Added Functions & Updated Requirements File

media_download.py

@@ -372,24 +372,51 @@ class YoutubeDownloader(MediaDownloader):
             video_path (str): Path to the input video file.
 
         Returns:
-            bool: True if extraction is successful, False otherwise.
+            str: Path of extracted audio.
         """
         try:
-            # Determine the file format (MP4 or WebM) based on the file extension
+            # Path for Extracted Audio File
             filename, extension = os.path.splitext(video_path)
-
-            # Extracted audio path
             audio_path = filename + '.mp3'
 
-            # Choose the appropriate codec for the output audio format (MP3)
+            # Choosing the Appropriate Codec for the Output Audio Format (MP3)
             audio_codec = "libmp3lame" if extension.lower() in (".mp4", ".webm") else "mp3"
 
-            # Run the ffmpeg command to extract audio
-            subprocess.run(["ffmpeg", "-i", video_path, "-vn", "-acodec", 
-                            audio_codec, audio_path, '-loglevel', 'quiet'], check=True)
-            
+            # Extracting Audio using FFMPEG Command
+            command = ["ffmpeg", "-i", video_path, "-vn", "-acodec", 
+                       audio_codec, audio_path, '-loglevel', 'quiet']
+            subprocess.run(command, check=True)
             
             return audio_path
         
+        except subprocess.CalledProcessError as e:
+            print(f"Error: {e}")
+    
+    @staticmethod
+    def burn_subtitles(video_file_path, subtitle_file_path):
+        '''
+        Burns the subtitles onto the video
+
+        Args:
+            video_file_path (str): Path to the input video file.
+            subtitle_file_path (str): Path to the subtitle file.
+
+        Returns:
+            str: Path of output video with subtitles.
+        '''
+        try:
+            # Getting Output File Path
+            video_filename, video_extension = os.path.splitext(video_file_path)
+            subtitle_filename, subtitle_extension = os.path.splitext(subtitle_file_path)
+            output_file_path = video_filename + subtitle_extension.replace('.', '_') + video_extension
+
+            # Burning the Subtitles onto Video using FFMPEG Command
+            command = ['ffmpeg', '-i', video_file_path, 
+                    '-vf', f'subtitles={subtitle_file_path}',
+                    output_file_path, '-loglevel', 'quiet']
+            subprocess.run(command, check=True)
+
+            return output_file_path
+            
         except subprocess.CalledProcessError as e:
             print(f"Error: {e}")

requirements.txt

@@ -1,7 +1,6 @@
 fastapi
 faster-whisper
 langchain
-nltk
 openai
 pandas
 pytube
@@ -12,3 +11,5 @@ uvicorn
 wordcloud
 youtube-transcript-api
 git+https://github.com/suno-ai/bark.git
+--extra-index-url https://download.pytorch.org/whl/cu113
+torch

summarizer.py

@@ -10,6 +10,10 @@ from sklearn.cluster import KMeans
 from sklearn.metrics import silhouette_score
 import os
 from langchain.docstore.document import Document
+import re
+from collections import Counter
+# import nltk
+from nltk.corpus import stopwords
 
 os.environ["OPENAI_API_KEY"] = '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' 
 
@@ -118,7 +122,7 @@ class Extract_Summary:
         prompt_template = """
             Extract Key Informtion from the text below. This key information can include People Names & their Role/rank, Locations, Organization,Nationalities,Religions,
             Events such as Historical, social, sporting and naturally occurring events, Products , Address & email, URL, Date & Time, Provide the list of Key information each 
-            should be labeled with thier crossponding category.if key information related to category is not present, dont add that category in Response.
+            should be labeled with thier crossponding category.if key information related to category is not present,add "Not mentioned" in the response.
                     {text}
 
                     """
@@ -133,7 +137,7 @@ class Extract_Summary:
         map_prompts = """
                     Extract Key Informtion from the text below. This key information can include People Names & their Role/rank, Locations, Organization,Nationalities,Religions,
                     Events such as Historical, social, sporting and naturally occurring events, Products , Address & email, URL, Date & Time, Provide the list of Key information each 
-                    should be labeled with thier crossponding category.if key information related to category is not present, dont add that category in Response.
+                    should be labeled with thier crossponding category.if key information related to category is not present, add Not mentioned in the response.
                             {text}
 
                             """
@@ -141,7 +145,7 @@ class Extract_Summary:
                     Below Text contains Key Information that was extracted from text. You job is to combine the Key Information and Return the results.This key information can include People Names & their Role/rank, 
                     Locations, Organization,Nationalities,Religions,Events such as Historical, social, sporting and naturally occurring events, Products , 
                     Address & email, URL, Date & Time, Provide the list of Key information each should be labeled with thier crossponding category.
-                    if key information related to category is not present, dont add that category in Response. 
+                    if key information related to category is not present, add Not mentioned in the response. 
                             {text}
 
                             """
@@ -253,11 +257,56 @@ class Extract_Summary:
         # display(Markdown(f"Text: {docs}"))
         # display(Markdown(f"Summary Response: {output}"))
         return output
+    
+
+    def parse_key_information(self,text):
+
+        lines = [line.strip() for line in text.split('\n') if line.strip()]
+        # Initialize the dictionary to store information
+        info_dict = {}
+
+        current_category = None
+
+        # Iterate through each line and process the information
+        for line in lines:
+            if re.match(r'^[A-Z][\w\s&/-]*:', line):
+                current_category = line.rstrip(':')
+                info_dict[current_category] = []
+            else:
+                if line != '- Not mentioned':
+                    info_dict[current_category].append(line.replace('- ', ''))
+
+        # Remove categories with no entries
+        info_dict = {category: entries for category, entries in info_dict.items() if entries}
+
+        return info_dict
+
 
-    def create_wordcloud(self, output):
-        wc = WordCloud(stopwords=STOPWORDS, height=500, width=300)
-        wc.generate(output)
-        wc.to_file('WordCloud.png')
+    # def create_wordcloud(self, output):
+    #     wc = WordCloud(stopwords=STOPWORDS, height=500, width=300)
+    #     wc.generate(output)
+    #     wc.to_file('WordCloud.png')
+
+
+    def create_word_count(text):
+        # Split the text into words, convert them to lowercase
+        words = text.split()
+        words = [word.lower() for word in words]
+
+        # Get a list of English stop words
+        stop_words = set(stopwords.words('english'))
+
+        # Filter out stop words from the list of words
+        filtered_words = [word for word in words if word not in stop_words]
+
+        # Count the frequencies of each word
+        word_counts = Counter(filtered_words)
+
+        # Convert the Counter object to a dictionary
+        word_count_dict = dict(word_counts)
+
+        return word_count_dict
+   
 
 
 class AudioBookNarration: