feat: あみたろ 6종 화자 모델 추가

あみたろの声素材工房 ITAコーパス読み上げ音声 기반 파인튜닝 모델 6종
- amitaro_normal, amitaro_runrun, amitaro_yofukashi, amitaro_punsuka, amitaro_sasayaki_a, amitaro_sasayaki_b
- PyTorch (safetensors) + ONNX (synthesizer + duration_predictor)
- License: CC-BY-SA-4.0
- Credit: あみたろの声素材工房 https://amitaro.net/

onnx/speakers/amitaro_normal/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_normal",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 21,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_normal/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_normal/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_normal": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

onnx/speakers/amitaro_normal/duration_predictor.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea72d3b807b7120a84cc533e04da80374352fa124adae13577ac36953b170e41
+size 31503027

onnx/speakers/amitaro_normal/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28f9059b9da6c20dd891a2f22df2bfe77311a51d337623bd88e701de2f942643
+size 1152

onnx/speakers/amitaro_normal/synthesizer.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b4840a47ff05b207e226bd936a143b5985ad24d0983be23a5b14705f6849965
+size 250644697

onnx/speakers/amitaro_punsuka/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_punsuka",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 11,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_punsuka/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_punsuka/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_punsuka": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

onnx/speakers/amitaro_punsuka/duration_predictor.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b6accd699eecf9b44c92a0fd497154aa5c6063e99937ee58d0c78d58f60b166
+size 31503027

onnx/speakers/amitaro_punsuka/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cdbc3e625231b57adb16c89499bd0f01fb74acac42abbd1c0aceb96346bfa68
+size 1152

onnx/speakers/amitaro_punsuka/synthesizer.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:241d74d32ea37a0c1cb3d37b0569563c9712d08514fe68136cd2f1a2743971bc
+size 250644697

onnx/speakers/amitaro_runrun/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_runrun",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 21,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_runrun/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_runrun/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_runrun": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

onnx/speakers/amitaro_runrun/duration_predictor.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97809075bf8aad4a482fe20bd2722a234d4a8ad916ad5e88fa70105d817430d5
+size 31503027

onnx/speakers/amitaro_runrun/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b59a7286e5ea0c549acbdb086aa52fd227b6e270ec20918bc8cc63b61cbf64b8
+size 1152

onnx/speakers/amitaro_runrun/synthesizer.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a183206be69b208edd3a229e884c85770304ca21ae8ee0b3b0e930700fd2ce44
+size 250644697

onnx/speakers/amitaro_sasayaki_a/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_sasayaki_a",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 100,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_sasayaki_a/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_sasayaki_a/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_sasayaki_a": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

onnx/speakers/amitaro_sasayaki_a/duration_predictor.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b1331719136678ce093301b88364ebfdca6cf70287cbb3b97e1e07d641cb29a
+size 31503027

onnx/speakers/amitaro_sasayaki_a/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab461895545a48c61e6d96ba8e9dc717d9e7edc06e42a549b934ed14f9c84c86
+size 1152

onnx/speakers/amitaro_sasayaki_a/synthesizer.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66b43fcf188b152cd9011d6c2d1ac3e16eb9e043589b28c7fe2bb729379afaa0
+size 250644697

onnx/speakers/amitaro_sasayaki_b/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_sasayaki_b",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 100,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_sasayaki_b/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_sasayaki_b/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_sasayaki_b": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

onnx/speakers/amitaro_sasayaki_b/duration_predictor.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7273327640ddb5345963ae5fea5a5517c65d7efa57c35db3811634f727180975
+size 31503027

onnx/speakers/amitaro_sasayaki_b/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fac4742f009abb2d8a3fb007319887ee664634c6eed7816a9b72b8c90bdcde3
+size 1152

onnx/speakers/amitaro_sasayaki_b/synthesizer.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:742e7ff7a39b7218a19ba3120c1281dd6eb8726b89c872ad3eeee1068006bd1b
+size 250644697

onnx/speakers/amitaro_yofukashi/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_yofukashi",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 21,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_yofukashi/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_yofukashi/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_yofukashi": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

onnx/speakers/amitaro_yofukashi/duration_predictor.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8899e93605fc44ce3e058882a3fd09110973416fa9510104a58a1d5c5a5777b1
+size 31503027

onnx/speakers/amitaro_yofukashi/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c249725ac0de81417da52cf0b3211bb7a14610f534df933f0d4a307fcaa659a0
+size 1152

onnx/speakers/amitaro_yofukashi/synthesizer.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:999faade6b925fb9008c8a97fd87e95e6e20266b4d142b106edc0f9fc4b18ba0
+size 250644697

pytorch/speakers/amitaro_normal/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_normal",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 21,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_normal/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_normal/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_normal": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

