add kei

app.py

@@ -90,6 +90,13 @@ def change_lang(language):
     else:
         return 0.6, 0.668, 1
 
+def search_speaker(search_value):
+    for s in speakers:
+        if search_value == s:
+            return s
+    for s in speakers:
+        if search_value in s:
+            return s
 
 download_audio_js = """
 () =>{{
@@ -196,6 +203,10 @@ if __name__ == '__main__':
                                         if sid == -1 or sid < 0:
                                             speaker = gr.Dropdown(label="Speaker", choices=speakers,
                                                                   type="index", value=speakers[0])
+                                            with gr.Row():
+                                                search = gr.Textbox(label="Search Speaker", lines=1)
+                                                btn2 = gr.Button(value="Search")
+                                                btn2.click(search_speaker, inputs=[search], outputs=[speaker])
                                         else:
                                             speaker = gr.Dropdown(label="Speaker", choices=speakers,
                                                                   visible=False, type="index", value=speakers[sid])
@@ -272,6 +283,10 @@ if __name__ == '__main__':
                                         if sid == -1 or sid < 0:
                                             speaker = gr.Dropdown(label="Speaker", choices=speakers,
                                                                   type="index", value=speakers[0])
+                                            with gr.Row():
+                                                search = gr.Textbox(label="Search Speaker", lines=1)
+                                                btn2 = gr.Button(value="Search")
+                                                btn2.click(search_speaker, inputs=[search], outputs=[speaker])
                                         else:
                                             speaker = gr.Dropdown(label="Speaker", choices=speakers,
                                                                   visible=False, type="index", value=speakers[sid])

config.py

@@ -1,4 +1,4 @@
-categories = ["Blue Archive", "Lycoris Recoil", "GalGame"]
+categories = ["GalGame"]
 
 categories_all = ["Honkai: Star Rail", "Blue Archive", "Lycoris Recoil", "Princess Connect! Re:Dive",
                   "Genshin Impact", "Honkai Impact 3rd", "Overwatch 2", "GalGame", "vits uma genshin", "moe tts"]

config/info.json

@@ -506,6 +506,19 @@
     "config_path": "./pretrained_models/my-vits/natsume_ai/config.json",
     "model_path": "./pretrained_models/my-vits/natsume_ai/natsume_ai.pth"
   },
+  "kei": {
+    "enable": true,
+    "name_en": "Natsume Kei",
+    "name_zh": "夏目圭",
+    "title": "GalGame-夏目圭",
+    "cover": "./pretrained_models/my-vits/kei/cover.jpg",
+    "sid": 0,
+    "example": "こんにちは。",
+    "language": "Japanese",
+    "type": "multi",
+    "config_path": "./pretrained_models/my-vits/kei/config.json",
+    "model_path": "./pretrained_models/my-vits/kei/kei.pth"
+  },
   "genshin": {
     "enable": true,
     "name_en": "Genshin Impact",

cuda_test.py

@@ -2,6 +2,8 @@ import torch
 
 
 if __name__ == "__main__":
+    print(torch.cuda.is_available())
+
     device_count = torch.cuda.device_count()
 
     if device_count > 0:

docker-compose.yml

@@ -1,4 +1,4 @@
-version: '3'
+version: "3"
 services:
   vits:
     image: vits-moe:latest
@@ -13,14 +13,6 @@ services:
       - ./config.py:/app/config.py
       - ./cuda_test.py:/app/cuda_test.py
       - ./entrypoint.sh:/app/entrypoint.sh
-      - /opt/cuda/lib64:/usr/local/cuda/lib64
-      - /opt/cuda/include:/usr/local/cuda/include
-    command: |
-      /bin/sh -c "
-      if ! python -c 'import torch; print(torch.cuda.is_available())'; then
-        echo 'CUDA is not available'
-        exit 0
-      fi"
     deploy:
       resources:
         reservations:
@@ -28,5 +20,3 @@ services:
             - driver: nvidia
               count: 1
               capabilities: [gpu]
-
-

pretrained_models/my-vits/kei/config.json

@@ -0,0 +1,112 @@
+{
+  "train": {
+    "log_interval": 10,
+    "eval_interval": 100,
+    "seed": 1234,
+    "epochs": 10000,
+    "learning_rate": 0.0002,
+    "betas": [0.8, 0.99],
+    "eps": 1e-9,
+    "batch_size": 16,
+    "fp16_run": true,
+    "lr_decay": 0.999875,
+    "segment_size": 8192,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0
+  },
+  "data": {
+    "training_files": "final_annotation_train.txt",
+    "validation_files": "final_annotation_val.txt",
+    "text_cleaners": ["zh_ja_mixture_cleaners"],
+    "max_wav_value": 32768.0,
+    "sampling_rate": 22050,
+    "filter_length": 1024,
+    "hop_length": 256,
+    "win_length": 1024,
+    "n_mel_channels": 80,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 3,
+    "cleaned_text": true
+  },
+  "model": {
+    "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],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [16, 16, 4, 4],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 256
+  },
+  "speakers": ["kei", "specialweek", "zhongli"],
+  "symbols": [
+    "_",
+    ",",
+    ".",
+    "!",
+    "?",
+    "-",
+    "~",
+    "\u2026",
+    "A",
+    "E",
+    "I",
+    "N",
+    "O",
+    "Q",
+    "U",
+    "a",
+    "b",
+    "d",
+    "e",
+    "f",
+    "g",
+    "h",
+    "i",
+    "j",
+    "k",
+    "l",
+    "m",
+    "n",
+    "o",
+    "p",
+    "r",
+    "s",
+    "t",
+    "u",
+    "v",
+    "w",
+    "y",
+    "z",
+    "\u0283",
+    "\u02a7",
+    "\u02a6",
+    "\u026f",
+    "\u0279",
+    "\u0259",
+    "\u0265",
+    "\u207c",
+    "\u02b0",
+    "`",
+    "\u2192",
+    "\u2193",
+    "\u2191",
+    " "
+  ]
+}

pretrained_models/my-vits/kei/kei.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f67b35330570f5d69244aefec23ddcbf9af89fe55836d0c04a20f487dbac01a
+size 158899433