Update app.py

app.py

@@ -34,7 +34,7 @@ def audio_extraction(video_file, output_format):
     # temp_filename = video_file.name
     # video_path = f"{temp_filename}"
     audio = extract_audio(
-        input_path=os.fspath(video_file), output_path=f"{str(video_file)[:-4]}.{output_format}", output_format=f"{output_format}"
+        input_path=os.fspath(video_file), output_path=f"{str(video_file)[:-4].mp3", output_format=f"{output_format}"
     )
     return audio
 
@@ -51,9 +51,9 @@ def load_asr_model():
     asr_model = pipeline(task="automatic-speech-recognition", model="openai/whisper-large-v3")
     return asr_model
     
-def transcribe_video(_processed_audio):
+def transcribe_video(processed_audio):
     transcriber_model = load_asr_model()
-    text_extract = transcriber_model(_processed_audio)
+    text_extract = transcriber_model(processed_audio)
     return text_extract['text']
 
 def generate_ai_summary(transcript):
@@ -76,7 +76,8 @@ with youtube_url_tab:
         if url:
            if st.button("Transcribe", key="yturl"):
                with st.spinner("Transcribing..."):
-                   audio = audio_extraction(yt_video, "wav")
+                   audio = audio_extraction(yt_video, "mp3")
+                   audio = audio_processing(audio)
                    ytvideo_transcript = transcribe_video(audio)
                st.success(f"Transcription successful")
                st.write(ytvideo_transcript)
@@ -98,6 +99,7 @@ with file_select_tab:
            if st.button("Transcribe", key="vidfile"):
                with st.spinner("Transcribing..."):
                    audio = audio_extraction(os.fspath(video_file), "mp3")
+                   audio = audio_processing(audio)
                    video_transcript = transcribe_video(audio) 
                    st.success(f"Transcription successful")
                    st.write(video_transcript)