pytorch/speakers/amitaro_normal/hayakoe_amitaro_normal_e11_s2000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:243d24c78d418b1b83001b5eb067e87372c398c25d36f65a5e47043dbf3484c5
+size 251155732

pytorch/speakers/amitaro_normal/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28f9059b9da6c20dd891a2f22df2bfe77311a51d337623bd88e701de2f942643
+size 1152

pytorch/speakers/amitaro_punsuka/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_punsuka",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 11,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_punsuka/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_punsuka/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_punsuka": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

pytorch/speakers/amitaro_punsuka/hayakoe_amitaro_punsuka_e11_s2000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:caa8a3a78795cab1b9d4c7507afab5b4dcf7f594c08fa49e826b917588db3cbe
+size 251155732

pytorch/speakers/amitaro_punsuka/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cdbc3e625231b57adb16c89499bd0f01fb74acac42abbd1c0aceb96346bfa68
+size 1152

pytorch/speakers/amitaro_runrun/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_runrun",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 21,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_runrun/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_runrun/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_runrun": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

pytorch/speakers/amitaro_runrun/hayakoe_amitaro_runrun_e11_s2000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00ac68eff452b4c2a3d1301ed46de128aa2a1a36e401954d7bd1c08408408473
+size 251155732

pytorch/speakers/amitaro_runrun/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b59a7286e5ea0c549acbdb086aa52fd227b6e270ec20918bc8cc63b61cbf64b8
+size 1152

pytorch/speakers/amitaro_sasayaki_a/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_sasayaki_a",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 100,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_sasayaki_a/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_sasayaki_a/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_sasayaki_a": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

pytorch/speakers/amitaro_sasayaki_a/hayakoe_amitaro_sasayaki_a_e87_s4000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bdef1fb086b368918a67cbc098ccf07f12d21f5db56df697c74056ebe18e967
+size 251155732

pytorch/speakers/amitaro_sasayaki_a/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab461895545a48c61e6d96ba8e9dc717d9e7edc06e42a549b934ed14f9c84c86
+size 1152

pytorch/speakers/amitaro_sasayaki_b/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_sasayaki_b",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 100,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_sasayaki_b/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_sasayaki_b/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_sasayaki_b": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

pytorch/speakers/amitaro_sasayaki_b/hayakoe_amitaro_sasayaki_b_e98_s4500.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7b1212c0d916a57cc9f1b9b09c93576aabe1fd0fd5e3dc7be0fdfa114bfaffd
+size 251155732

pytorch/speakers/amitaro_sasayaki_b/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fac4742f009abb2d8a3fb007319887ee664634c6eed7816a9b72b8c90bdcde3
+size 1152

pytorch/speakers/amitaro_yofukashi/config.json

@@ -0,0 +1,118 @@
+{
+  "model_name": "hayakoe_amitaro_yofukashi",
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 500,
+    "seed": 42,
+    "epochs": 21,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 2,
+    "bf16_run": false,
+    "fp16_run": false,
+    "lr_decay": 0.99996,
+    "segment_size": 16384,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0,
+    "c_commit": 100,
+    "skip_optimizer": false,
+    "freeze_ZH_bert": false,
+    "freeze_JP_bert": false,
+    "freeze_EN_bert": false,
+    "freeze_emo": false,
+    "freeze_style": false,
+    "freeze_decoder": false
+  },
+  "data": {
+    "use_jp_extra": true,
+    "training_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_yofukashi/train.list",
+    "validation_files": "/home/lemon/claude-projects/vits-rnd/hayakoe/data/dataset/amitaro_yofukashi/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 44100,
+    "filter_length": 2048,
+    "hop_length": 512,
+    "win_length": 2048,
+    "n_mel_channels": 128,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 1,
+    "cleaned_text": true,
+    "spk2id": {
+      "amitaro_yofukashi": 0
+    },
+    "num_styles": 1,
+    "style2id": {
+      "Neutral": 0
+    }
+  },
+  "model": {
+    "use_spk_conditioned_encoder": true,
+    "use_noise_scaled_mas": true,
+    "use_mel_posterior_encoder": false,
+    "use_duration_discriminator": false,
+    "use_wavlm_discriminator": true,
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      8,
+      2,
+      2
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 512,
+    "slm": {
+      "model": "microsoft/wavlm-base-plus",
+      "sr": 16000,
+      "hidden": 768,
+      "nlayers": 13,
+      "initial_channel": 64
+    }
+  },
+  "version": "2.7.0-JP-Extra"
+}

pytorch/speakers/amitaro_yofukashi/hayakoe_amitaro_yofukashi_e11_s2000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:836dfc7745d9616a26afcad054b6140143dd2157b45cac0b71fa4f86cb10b425
+size 251155732

pytorch/speakers/amitaro_yofukashi/style_vectors.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c249725ac0de81417da52cf0b3211bb7a14610f534df933f0d4a307fcaa659a0
+size 1152