app.py

import gradio as gr
import os
import torch
import io
from pyannote.audio import Pipeline
from pyannote.audio import Audio
from pyannote.audio.pipelines.utils.hook import TimingHook
from pyannote.core import Segment

pipeline = Pipeline.from_pretrained(
  "pyannote/speaker-diarization-3.1",
  use_auth_token=os.environ['api'])

def process_audio(audio_file):
    # Save the uploaded audio file to a temporary location
    temp_file = "temp_audio.wav"
    with open(temp_file, "wb") as f:
        f.write(audio_file.read())
    
    # Use the diarization pipeline to process the audio file
    diarization = pipeline(temp_file)
    
    # Remove the temporary file
    os.remove(temp_file)
    
    # Return the diarization output
    return diarization

with gr.Blocks() as demo:
    audio_input = gr.File(label="Upload Audio", file_types=["audio"])
    process_button = gr.Button("Process")
    diarization_output = gr.JSON(label="Diarization Output")

    process_button.click(fn=process_audio, inputs=audio_input, outputs=diarization_output)

demo.launch()