app.py

import gradio as gr
from gtts import gTTS
from pydub import AudioSegment
from pydub.playback import play
from io import BytesIO

# Step 1: Define a function to generate and merge TTS audio for multiple languages
def multilingual_tts(korean_text, english_text, russian_text, bulgarian_text):
    # Language mapping
    texts = {
        "ko": korean_text,
        "en": english_text,
        "ru": russian_text,
        "bg": bulgarian_text,
    }
    
    combined_audio = AudioSegment.silent(duration=0)  # Empty audio to start
    
    for lang, text in texts.items():
        if text.strip():  # Process only if text is provided
            tts = gTTS(text, lang=lang)
            audio_file = BytesIO()
            tts.write_to_fp(audio_file)
            audio_file.seek(0)
            tts_audio = AudioSegment.from_file(audio_file, format="mp3")
            combined_audio += tts_audio + AudioSegment.silent(duration=500)  # Add silence between languages
    
    # Save combined audio to a file
    output_file = "combined_output.mp3"
    combined_audio.export(output_file, format="mp3")
    
    return output_file

# Step 2: Create Gradio interface
with gr.Blocks() as demo:
    gr.Markdown("## Multilingual TTS: Generate a Single Audio File")
    
    with gr.Row():
        korean_input = gr.Textbox(label="Enter Korean Text:", placeholder="안녕하세요")
        english_input = gr.Textbox(label="Enter English Text:", placeholder="Hello")
        russian_input = gr.Textbox(label="Enter Russian Text:", placeholder="Привет")
        bulgarian_input = gr.Textbox(label="Enter Bulgarian Text:", placeholder="Здравейте")
    
    output_audio = gr.Audio(label="Generated Speech", type="filepath")
    generate_button = gr.Button("Generate Speech")
    
    generate_button.click(
        multilingual_tts, 
        inputs=[korean_input, english_input, russian_input, bulgarian_input], 
        outputs=output_audio
    )

# Run the app
if __name__ == "__main__":
    demo.launch()