app.py

from huggingface_hub import login
from diarization import start_diarization
from transcribe import start_transcribe
import ffmpeg
import gradio as gr
import os


def prepare_input(input_file):
    output_file = "input.wav"
    ffmpeg.input(input_file).audio.output(
        output_file, format="wav").run()

    progress = gr.Progress()
    start_diarization(output_file, progress)
    return start_transcribe(progress)


video_interface = gr.Interface(
    fn=prepare_input,
    inputs=gr.Video(type="file"),
    outputs="text",
    title="Test 1"
)

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