app.py

import gradio as gr
import whisper

# Function to process the audio file
def transcribe_audio(model_size, audio):
    # Load the Whisper model based on the user's choice
    model = whisper.load_model(model_size)
    
    # Transcribe the audio file
    result = model.transcribe(audio)
    
    return result['text']

# Gradio interface with model selection
iface = gr.Interface(
    fn=transcribe_audio,                           # The function that will process the audio and model choice
    inputs=[
        gr.Dropdown(label="Choose Whisper Model", choices=["tiny", "base", "small", "medium", "large"], value="base"),  # Model selection
        gr.Audio(type="filepath")  # Audio upload input (removed 'source')
    ],
    outputs="text",                                # Output transcription as text
    title="Whisper Audio Transcription",
    description="Upload an audio file and select a Whisper model to get the transcription."
)

# Launch the interface
iface.launch()