Prepared some files for an upcoming update.

app.py

@@ -150,7 +150,9 @@ def tts_fn(
 
 def load_voicedata():
     print("Loading voice data...")
-    voices = []
+    #voices = []
+    envoices = []
+    jpvoices = []
     styledict = {}
     with open("voicelist.json", "r", encoding="utf-8") as f:
         voc_info = json.load(f)
@@ -168,8 +170,11 @@ def load_voicedata():
            s2id = hps.data.style2id
            styledict[model_path] = s2id.keys()
         print(f"Indexed voice {voice_name}")
-        voices.append((name, model_path, voice_name, speakerid, datasetauthor, image))
-    return voices, styledict
+        if(info['primarylang']=="JP"):
+            jpvoices.append((name, model_path, voice_name, speakerid, datasetauthor, image))
+        else:
+            envoices.append((name, model_path, voice_name, speakerid, datasetauthor, image))
+    return [envoices, jpvoices], styledict
         
 
 initial_text = "Hello there! This is test audio of a new Hololive text to speech tool."
@@ -224,6 +229,7 @@ if __name__ == "__main__":
     model_holder = ModelHolder(model_dir, device)
 
     languages = ["EN", "JP", "ZH"]
+    langnames = ["English", "Japanese"]
 
     model_names = model_holder.model_names
     if len(model_names) == 0:
@@ -287,63 +293,66 @@ if __name__ == "__main__":
         )
         use_tone = gr.Checkbox(label="Use accent adjustment", value=False, visible=False)
 
-        for (name, model_path, voice_name, speakerid, datasetauthor, image) in voicedata:
-            with gr.TabItem(name):
-                mn = gr.Textbox(value=model_path, visible=False, interactive=False)
-                mp = gr.Textbox(value=f"model_assets/{model_path}/{model_path}.safetensors", visible=False, interactive=False)
-                spk = gr.Textbox(value=speakerid, visible=False, interactive=False)
-                with gr.Row():
-                    with gr.Column():
-                        gr.Markdown(f"**{voice_name}**\n\nModel name: {model_path} | Dataset author: {datasetauthor}")
-                        gr.Image(f"images/{image}", label=None, show_label=False, width=300, show_download_button=False, container=False, show_share_button=False)
-                    with gr.Column():
-                        with gr.TabItem("Style using a preset"):
-                            style = gr.Dropdown(
-                                label="Current style (Neutral is an average style)",
-                                choices=styledict[model_path],
-                                value="Neutral",
-                            )
-                        with gr.TabItem("Style using existing audio"):
-                            ref_audio_path = gr.Audio(label="Reference Audio", type="filepath")
-                        style_weight = gr.Slider(
-                            minimum=0,
-                            maximum=50,
-                            value=5,
-                            step=0.1,
-                            label="Style strength",
-                        )
-                    with gr.Column():
-                        tts_button = gr.Button(
-                            "Synthesize", variant="primary", interactive=True
-                        )
-                        text_output = gr.Textbox(label="Info")
-                        audio_output = gr.Audio(label="Result")
-
-                        tts_button.click(
-                            tts_fn,
-                            inputs=[
-                                mn,
-                                mp,
-                                text_input,
-                                language,
-                                ref_audio_path,
-                                sdp_ratio,
-                                noise_scale,
-                                noise_scale_w,
-                                length_scale,
-                                line_split,
-                                split_interval,
-                                style_text,
-                                style_text_weight,
-                                use_style_text,
-                                style,
-                                style_weight,
-                                tone,
-                                use_tone,
-                                spk,
-                            ],
-                            outputs=[text_output, audio_output, tone],
-                        )
+        #for (name, model_path, voice_name, speakerid, datasetauthor, image) in voicedata:
+        for vi in range(len(voicedata)):
+            with gr.TabItem(langnames[vi]):
+                for (name, model_path, voice_name, speakerid, datasetauthor, image) in voicedata[vi]:
+                    with gr.TabItem(name):
+                        mn = gr.Textbox(value=model_path, visible=False, interactive=False)
+                        mp = gr.Textbox(value=f"model_assets/{model_path}/{model_path}.safetensors", visible=False, interactive=False)
+                        spk = gr.Textbox(value=speakerid, visible=False, interactive=False)
+                        with gr.Row():
+                            with gr.Column():
+                                gr.Markdown(f"**{voice_name}**\n\nModel name: {model_path} | Dataset author: {datasetauthor}")
+                                gr.Image(f"images/{image}", label=None, show_label=False, width=300, show_download_button=False, container=False, show_share_button=False)
+                            with gr.Column():
+                                with gr.TabItem("Style using a preset"):
+                                    style = gr.Dropdown(
+                                        label="Current style (Neutral is an average style)",
+                                        choices=styledict[model_path],
+                                        value="Neutral",
+                                    )
+                                with gr.TabItem("Style using existing audio"):
+                                    ref_audio_path = gr.Audio(label="Reference Audio", type="filepath")
+                                style_weight = gr.Slider(
+                                    minimum=0,
+                                    maximum=50,
+                                    value=5,
+                                    step=0.1,
+                                    label="Style strength",
+                                )
+                            with gr.Column():
+                                tts_button = gr.Button(
+                                    "Synthesize", variant="primary", interactive=True
+                                )
+                                text_output = gr.Textbox(label="Info")
+                                audio_output = gr.Audio(label="Result")
+
+                                tts_button.click(
+                                    tts_fn,
+                                    inputs=[
+                                        mn,
+                                        mp,
+                                        text_input,
+                                        language,
+                                        ref_audio_path,
+                                        sdp_ratio,
+                                        noise_scale,
+                                        noise_scale_w,
+                                        length_scale,
+                                        line_split,
+                                        split_interval,
+                                        style_text,
+                                        style_text_weight,
+                                        use_style_text,
+                                        style,
+                                        style_weight,
+                                        tone,
+                                        use_tone,
+                                        spk,
+                                    ],
+                                    outputs=[text_output, audio_output, tone],
+                                )
 
