app.py

import gradio as gr
import whisper

def convert_to_text(audio_path: str) -> str:  
    model = whisper.load_model("base")
    result = model.transcribe(audio_path)
    return result["text"]

audio_input = gr.inputs.Audio(type="file", label="Upload an Audio")

iface = gr.Interface(
    fn=convert_to_text,
    inputs=audio_input,
    outputs="text",
    title="Transform your audio into text",
    description="Upload an audio here, either from your mobile or laptop and get it transcribed in a matter of seconds.",
    theme="Monochrome"
)

iface.launch()