try with different model

app.py

@@ -2,13 +2,14 @@ from transformers import pipeline
 import gradio as gr
 from pytube import YouTube
 
-model = pipeline(model="SofiaK/dataset")
+model = pipeline(model="SofiaK/checkpoints")
 
 
 def youtube_to_text(youtube_url):
     video = YouTube(youtube_url).streams.filter(only_audio=True).all()
     audio = video[0].download()
     text = model(audio)["text"]
+    # todo : remove the audio file after the inference
     return text