1].startswith("00"):
+                end = digits[int(parts[1][2:])]
+                return read_number(parts[0]) + " ngàn lẻ " + end
+            else:
+                return read_number(parts[0]) + " ngàn " + read_number(parts[1])
+        elif len(parts) == 3:
+            return (
+                read_number(parts[0])
+                + " triệu "
+                + read_number(parts[1])
+                + " ngàn "
+                + read_number(parts[2])
+            )
+    return num
+
+
+def load_model():
+    config_path = 'vits/vits_config.json'
+    checkpoint_path = 'vits/best_model_vits_22951.pth'
+
+    tts = TTS(model_name='my_tts',
+              model_path=checkpoint_path,
+              config_path=config_path)
+
+    return tts
+
+
+def normalize_text(text):
+    # lowercase
+    text = text.lower()
+    # unicode normalize
+    text = unicodedata.normalize("NFKC", text)
+    text = text.replace(".", "")
+    text = text.replace(",", "")
+    text = text.replace(";", "")
+    text = text.replace(":", "")
+    text = text.replace("!", "")
+    text = text.replace("?", "")
+    text = text.replace("(", "")
+
+    text = num_re.sub(r" \1 ", text)
+    words = text.split()
+    words = [read_number(w) if num_re.fullmatch(w) else w for w in words]
+    text = " ".join(words)
+
+    return text

vits/best_model_vits_22951.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0dfc9c3865ccf2359258c5c6e2145365e643ba3be0208b57efedb6bfac20e428
+size 997817797

vits/vits_config.json

@@ -0,0 +1,255 @@
+{
+    "output_path": "/kaggle/working/",
+    "logger_uri": null,
+    "run_name": "vits_viet",
+    "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": null,
+    "save_step": 10000,
+    "save_n_checkpoints": 5,
+    "save_checkpoints": true,
+    "save_all_best": false,
+    "save_best_after": 10000,
+    "target_loss": null,
+    "print_eval": true,
+    "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": 2000,
+    "batch_size": 32,
+    "eval_batch_size": 16,
+    "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": null,
+    "use_grad_scaler": false,
+    "allow_tf32": false,
+    "cudnn_enable": true,
+    "cudnn_deterministic": false,
+    "cudnn_benchmark": false,
+    "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": false,
+    "phonemizer": null,
+    "phoneme_language": null,
+    "compute_input_seq_cache": true,
+    "text_cleaner": null,
+    "enable_eos_bos_chars": false,
+    "test_sentences_file": "",
+    "phoneme_cache_path": null,
+    "characters": {
+        "characters_class": "TTS.tts.utils.text.characters.Graphemes",
+        "vocab_dict": null,
+        "pad": "<PAD>",
+        "eos": "<EOS>",
+        "bos": "<BOS>",
+        "blank": "<BLNK>",
+        "characters": "abcdefghijklmnopqrstuvwxyz\u00e0\u00e1\u00e2\u00e3\u00e8\u00e9\u00ea\u00ec\u00ed\u00f2\u00f3\u00f4\u00f5\u00f9\u00fa\u00fd\u0103\u0111\u0129\u0169\u01a1\u01b0\u1ea1\u1ea3\u1ea5\u1ea7\u1ea9\u1eab\u1ead\u1eaf\u1eb1\u1eb3\u1eb5\u1eb7\u1eb9\u1ebb\u1ebd\u1ebf\u1ec1\u1ec3\u1ec5\u1ec7\u1ec9\u1ecb\u1ecd\u1ecf\u1ed1\u1ed3\u1ed5\u1ed7\u1ed9\u1edb\u1edd\u1edf\u1ee1\u1ee3\u1ee5\u1ee7\u1ee9\u1eeb\u1eed\u1eef\u1ef1\u1ef3\u1ef5\u1ef7\u1ef9",
+        "punctuations": "!'(),-.:;? ",
+        "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": Infinity,
+    "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": "/kaggle/input/infore/inforedataset",
+            "meta_file_train": "metadata.tsv",
+            "ignored_speakers": null,
+            "language": "",
+            "phonemizer": "",
+            "meta_file_val": "",
+            "meta_file_attn_mask": ""
+        }
+    ],
+    "test_sentences": [
+        "xin ch\u00e0o, t\u1ea5t c\u1ea3 m\u1ecdi ng\u01b0\u1eddi"
+    ],
+    "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": 108,
+        "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": false,
+        "num_speakers": 0,
+        "speakers_file": null,
+        "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": 0.0001,
+    "lr_disc": 0.0001,
+    "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": null,
+    "weighted_sampler_multipliers": null,
+    "r": 1,
+    "num_speakers": 0,
+    "use_speaker_embedding": false,
+    "speakers_file": null,
+    "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,
+    "github_branch": "* master"
+}