app.py

import gradio as gr
from pytube import YouTube
import subprocess
from huggingsound import SpeechRecognitionModel
import torch
import librosa
from transformers import pipeline

def process_video(video_url):
    try:
        # Download audio from YouTube
        yt = YouTube(video_url)
        audio_file = yt.streams.filter(only_audio=True, file_extension='mp4').first().download(filename='ytaudio.mp4')
    except Exception as e:
        return f"Error downloading audio from YouTube: {e}"

    try:
        # Convert to suitable format for speech recognition
        subprocess.run(['ffmpeg', '-i', 'ytaudio.mp4', '-acodec', 'pcm_s16le', '-ar', '16000', 'ytaudio.wav'], check=True)
    except subprocess.CalledProcessError as e:
        return f"Error converting audio file: {e}"

    try:
        # Speech Recognition
        device = "cuda" if torch.cuda.is_available() else "cpu"
        model = SpeechRecognitionModel("jonatasgrosman/wav2vec2-large-xlsr-53-english", device=device)
        transcription = model.transcribe(['ytaudio.wav'])[0]['transcription']
    except Exception as e:
        return f"Error in speech recognition: {e}"

    try:
        # Summarize Transcription
        summarization = pipeline('summarization')
        summarized_text = summarization(transcription, max_length=130, min_length=30, do_sample=False)
        return summarized_text[0]['summary_text']
    except Exception as e:
        return f"Error summarizing text: {e}"

iface = gr.Interface(fn=process_video, inputs=gr.inputs.Textbox(lines=2, placeholder="Enter YouTube Video URL Here..."), outputs="text", title="YouTube Video Summarizer", description="This tool extracts audio from a YouTube video, transcribes it, and provides a summary.")
iface.launch()