Merge pull request #27 from TommasU/saurabh-dev

Code documentation review and edits

source/main/transcribe.py

@@ -42,7 +42,7 @@ def __init(self):
 
     def transcribe_video(self, ip_path):
         """
-        Generate summary from video without Closed Captions
+        Generate summary on punctuated transcript from video without Closed Captions
         """
         # Read video input
         video = mp.VideoFileClip(ip_path)
@@ -67,11 +67,12 @@ def transcribe_video(self, ip_path):
             # Perform speech to text and store the text
             transcript_text += recognizer.recognize_google(audio_data)
 
+        # Adding punctuation to transcript
         punctuated_transcription = Punctuation.add_punctuation_transcript(
             transcript_text
         )
 
-        # Call the summarization script
+        # Call the summarization script on the punctuated transcript
         transcript_summary = Summary(punctuated_transcription)
         summary = transcript_summary.summarize_text()
         for lines in summary:

source/main/transcribe_yt.py

@@ -70,7 +70,7 @@ def check_yt_cc(self):
 
     def transcribe_yt_video(self):
         """
-        Caller function for methods in the class
+        Calls correct transcribe function after checking youtube video has captions or not
         """
         check_cc = self.check_yt_cc()
         # If captions are not present
@@ -84,7 +84,7 @@ def transcribe_yt_video(self):
 
     def transcribe_yt_video_w_cc(self):
         """
-        Generate summary for Youtube videos with Closed Captions
+        Generate summary on punctuated transcript for Youtube videos with Closed Captions
         """
         # Get transcript from youtube video
         transcript_json = YouTubeTranscriptApi.get_transcript(self.yt_id)
@@ -93,11 +93,12 @@ def transcribe_yt_video_w_cc(self):
         for rec in transcript_json:
             transcript_text += " " + rec["text"]
 
+        # Add punctuation to transcript
         punctuated_transcription = Punctuation.add_punctuation_transcript(
             transcript_text
         )
 
-        # Call the summarization script
+        # Call the summarization script on punctuated transcript
         transcript_summary = Summary(punctuated_transcription)
         summary = transcript_summary.summarize_text()
         for lines in summary:

source/scrivener_user_interface.py

@@ -63,6 +63,8 @@
 if not os.path.exists('source/punct_model_full.pcl'):
     print("Creating punct_model_full.pcl file for ML model...")
 
+
+    # Path to model files parts that needs to be combined
     # Storing these models in github causes an issue with the Heroku deployment and exceeds 500 MB (it is 618 MB)
     # slug/payload limit. Therefore, using this alternative to get it from Github during runtime.
     if not os.path.exists('source/punct_model_part1.pcl'):
@@ -83,11 +85,15 @@
         filename = wget.download(url3, out='source/punct_model_part3.pcl')
         print("\nDownloaded file: " + filename)
 
+
     first_file = os.path.abspath('source/punct_model_part1.pcl')
     second_file = os.path.abspath('source/punct_model_part2.pcl')
     third_file = os.path.abspath('source/punct_model_part3.pcl')
+
+    # Path to combined model file
     new_file = os.path.abspath('source/punct_model_full.pcl')
 
+    # Read content of model file parts and write it to the combined model file
     with open(new_file, "wb") as wfd:
         for f in [first_file, second_file, third_file]:
             with open(f, "rb") as fd:
@@ -135,8 +141,7 @@ def save_file(file):
         # Display Summary
         st.header('Summary')
         st.write(summary)
-
-
+
     # If user inputs an invalid Youtube link
     elif youtube_link!='':
         st.error('Please enter a valid Youtube Link!')