         with gr.Row():
             with gr.Column():

voicelist.json

@@ -5,6 +5,7 @@
     "title": "Mori Calliope",
 	"speakerid": "MoriCalliope",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "calli.png"
   },
   "Kiara": {
@@ -13,6 +14,7 @@
     "title": "Takanashi Kiara",
 	"speakerid": "TakanashiKiara",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "kiara.png"
   },
   "Ina": {
@@ -21,6 +23,7 @@
     "title": "Ninomae Ina'nis",
 	"speakerid": "NinomaeInanis",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "ina.png"
   },
   "Gura": {
@@ -29,6 +32,7 @@
     "title": "Gawr Gura",
 	"speakerid": "GawrGura",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "gura.png"
   },
   "Amelia": {
@@ -37,6 +41,7 @@
     "title": "Amelia Watson",
 	"speakerid": "AmeliaWatson",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "ame.png"
   },
   "IRyS": {
@@ -45,6 +50,7 @@
     "title": "IRyS",
 	"speakerid": "IRyS",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "irys.png"
   },
   "Sana": {
@@ -53,6 +59,7 @@
     "title": "Tsukumo Sana",
 	"speakerid": "TsukumoSana",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "sana.png"
   },
   "Fauna": {
@@ -61,6 +68,7 @@
     "title": "Ceres Fauna",
 	"speakerid": "CeresFauna",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "fauna.png"
   },
   "Kronii": {
@@ -69,6 +77,7 @@
     "title": "Ouro Kronii",
 	"speakerid": "OuroKronii",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "kronii.png"
   },
   "Mumei": {
@@ -77,6 +86,7 @@
     "title": "Nanashi Mumei",
 	"speakerid": "NanashiMumei",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "mumei.png"
   },
   "Baelz": {
@@ -85,6 +95,7 @@
     "title": "Hakos Baelz",
 	"speakerid": "HakosBaelz",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "bae.png"
   },
   "Shiori": {
@@ -93,6 +104,7 @@
     "title": "Shiori Novella",
 	"speakerid": "ShioriNovella",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "shiori.png"
   },
   "Bijou": {
@@ -101,6 +113,7 @@
     "title": "Koseki Bijou",
 	"speakerid": "KosekiBijou",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "bijou.png"
   },
   "Nerissa": {
@@ -109,6 +122,7 @@
     "title": "Nerissa Ravencroft",
 	"speakerid": "NerissaRavencroft",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "nerissa.png"
   },
   "Iofi": {
@@ -117,6 +131,7 @@
     "title": "Airani Iofifteen",
 	"speakerid": "AiraniIofifteen",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "iofi.png"
   },
   "Ollie": {
@@ -125,6 +140,7 @@
     "title": "Kureiji Ollie",
 	"speakerid": "KureijiOllie",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "ollie.png"
   },
   "Anya": {
@@ -133,6 +149,7 @@
     "title": "Anya Melfissa",
 	"speakerid": "AnyaMelfissa",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "anya.png"
   },
   "Zeta": {
@@ -141,14 +158,61 @@
     "title": "Vestia Zeta",
 	"speakerid": "VestiaZeta",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "EN",
     "cover": "zeta.png"
   },
+  "Sora": {
