import os
import gradio as gr
from faster_whisper import WhisperModel
from pytube import YouTube

# Inicializar el modelo Whisper
model = WhisperModel("base", device="cpu", compute_type="int8")

def transcribe_audio(audio_path):
    segments, _ = model.transcribe(audio_path, beam_size=5)
    return " ".join([segment.text for segment in segments])

def process_youtube(youtube_url):
    try:
        yt = YouTube(youtube_url)
        audio_stream = yt.streams.filter(only_audio=True).first()
        
        if not os.path.exists("temp"):
            os.makedirs("temp")
        
        output_path = audio_stream.download(output_path="temp")
        return transcribe_audio(output_path)
    except Exception as e:
        return f"Error processing YouTube URL: {str(e)}"

def transcribe(audio_file, youtube_url):
    if audio_file:
        return transcribe_audio(audio_file)
    elif youtube_url:
        return process_youtube(youtube_url)
    else:
        return "Please provide either an audio file or a YouTube URL."

# Definir la interfaz de Gradio
iface = gr.Interface(
    fn=transcribe,
    inputs=[
        gr.Audio(type="filepath", label="Upload Audio File"),
        gr.Textbox(label="Or Enter YouTube URL")
    ],
    outputs="text",
    title="Whisper Transcription App",
    description="Upload an audio file or provide a YouTube URL to transcribe. Note: This is running on CPU, so processing might be slower."
)

# Lanzar la aplicación
iface.launch()