Update main.py

main.py

@@ -5,9 +5,13 @@ import numpy as np
 import tempfile
 from fastapi import FastAPI, UploadFile, File, HTTPException
 from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC
-from librosa.sequence import dtw
+from librosa.sequence import dtw  # Ensure librosa==0.9.2 is installed
 
-app = FastAPI(title="Quran Recitation Comparer API", description="Compares two Quran recitations using a deep wav2vec2 model.", version="1.0")
+app = FastAPI(
+    title="Quran Recitation Comparer API",
+    description="Compares two Quran recitations using a deep wav2vec2 model.",
+    version="1.0"
+)
 
 # --- Core Class Definition ---
 class QuranRecitationComparer:
@@ -118,13 +122,11 @@ class QuranRecitationComparer:
         """Clear the embedding cache to free memory."""
         self.embedding_cache = {}
 
-
 # --- FastAPI Startup Event ---
-# In production, consider loading sensitive tokens from environment variables or configuration files.
 @app.on_event("startup")
 def startup_event():
     global comparer
-    # For production, do not hardcode tokens; use os.environ.get(...) or a configuration system.
+    # In production, use environment variables or configuration management for tokens.
     auth_token = os.environ.get("HF_TOKEN")
     comparer = QuranRecitationComparer(
         model_name="jonatasgrosman/wav2vec2-large-xlsr-53-arabic",
@@ -132,13 +134,11 @@ def startup_event():
     )
     print("Model initialized and ready for predictions!")
 
-
 # --- API Endpoints ---
 @app.get("/", summary="Health Check")
 async def root():
     return {"message": "Quran Recitation Comparer API is up and running."}
 
-
 @app.post("/predict", summary="Compare Two Audio Files", response_model=dict)
 async def predict(file1: UploadFile = File(...), file2: UploadFile = File(...)):
     """
@@ -177,7 +177,6 @@ async def predict(file1: UploadFile = File(...), file2: UploadFile = File(...)):
         if tmp2_path and os.path.exists(tmp2_path):
             os.remove(tmp2_path)
 
-
 @app.post("/clear_cache", summary="Clear Embedding Cache", response_model=dict)
 async def clear_cache():
     """