app.py

import os
import subprocess
import gradio as gr

# Supported languages
LANGUAGE_CODES = {
    "English": "eng",
    "Spanish": "spa",
    "French": "fra",
    "German": "deu",
    "Italian": "ita",
    "Chinese": "cmn"
}

def translate_speech(audio_file, target_language):
    """
    Translate input speech (audio file) to the specified target language.
    
    Args:
    audio_file (str): Path to the input audio file.
    target_language (str): The target language for translation.
    
    Returns:
    str: Path to the translated audio file.
    """
    language_code = LANGUAGE_CODES[target_language]
    output_file = "translated_audio.wav"
    
    command = [
        "expressivity_predict",
        audio_file,
        "--tgt_lang", language_code,
        "--model_name", "seamless_expressivity",
        "--vocoder_name", "vocoder_pretssel",
        "--gated-model-dir", "seamlessmodel",
        "--output_path", output_file
    ]
    
    subprocess.run(command, check=True)

    if os.path.exists(output_file):
        print(f"File created successfully: {output_file}")
    else:
        print(f"File not found: {output_file}")

    return output_file

def create_interface():
    """Create and configure the Gradio interface."""
    inputs = [
        gr.Audio(type="filepath", label="Audio File"),
        gr.Dropdown(list(LANGUAGE_CODES.keys()), label="Target Language")
    ]

    return gr.Interface(
        fn=translate_speech,
        inputs=inputs,
        outputs=gr.Audio(label="Translated Audio"),
        title="Seamless Expressive Speech-To-Speech Translator",
        description="Hear how you sound in another language.",
    )

if __name__ == "__main__":
    iface = create_interface()
    iface.launch()