Test: ran the code and checked for ffmpeg functionality

README.md

@@ -12,11 +12,17 @@ Before running this script, make sure to install the necessary python packages u
 pip install -r requirements.txt
 ```
 
+If your system does not have ffmpeg installed you can install it via:
+
+```bash
+sudo apt install ffmpeg
+```
+
 ## Running the script
 
 To generate transcripts create a new job with your specifications (see jobs section below).
 <br/>
-If you need to download the transcriber model (not saved in `data/models` yet) set your huggingface token:
+If you need to download the transcriber model set your huggingface token:
 
 On Linux/Mac:
 

requirements.txt

@@ -1,5 +1,7 @@
 youtube-dl
 pytube
 pydub
+ffmpeg-python
+ffprobe
 torch
 git+https://github.com/m-bain/whisperx.git

src/transcriber.py

@@ -102,14 +102,12 @@ class Transcriber:
             logger.exception("Failed to save transcript: %s", e)
             raise
 
-    def _clear_memory(self):
+    @staticmethod
+    def _clear_memory():
         """
         Clear GPU memory.
         """
         try:
-            del self.transcription_model
-            del self.alignment_model
-            del self.metadata
             gc.collect()
             torch.cuda.empty_cache()
         except Exception as e: