Update app.py

app.py

@@ -4,14 +4,10 @@ import os
 import streamlit as st
 import transformers
 from transformers import pipeline
-import torch
 from summarizer import Summarizer
+import torch
 import ffmpeg
 import speech_recognition as sr
-from pytube import YouTube
-import pandas as pd
-import string
-import whisper
 from moviepy.editor import AudioFileClip
 
 #Download youtube video
@@ -19,22 +15,18 @@ def download_audio(link):
     with yt_dlp.YoutubeDL({'extract_audio': True, 'format': 'bestaudio', 'outtmpl': 'video.mp3'}) as video:
         video.download(link)
 
-#Convert video format
-def convert_mp4_to_wav(mp4_file, wav_file):
-    video = AudioFileClip(mp4_file)
-    video.write_audiofile(wav_file)
+def get_transcription_en(mp3_file):
+    model = whisper.load_model("tiny.en")
+    directory = os.getcwd()
+    result = model.transcribe(os.path.join(directory, mp3_file))
+    return result["text"]
 
 #def portuguese_sum_pipeline(link):
 #    video_title = download_audio(link)
 
 def english_sum_pipeline(link):
     download_audio(link)
-
-    #audio-to-text
-    transcriptor = whisper.load_model("base.en")
-    directory = os.getcwd()
-    result = transcriptor.transcribe(os.path.join(directory, "video.mp3"))
-    transcript_text = result["text"]
+    transcript_text = get_transcription_en("video.mp3")
 
     #extractive summarization
     extractive_model = Summarizer()
@@ -51,9 +43,7 @@ def english_qa_pipeline(question, context):
     nlp = pipeline("question-answering", model='distilbert-base-uncased-distilled-squad')
     result = nlp(question=question, context=context)
     return result["answer"]
-    
-    
-                    
+               
 #Collect inputs and create the interface
 def main():
     header = st.container()