+    "enable": false,
+    "model_path": "SBV2_HoloJPTest2",
+    "title": "Tokino Sora",
+	"speakerid": "TokinoSora",
+	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
+    "cover": "sora.png"
+  },
+  "Suisei": {
+    "enable": false,
+    "model_path": "SBV2_HoloJPTest2",
+    "title": "Hoshimachi Suisei",
+	"speakerid": "HoshimachiSuisei",
+	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
+    "cover": "suisei.png"
+  },
+  "Matsuri": {
+    "enable": false,
+    "model_path": "SBV2_HoloJPTest2",
+    "title": "Natsuiro Matsuri",
+	"speakerid": "NatsuiroMatsuri",
+	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
+    "cover": "matsuri.png"
+  },
+  "Aqua": {
+    "enable": false,
+    "model_path": "SBV2_HoloJPTest2",
+    "title": "Minato Aqua",
+	"speakerid": "MinatoAqua",
+	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
+    "cover": "aqua.png"
+  },
+  "Ayame": {
+    "enable": false,
+    "model_path": "SBV2_HoloJPTest2",
+    "title": "Nakiri Ayame",
+	"speakerid": "NakiriAyame",
+	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
+    "cover": "ayame.png"
+  },
   "Okayu": {
     "enable": true,
     "model_path": "SBV2_HoloJPTest",
     "title": "Nekomata Okayu",
 	"speakerid": "NekomataOkayu",
 	"datasetauthor": "dacoolkid44",
+	"primarylang": "JP",
     "cover": "okayu.png"
   },
   "Flare": {
@@ -157,6 +221,7 @@
     "title": "Shiranui Flare",
 	"speakerid": "ShiranuiFlare",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
     "cover": "flare.png"
   },
   "Noel": {
@@ -165,6 +230,7 @@
     "title": "Shirogane Noel",
 	"speakerid": "ShiroganeNoel",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
     "cover": "noel.png"
   },
   "Marine": {
@@ -173,22 +239,52 @@
     "title": "Houshou Marine",
 	"speakerid": "HoushouMarine",
 	"datasetauthor": "dacoolkid44",
+	"primarylang": "JP",
     "cover": "marine.png"
   },
+  "Lamy": {
+    "enable": false,
+    "model_path": "SBV2_HoloJPTest2",
+    "title": "Yukihana Lamy",
+	"speakerid": "YukihanaLamy",
+	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
+    "cover": "lamy.png"
+  },
   "Laplus": {
     "enable": true,
     "model_path": "SBV2_HoloJPTest",
     "title": "La+ Darknesss",
 	"speakerid": "LaplusDarknesss",
 	"datasetauthor": "dacoolkid44",
+	"primarylang": "JP",
     "cover": "laplus.png"
   },
+  "Koyori": {
+    "enable": false,
+    "model_path": "SBV2_HoloJPTest2",
+    "title": "Hakui Koyori",
+	"speakerid": "HakuiKoyori",
+	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
+    "cover": "koyori.png"
+  },
+  "Chloe": {
+    "enable": false,
+    "model_path": "SBV2_HoloJPTest2",
+    "title": "Sakamata Chloe",
+	"speakerid": "SakamataChloe",
+	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
+    "cover": "chloe.png"
+  },
   "Ririka": {
     "enable": true,
     "model_path": "SBV2_HoloJPTest",
     "title": "Ichijou Ririka",
 	"speakerid": "IchijouRirika",
 	"datasetauthor": "Kit Lemonfoot",
+	"primarylang": "JP",
     "cover": "ririka.png"
   }
 }