Try replacing distil-large-v2 with distil-medium.en for faster transcription.

app.py

@@ -15,7 +15,7 @@ def init_speech_to_text_model():
     device = "cuda:0" if torch.cuda.is_available() else "cpu"
     torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
 
-    model_id = "distil-whisper/distil-large-v2"
+    model_id = "distil-whisper/distil-medium.en"
     model = AutoModelForSpeechSeq2Seq.from_pretrained(
         model_id, torch_dtype=torch_dtype, low_cpu_mem_usage=True, use_safetensors=True
     )