Update app.py

app.py

@@ -117,6 +117,10 @@ if __name__ == "__main__":
     parser = argparse.ArgumentParser()
     parser.add_argument("--share", action="store_true", default=False, help="share gradio app")
     args = parser.parse_args()
+    categories = ["Princess Connect! Re:Dive"]
+    others = {
+        "Umamusume": "https://huggingface.co/spaces/FrankZxShen/vits-fast-finetuning-umamusume",
+    }
     for info in models_info:
         lang = info['languages']
         examples = info['examples']
@@ -154,85 +158,113 @@ if __name__ == "__main__":
         gr.Markdown("# TTS&Voice Conversion for Princess Connect! Re:Dive\n\n"
                     )
         with gr.Tabs():
-            with gr.Tab("TTS"):
-                for i, (description, speakers, lang, example, symbols, tts_fn, to_symbol_fn) in enumerate(
-                        models_tts):
-                    gr.Markdown(description)
-                    with gr.Row():
-                        with gr.Column():
-                            textbox = gr.TextArea(label="Text",
-                                                    placeholder="Type your sentence here ",
-                                                    value="新たなキャラを解放できるようになったようですね。", elem_id=f"tts-input")
-                            with gr.Accordion(label="Phoneme Input", open=False):
-                                temp_text_var = gr.Variable()
-                                symbol_input = gr.Checkbox(value=False, label="Symbol input")
-                                symbol_list = gr.Dataset(label="Symbol list", components=[textbox],
-                                                            samples=[[x] for x in symbols],
-                                                            elem_id=f"symbol-list")
-                                symbol_list_json = gr.Json(value=symbols, visible=False)
-                            symbol_input.change(to_symbol_fn,
-                                                [symbol_input, textbox, temp_text_var],
-                                                [textbox, temp_text_var])
-                            symbol_list.click(None, [symbol_list, symbol_list_json], textbox,
-                                                _js=f"""
-                            (i, symbols, text) => {{
-                                let root = document.querySelector("body > gradio-app");
-                                if (root.shadowRoot != null)
-                                    root = root.shadowRoot;
-                                let text_input = root.querySelector("#tts-input").querySelector("textarea");
-                                let startPos = text_input.selectionStart;
-                                let endPos = text_input.selectionEnd;
-                                let oldTxt = text_input.value;
-                                let result = oldTxt.substring(0, startPos) + symbols[i] + oldTxt.substring(endPos);
-                                text_input.value = result;
-                                let x = window.scrollX, y = window.scrollY;
-                                text_input.focus();
-                                text_input.selectionStart = startPos + symbols[i].length;
-                                text_input.selectionEnd = startPos + symbols[i].length;
-                                text_input.blur();
-                                window.scrollTo(x, y);
-                                text = text_input.value;
-                                return text;
-                            }}""")
-                            # select character
-                            char_dropdown = gr.Dropdown(choices=speakers, value=speakers[0], label='character')
-                            language_dropdown = gr.Dropdown(choices=lang, value=lang[0], label='language')
-                            ns = gr.Slider(label="noise_scale", minimum=0.1, maximum=1.0, step=0.1, value=0.6, interactive=True)
-                            nsw = gr.Slider(label="noise_scale_w", minimum=0.1, maximum=1.0, step=0.1, value=0.668, interactive=True)
-                            duration_slider = gr.Slider(minimum=0.1, maximum=5, value=1, step=0.1,
-                                                        label='速度 Speed')
-                        with gr.Column():
-                            text_output = gr.Textbox(label="Message")
-                            audio_output = gr.Audio(label="Output Audio", elem_id="tts-audio")
-                            btn = gr.Button("Generate!")
-                            btn.click(tts_fn,
-                                        inputs=[textbox, char_dropdown, language_dropdown, ns, nsw, duration_slider,
-                                                symbol_input],
-                                        outputs=[text_output, audio_output])
-                    gr.Examples(
-                        examples=example,
-                        inputs=[textbox, char_dropdown, language_dropdown,
-                                duration_slider, symbol_input],
-                        outputs=[text_output, audio_output],
-                        fn=tts_fn
+            for category in categories:
+                with gr.TabItem(category):
+                    with gr.Tab("TTS"):
+                        for i, (description, speakers, lang, example, symbols, tts_fn, to_symbol_fn) in enumerate(
+                                models_tts):
+                            gr.Markdown(description)
+                            with gr.Row():
+                                with gr.Column():
+                                    textbox = gr.TextArea(label="Text",
+                                                          placeholder="Type your sentence here ",
+                                                          value="よーし、私もがんばらないと！", elem_id=f"tts-input")
+                                    with gr.Accordion(label="Phoneme Input", open=False):
+                                        temp_text_var = gr.Variable()
+                                        symbol_input = gr.Checkbox(
+                                            value=False, label="Symbol input")
+                                        symbol_list = gr.Dataset(label="Symbol list", components=[textbox],
+                                                                 samples=[[x]
+                                                                          for x in symbols],
+                                                                 elem_id=f"symbol-list")
+                                        symbol_list_json = gr.Json(
+                                            value=symbols, visible=False)
+                                    symbol_input.change(to_symbol_fn,
+                                                        [symbol_input, textbox,
+                                                            temp_text_var],
