app.py

import gradio as gr
import whisper
import tempfile
import os

model = whisper.load_model("base")


def video_to_text(video_file):
    video_path = video_file.name
    transcription = model.transcribe(video_path)

    return transcription['text']


iface = gr.Interface(
    fn=video_to_text,
    inputs=gr.File(file_types=["video"]),
    outputs="text",
    title="Video to Text Transcription",
    description="Upload a video and get the transcribed text"
)

iface.launch()