app.py

import os
import gradio as gr
from TTS.api import TTS

# Set the environment variable to agree to the terms of service
os.environ["COQUI_TOS_AGREED"] = "1"

# Initialize the TTS model
tts = TTS("tts_models/multilingual/multi-dataset/xtts_v2")

# Get available voices and languages
available_voices = tts.list_speakers()
available_languages = tts.list_languages()

def generate_speech(text, voice, language):
    # Generate speech
    output_path = "output.wav"
    tts.tts_to_file(text=text, speaker_wav=voice, language=language, file_path=output_path)
    return output_path

# Create Gradio interface
with gr.Blocks() as demo:
    gr.Markdown("# XTTS v2 Text-to-Speech Generator")

    with gr.Row():
        text_input = gr.Textbox(label="Input Text", placeholder="Enter text to be synthesized")

    with gr.Row():
        voice_dropdown = gr.Dropdown(label="Select Voice", choices=available_voices)
        language_dropdown = gr.Dropdown(label="Select Language", choices=available_languages)

    with gr.Row():
        generate_button = gr.Button("Generate Speech")

    with gr.Row():
        audio_output = gr.Audio(label="Generated Speech")

    generate_button.click(generate_speech, inputs=[text_input, voice_dropdown, language_dropdown], outputs=audio_output)

demo.launch(debug=True)