1st version of xtts-v2 with Mandarin accent

checkpoint.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36b1d450f803c532393d0aba924fb05f083fa3fbc545235b0ed17d8523d6cd68
+size 5607928213

config.json

@@ -0,0 +1,246 @@
+{
+    "output_path": "zh-tw-xtts-v2",
+    "logger_uri": null,
+    "run_name": "GPT_XTTS_v2.0_LJSpeech_FT",
+    "project_name": "XTTS_trainer",
+    "run_description": "\n            GPT XTTS training\n            ",
+    "print_step": 50,
+    "plot_step": 100,
+    "model_param_stats": false,
+    "wandb_entity": null,
+    "dashboard_logger": "tensorboard",
+    "save_on_interrupt": true,
+    "log_model_step": 100,
+    "save_step": 100,
+    "save_n_checkpoints": 1,
+    "save_checkpoints": true,
+    "save_all_best": true,
+    "save_best_after": 0,
+    "target_loss": null,
+    "print_eval": false,
+    "test_delay_epochs": 0,
+    "run_eval": true,
+    "run_eval_steps": null,
+    "distributed_backend": "nccl",
+    "distributed_url": "tcp://localhost:54321",
+    "mixed_precision": false,
+    "precision": "fp16",
+    "epochs": 200,
+    "batch_size": 3,
+    "eval_batch_size": 3,
+    "grad_clip": 0.0,
+    "scheduler_after_epoch": true,
+    "lr": 5e-06,
+    "optimizer": "AdamW",
+    "optimizer_params": {
+        "betas": [
+            0.9,
+            0.96
+        ],
+        "eps": 1e-08,
+        "weight_decay": 0.01
+    },
+    "lr_scheduler": "MultiStepLR",
+    "lr_scheduler_params": {
+        "milestones": [
+            900000,
+            2700000,
+            5400000
+        ],
+        "gamma": 0.5,
+        "last_epoch": -1
+    },
+    "use_grad_scaler": false,
+    "allow_tf32": false,
+    "cudnn_enable": true,
+    "cudnn_deterministic": false,
+    "cudnn_benchmark": false,
+    "training_seed": 1,
+    "model": "xtts",
+    "num_loader_workers": 8,
+    "num_eval_loader_workers": 0,
+    "use_noise_augment": false,
+    "audio": {
+        "sample_rate": 22050,
+        "output_sample_rate": 24000,
+        "dvae_sample_rate": 22050
+    },
+    "use_phonemes": false,
+    "phonemizer": null,
+    "phoneme_language": null,
+    "compute_input_seq_cache": false,
+    "text_cleaner": null,
+    "enable_eos_bos_chars": false,
+    "test_sentences_file": "",
+    "phoneme_cache_path": null,
+    "characters": null,
+    "add_blank": false,
+    "batch_group_size": 48,
+    "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": false,
+    "precompute_num_workers": 0,
+    "start_by_longest": false,
+    "shuffle": false,
+    "drop_last": false,
+    "datasets": [
+        {
+            "formatter": "",
+            "dataset_name": "",
+            "path": "",
+            "meta_file_train": "",
+            "ignored_speakers": null,
+            "language": "",
+            "phonemizer": "",
+            "meta_file_val": "",
+            "meta_file_attn_mask": ""
+        }
+    ],
+    "test_sentences": [
+        {
+            "text": "\u8fd8\u662fGOOGLE\u7684ERICSCHMIDT\u4f1a\u8f9e\u53bb\u82f9\u679c\u8463\u4e8b\u4e00\u804c\u5462?",
+            "speaker_wav": "../tts_audio/recipes/taiwanvoice/wav/taiwanvoice_1.wav",
+            "language": "zh-cn"
+        },
+        {
+            "text": "MLB\u662f\u7f8e\u56fd\u56db\u5927\u804c\u4e1a\u4f53\u80b2\u8054\u76df\u4e4b\u4e00\u3002",
+            "speaker_wav": "../tts_audio/recipes/taiwanvoice/wav/taiwanvoice_2.wav",
+            "language": "zh-cn"
+        },
+        {
+            "text": "\u8fd9\u9879\u6d3b\u52a8\u7531APC\u7b56\u5212\u3002",
+            "speaker_wav": "../tts_audio/recipes/taiwanvoice/wav/taiwanvoice_3.wav",
+            "language": "zh-cn"
+        },
+        {
+            "text": "ITS\u5229\u7528\u5148\u8fdb\u4e4b\u7535\u5b50\u3001\u901a\u4fe1\u3001\u7535\u8111\u3001\u63a7\u5236\u53ca\u611f\u6d4b\u7b49\u6280\u672f\u4e8e\u5404\u79cd\u8fd0\u8f93\u7cfb\u7edf\u3002",
+            "speaker_wav": "../tts_audio/recipes/taiwanvoice/wav/taiwanvoice_4.wav",
+            "language": "zh-cn"
+        },
+        {
+            "text": "NVIDIA\u5bf9GOOGLEANDROID\u5e94\u7528\u5230\u8fd9\u7c7b\u88c5\u7f6e\u6709\u5f88\u9ad8\u7684\u671f\u671b\u3002",
+            "speaker_wav": "../tts_audio/recipes/taiwanvoice/wav/taiwanvoice_5.wav",
