app.py

import gradio as gr
import edge_tts
import asyncio

# TTS synthesis function
async def synthesize_tts(text, voice, rate, pitch):
    tts = edge_tts.Communicate(text, voice, rate=f"{rate}%", pitch=f"{pitch}%")
    output_file = "output.mp3"
    await tts.save(output_file)
    return output_file

# Gradio-compatible function for TTS
def generate_tts(text, voice, rate, pitch):
    loop = asyncio.new_event_loop()
    asyncio.set_event_loop(loop)
    output_file = loop.run_until_complete(synthesize_tts(text, voice, rate, pitch))
    return output_file

# API-friendly Gradio Interface
interface = gr.Interface(
    fn=generate_tts,
    inputs=[
        gr.Textbox(label="Enter Text"),
        gr.Dropdown(choices=["en-US-GuyNeural", "en-US-JennyNeural"], label="Voice"),  # Add more voices
        gr.Slider(-50, 50, value=0, label="Rate (%)"),
        gr.Slider(-50, 50, value=0, label="Pitch (%)"),
    ],
    outputs=gr.Audio(label="Generated Audio"),
    allow_flagging="never",  # Disables user feedback
    live=True  # Supports real-time processing
)

if __name__ == "__main__":
    interface.launch(server_name="0.0.0.0", server_port=7860)