app.py

import spaces
import gradio as gr

from audio_processing import process_audio


@spaces.GPU
def gradio_process_audio(audio):
    try:
        if audio is None:
            return "No file uploaded", "", ""
        
        # Extract the sample rate and audio data from the Gradio input
        sample_rate, audio_data = audio
        
        # Call the existing process_audio function
        detected_lang, transcription, translation = process_audio((sample_rate, audio_data))
        
        return detected_lang, transcription, translation
    except Exception as e:
        return str(e), "", ""

iface = gr.Interface(
    fn=gradio_process_audio,
    inputs=gr.Audio(type="numpy"),
    outputs=[
        gr.Textbox(label="Detected Language"),
        gr.Textbox(label="Transcription", lines=5),
        gr.Textbox(label="Translation", lines=5)
    ],
    title="Audio Transcription and Translation",
    description="Upload an audio file to detect its language, transcribe, and translate it.",
    allow_flagging="never",
    css=".output-textbox { font-family: 'Noto Sans Devanagari', sans-serif; font-size: 18px; }"
)

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