app.py

import gradio as gr
from rvc import *

initial_md = """
![banner that says mikutts](https://huggingface.co/spaces/NoCrypt/mikuTTS/resolve/main/imgs/banner_mikutts.webp)
"""

app = gr.Blocks(theme='NoCrypt/miku')
with app:
    gr.Markdown(initial_md)
    with gr.Row():
        with gr.Column():
            model_name = gr.Dropdown(label="Model",choices=models,value=models[0])
            f0_key_up = gr.Number(label="Tune",value=6)
        with gr.Column():
            f0_method = gr.Radio(label="Pitch extraction method (pm: very fast, low quality, rmvpe: a little slow, high quality)",choices=["pm", "rmvpe"], value="rmvpe",visible=False)       
            index_rate = gr.Slider(minimum=0,maximum=1,label="Index rate",value=1,interactive=True)
            protect0 = gr.Slider(minimum=0,maximum=0.5,label="Protect",value=0.33,step=0.01,interactive=True)
                                                          
            
    with gr.Row():
        with gr.Column():
            tts_voice = gr.Dropdown(
                label="Edge-tts speaker (format: language-Country-Name-Gender), make sure the gender matches the model",
                choices=tts_voices,
                allow_custom_value=False,
                value="ja-JP-NanamiNeural-Female",
            )
            speed = gr.Slider(
                minimum=-100,
                maximum=100,
                label="Speech speed (%)",
                value=0,
                step=10,
                interactive=True,
            )
            tts_text = gr.Textbox(label="Input Text", value="こんにちは、私の名前は初音ミクです!")
        with gr.Column():
            but0 = gr.Button("Convert", variant="primary")
            info_text = gr.Textbox(label="Output info")
        with gr.Column():
            with gr.Accordion("Edge Voice", visible=False):
                edge_tts_output = gr.Audio(label="Edge Voice", type="filepath")
            tts_output = gr.Audio(label="Result")
        but0.click(
            tts,
            [
                model_name,
                speed,
                tts_text,
                tts_voice,
                f0_key_up,
                f0_method,
                index_rate,
                protect0,
            ],
            [info_text, edge_tts_output, tts_output],
        )
    with gr.Row():
        examples = gr.Examples(
            examples_per_page=100,
            examples=[
                ["こんにちは、私の名前は初音ミクです!", "ja-JP-NanamiNeural-Female", 6],
                ["Hello there. My name is Hatsune Miku!","en-CA-ClaraNeural-Female", 6],
                ["Halo. Nama saya Hatsune Miku!","id-ID-GadisNeural-Female", 4],
                ["Halo. Jenengku Hatsune Miku!","jv-ID-SitiNeural-Female", 10],
            ],
            inputs=[tts_text, tts_voice, f0_key_up],
        )

app.launch()