transcription uses an executable

.gitattributes

@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+exec/yt-dlp_linux filter=lfs diff=lfs merge=lfs -text

exec/yt-dlp_linux

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7494af600c4389d19961b44e406472bba7c54c28dc1f62bb72bdde7181708a9a
+size 29333576

transcription.py

@@ -5,6 +5,7 @@ from pydub import AudioSegment
 from pydub.utils import make_chunks
 import pydub
 from pathlib import Path
+import subprocess
 
 # For getting text from PDF
 from zipfile import ZipFile
@@ -57,7 +58,7 @@ class DownloadAudio:
                 self.yt = YouTube(self.link)
                 continue
 
-    def download(self, pathname:str):
+    def download(self, pathname:str) -> str:
         """
         Download the audio from the youtube video and saves it to multiple .wav files
         in the specified folder. Returns a list of the paths to the .wav files.
@@ -69,13 +70,15 @@ class DownloadAudio:
         FINAL_WAV_PATH = f"{pathname}/{self.WAV_FILE_NAME}"
 
         if not os.path.exists(FINAL_WAV_PATH):
-            # Download the .mp4 file
-            audiostream = self.yt.streams.filter(only_audio=True).first()
-            outfile_path = audiostream.download(pathname)
-
-            # Convert the .mp4 file to .wav
-            wav_file = AudioFileClip(outfile_path)
-            wav_file.write_audiofile(FINAL_WAV_PATH, bitrate="16k", fps=16000)
+            print("\n\n\n DOWNLOADING AUDIO \n\n\n")
+            current_dir = os.getcwd()
+            print(current_dir)
+            executable_path = os.path.join(current_dir, "exec/yt-dlp_linux")
+            # Download the video as an audio file using youtube-dl
+            result = subprocess.run([executable_path, "-x", "--audio-format", "wav", "-o", FINAL_WAV_PATH, self.link])
+            if result.returncode != 0:
+                print("Failed to download audio. Retrying...")
+                return "FAILED"
 
         # Load the input .wav file
         audio = AudioSegment.from_wav(FINAL_WAV_PATH)