+                                                        [textbox, temp_text_var])
+                                    symbol_list.click(None, [symbol_list, symbol_list_json], textbox,
+                                                      _js=f"""
+                                    (i, symbols, text) => {{
+                                        let root = document.querySelector("body > gradio-app");
+                                        if (root.shadowRoot != null)
+                                            root = root.shadowRoot;
+                                        let text_input = root.querySelector("#tts-input").querySelector("textarea");
+                                        let startPos = text_input.selectionStart;
+                                        let endPos = text_input.selectionEnd;
+                                        let oldTxt = text_input.value;
+                                        let result = oldTxt.substring(0, startPos) + symbols[i] + oldTxt.substring(endPos);
+                                        text_input.value = result;
+                                        let x = window.scrollX, y = window.scrollY;
+                                        text_input.focus();
+                                        text_input.selectionStart = startPos + symbols[i].length;
+                                        text_input.selectionEnd = startPos + symbols[i].length;
+                                        text_input.blur();
+                                        window.scrollTo(x, y);
+                                        text = text_input.value;
+                                        return text;
+                                    }}""")
+                                    # select character
+                                    char_dropdown = gr.Dropdown(
+                                        choices=speakers, value=speakers[0], label='character')
+                                    language_dropdown = gr.Dropdown(
+                                        choices=lang, value=lang[0], label='language')
+                                    ns = gr.Slider(
+                                        label="noise_scale", minimum=0.1, maximum=1.0, step=0.1, value=0.6, interactive=True)
+                                    nsw = gr.Slider(label="noise_scale_w", minimum=0.1,
+                                                    maximum=1.0, step=0.1, value=0.668, interactive=True)
+                                    duration_slider = gr.Slider(minimum=0.1, maximum=5, value=1, step=0.1,
+                                                                label='速度 Speed')
+                                with gr.Column():
+                                    text_output = gr.Textbox(label="Message")
+                                    audio_output = gr.Audio(
+                                        label="Output Audio", elem_id="tts-audio")
+                                    btn = gr.Button("Generate!")
+                                    btn.click(tts_fn,
+                                              inputs=[textbox, char_dropdown, language_dropdown, ns, nsw, duration_slider,
+                                                      symbol_input],
+                                              outputs=[text_output, audio_output])
+                            gr.Examples(
+                                examples=example,
+                                inputs=[textbox, char_dropdown, language_dropdown,
+                                        duration_slider, symbol_input],
+                                outputs=[text_output, audio_output],
+                                fn=tts_fn
+                            )
+                    with gr.Tab("Voice Conversion"):
+                        for i, (description, speakers, vc_fn) in enumerate(
+                                models_vc):
+                            gr.Markdown("""
+                                            录制或上传声音，并选择要转换的音色。
+                            """)
+                            with gr.Column():
+                                record_audio = gr.Audio(
+                                    label="record your voice", source="microphone")
+                                upload_audio = gr.Audio(
+                                    label="or upload audio here", source="upload")
+                                source_speaker = gr.Dropdown(
+                                    choices=speakers, value=speakers[0], label="source speaker")
+                                target_speaker = gr.Dropdown(
+                                    choices=speakers, value=speakers[0], label="target speaker")
+                            with gr.Column():
+                                message_box = gr.Textbox(label="Message")
+                                converted_audio = gr.Audio(
+                                    label='converted audio')
+                            btn = gr.Button("Convert!")
+                            btn.click(vc_fn, inputs=[source_speaker, target_speaker, record_audio, upload_audio],
+                                      outputs=[message_box, converted_audio])
+            for category, link in others.items():
+                with gr.TabItem(category):
+                    gr.Markdown(
+                        f'''
+                        <center>
+                          <h2>Click to Go</h2>
+                          <a href="{link}">
+                            <img src="https://huggingface.co/datasets/huggingface/badges/raw/main/open-in-hf-spaces-xl-dark.svg"
+                          </a>
+                        </center>
+                        '''
                     )
-            with gr.Tab("Voice Conversion"):
-                for i, (description, speakers, vc_fn) in enumerate(
-                        models_vc):
-                    gr.Markdown("""
-                                    录制或上传声音，并选择要转换的音色。
-                    """)
-                    with gr.Column():
-                        record_audio = gr.Audio(label="record your voice", source="microphone")
-                        upload_audio = gr.Audio(label="or upload audio here", source="upload")
-                        source_speaker = gr.Dropdown(choices=speakers, value=speakers[0], label="source speaker")
-                        target_speaker = gr.Dropdown(choices=speakers, value=speakers[0], label="target speaker")
-                    with gr.Column():
-                        message_box = gr.Textbox(label="Message")
-                        converted_audio = gr.Audio(label='converted audio')
-                    btn = gr.Button("Convert!")
-                    btn.click(vc_fn, inputs=[source_speaker, target_speaker, record_audio, upload_audio],
-                            outputs=[message_box, converted_audio])
 
     app.queue(concurrency_count=3).launch(show_api=False, share=args.share)