app.py

import gradio as gr
import os
from pyannote.audio import Pipeline
from pyannote.core import Segment
from pyannote.audio import Audio
from pydub import AudioSegment

api_k = os.getenv("API_KEY")
# Initialize the pyannote pipeline for speaker diarization
pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization-3.1",use_auth_token=api_k)

# Function to process the audio input and return both text and audio response
def respond(audio_input):
    # Save the audio input to a file
    audio_file = audio_input
    # Perform diarization on the whole file
    diarization = pipeline(audio_file)

    return text_response

input_audio = gr.Audio(
    sources=["microphone"],
    waveform_options=gr.WaveformOptions(
        waveform_color="#01C6FF",
        waveform_progress_color="#0066B4",
        skip_length=2,
        show_controls=False,
    ),
)

gr.Interface(
    fn=respond,
    inputs=input_audio,
    outputs="text",
    title="Tommy Vercetti Chatbot",
    description="Chat with Tommy Vercetti from GTA Vice City. Get responses in both text and voice!"
).launch(debug=True)