Update functions.py

functions.py

@@ -113,7 +113,7 @@ def get_yt_audio(url):
     title = yt.title
 
     # Get the first available audio stream and download it
-    audio_stream =  yt.streams.filter(progressive=True, file_extension='mp4').order_by('resolution').desc().first().download()
+    audio_stream =  yt.streams.filter(progressive=True, file_extension='mp3').order_by('resolution').desc().first().download()
 
     return audio_stream, title
 
@@ -185,7 +185,7 @@ def transcribe_yt_video(link, py_tube=True):
 
             st.warning('File size larger than 24mb, applying chunking and transcription',icon="⚠️")
 
-            song = AudioSegment.from_file(st.session_state['audio'], format='mp4')
+            song = AudioSegment.from_file(st.session_state['audio'], format='mp3')
 
             # PyDub handles time in milliseconds
             twenty_minutes = 20 * 60 * 1000
@@ -199,8 +199,8 @@ def transcribe_yt_video(link, py_tube=True):
             print(video_id)
             
             for i, chunk in enumerate(chunks):
-                chunk.export(f'output/chunk_{i}_{video_id}.mp4', format='mp4')
-                transcriptions.append(load_whisper_api(f'output/chunk_{i}_{video_id}.mp4'))
+                chunk.export(f'output/chunk_{i}_{video_id}.mp4', format='mp3')
+                transcriptions.append(load_whisper_api(f'output/chunk_{i}_{video_id}.mp4')['text'])
 
             results = ','.join(transcriptions)
 
@@ -231,7 +231,7 @@ def transcribe_yt_video(link, py_tube=True):
 
             st.warning('File size larger than 24mb, applying chunking and transcription',icon="⚠️")
 
-            song = AudioSegment.from_file(st.session_state['audio'], format='mp4')
+            song = AudioSegment.from_file(st.session_state['audio'], format='mp3')
 
             # PyDub handles time in milliseconds
             twenty_minutes = 20 * 60 * 1000
@@ -243,7 +243,7 @@ def transcribe_yt_video(link, py_tube=True):
             video_id = extract.video_id(link)
             
             for i, chunk in enumerate(chunks):
-                chunk.export(f'output/chunk_{i}_{video_id}.mp3', format='mp4')
+                chunk.export(f'output/chunk_{i}_{video_id}.mp3', format='mp3')
                 transcriptions.append(load_whisper_api(f'output/chunk_{i}_{video_id}.mp4')['text'])
 
             results = ','.join(transcriptions)