app.py

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

# ⛳ License agreement auto accept
os.environ["COQUI_TOS_AGREED"] = "1"

# 🎤 Load model
tts = TTS(model_name="tts_models/multilingual/multi-dataset/xtts_v2", progress_bar=False, gpu=False)

# 📢 Voice clone + TTS function
def clone_and_speak(audio, text, speed):
    if audio is None or text.strip() == "":
        return "Please upload a voice sample and enter some text."
    output_path = "output.wav"
    tts.tts_to_file(text=text, speaker_wav=audio, file_path=output_path, speed=speed)
    return output_path

# 🎨 UI Design
with gr.Blocks() as demo:
    gr.Markdown("<h1 style='text-align: center; color: #4CAF50;'>🎙️ Voice Cloning Tool</h1>")
    with gr.Row():
        with gr.Column():
            audio_input = gr.Audio(type="filepath", label="Upload Your Voice Sample")
            text_input = gr.Textbox(label="Enter text to speak in your voice")
            speed_input = gr.Slider(minimum=0.5, maximum=1.5, value=1.0, step=0.1, label="Speed")
            submit_button = gr.Button("Clone Voice")
        with gr.Column():
            audio_output = gr.Audio(label="Cloned Voice Output")

    submit_button.click(fn=clone_and_speak, inputs=[audio_input, text_input, speed_input], outputs=audio_output)

demo.launch()