+            "language": "zh-cn"
+        },
+        {
+            "text": "\u53f0\u5927\u533b\u9662\u91d1\u5c71\u5206\u9662",
+            "speaker_wav": "../tts_audio/recipes/common_voice/wav/common_voice13_6793.wav",
+            "language": "zh-cn"
+        },
+        {
+            "text": "\u53f0\u4e2d\u5e02\u7acb\u56fe\u4e66\u9986\u6c99\u9e7f\u6df1\u6ce2\u5206\u9986",
+            "speaker_wav": "../tts_audio/recipes/common_voice/wav/common_voice13_6794.wav",
+            "language": "zh-cn"
+        },
+        {
+            "text": "\u65b0\u5e73\u7acb\u4f53\u505c\u8f66\u573a",
+            "speaker_wav": "../tts_audio/recipes/common_voice/wav/common_voice13_6795.wav",
+            "language": "zh-cn"
+        },
+        {
+            "text": "\u516d\u987a\u591a\u529f\u80fd\u6d3b\u52a8\u4e2d\u5fc3",
+            "speaker_wav": "../tts_audio/recipes/common_voice/wav/common_voice13_6796.wav",
+            "language": "zh-cn"
+        },
+        {
+            "text": "\u4eba\u56e0\u5e7b\u60f3\u800c\u4f1f\u5927",
+            "speaker_wav": "../tts_audio/recipes/common_voice/wav/common_voice13_6797.wav",
+            "language": "zh-cn"
+        },
+        {
+            "text": "\u5982\u679c\u4f60\u4e5f\u6709\u5174\u8da3\uff0c\u6b22\u8fce\u4e00\u540c\u9009\u8d2d",
+            "speaker_wav": "../tts_audio/recipes/common_voice/wav/common_voice13_6798.wav",
+            "language": "zh-cn"
+        }
+    ],
+    "eval_split_max_size": 256,
+    "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": {
+        "gpt_batch_size": 1,
+        "enable_redaction": false,
+        "kv_cache": true,
+        "gpt_checkpoint": "",
+        "clvp_checkpoint": null,
+        "decoder_checkpoint": null,
+        "num_chars": 255,
+        "tokenizer_file": "zh-tw-xtts-v2/vocab.json",
+        "gpt_max_audio_tokens": 605,
+        "gpt_max_text_tokens": 402,
+        "gpt_max_prompt_tokens": 70,
+        "gpt_layers": 30,
+        "gpt_n_model_channels": 1024,
+        "gpt_n_heads": 16,
+        "gpt_number_text_tokens": 6681,
+        "gpt_start_text_token": 261,
+        "gpt_stop_text_token": 0,
+        "gpt_num_audio_tokens": 1026,
+        "gpt_start_audio_token": 1024,
+        "gpt_stop_audio_token": 1025,
+        "gpt_code_stride_len": 1024,
+        "gpt_use_masking_gt_prompt_approach": true,
+        "gpt_use_perceiver_resampler": true,
+        "input_sample_rate": 22050,
+        "output_sample_rate": 24000,
+        "output_hop_length": 256,
+        "decoder_input_dim": 1024,
+        "d_vector_dim": 512,
+        "cond_d_vector_in_each_upsampling_layer": true,
+        "duration_const": 102400,
+        "min_conditioning_length": 66150,
+        "max_conditioning_length": 132300,
+        "gpt_loss_text_ce_weight": 0.1,
+        "gpt_loss_mel_ce_weight": 0.9,
+        "debug_loading_failures": false,
+        "max_wav_length": 255995,
+        "max_text_length": 200,
+        "mel_norm_file": "zh-tw-xtts-v2/mel_stats.pth",
+        "dvae_checkpoint": "zh-tw-xtts-v2/dvae.pth",
+        "xtts_checkpoint": "zh-tw-xtts-v2/model.pth",
+        "vocoder": ""
+    },
+    "model_dir": null,
+    "languages": [
+        "en",
+        "es",
+        "fr",
+        "de",
+        "it",
+        "pt",
+        "pl",
+        "tr",
+        "ru",
+        "nl",
+        "cs",
+        "ar",
+        "zh-cn",
+        "hu",
+        "ko",
+        "ja",
+        "hi"
+    ],
+    "temperature": 0.85,
+    "length_penalty": 1.0,
+    "repetition_penalty": 2.0,
+    "top_k": 50,
+    "top_p": 0.85,
+    "num_gpt_outputs": 1,
+    "gpt_cond_len": 12,
+    "gpt_cond_chunk_len": 4,
+    "max_ref_len": 10,
+    "sound_norm_refs": false,
+    "optimizer_wd_only_on_weights": true,
+    "weighted_loss_attrs": {},
+    "weighted_loss_multipliers": {},
+    "github_branch": "* dev"
+}

dvae.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b29bc227d410d4991e0a8c09b858f77415013eeb9fba9650258e96095557d97a
+size 210514388

mel_stats.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f69422a8a8f344c4fca2f0c6b8d41d2151d6615b7321e48e6bb15ae949b119c
+size 1067

model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7ea20001c6a0a841c77e252d8409f6a74fb423e79b3206a0771ba5989776187
+size 1867929118