tabs/tts/ui.py

import gradio as gr

from config import Config
from .events import *


def audio_tab():
    with gr.Row():
        with gr.Column():
            with gr.Group():
                with gr.Group():
                    text = gr.Textbox(lines=5, label="Enter text")
                    mode = gr.Radio(["SFT", "VC", "VC-CrossLingual", "Instruct"], label="Mode", value="SFT",) # automate with speech recognition pipeline
                    sft_speaker = gr.Radio(sft_speakers, label="Select speaker")
                with gr.Accordion('Voice Clone', open=False):
                    speaker_audio = gr.Audio(label="Upload Audio", type='numpy', visible=False)
                    clear_speaker_audio = gr.Button(value="Clear Audio")
                
                with gr.Accordion('Instruct', open=False):
                    voice_instructions = gr.Textbox(lines=5, label="Enter voice instructions", visible=False)
        
        with gr.Column():
            output_audio = gr.Audio(label="Output Audio", interactive=False, show_download_button=True)
            clear_output_audio = gr.Button(value="Clear Audio")
            generate_audio = gr.Button(value="Generate Audio")
    
    
    # Events
    # Clear Audio
    clear_speaker_audio.click(clear_audio, speaker_audio, speaker_audio)
    clear_output_audio.click(clear_audio, output_audio, output_audio)
    
    # Mode
    mode.change(
        update_mode,
        [mode, sft_speaker, speaker_audio, voice_instructions],
        [sft_speaker, speaker_audio, voice_instructions]
    )
    
    # Generate Audio
    generate_audio.click(
        gen_audio,
        [text, mode, sft_speaker, speaker_audio, voice_instructions],
        [output_audio]
    )