Update app.py

app.py

@@ -10,24 +10,24 @@ import torch
 from pathlib import Path
 from datetime import datetime
 
-# קריאת טוכן מ-Secrets
+# Read token from environment
 HF_TOKEN = os.getenv("HF_TOKEN")
 if not HF_TOKEN:
-    raise RuntimeError("❌ HF_TOKEN environment variable is required")
+    raise RuntimeError("HF_TOKEN environment variable is required")
 
-# טעינת המודל
+# Load model
 device = "cuda" if torch.cuda.is_available() else "cpu"
-print(f"🚀 Loading model on {device}...")
+print(f"Loading model on {device}...")
 
 try:
     pipeline = Pipeline.from_pretrained(
         "ivrit-ai/pyannote-speaker-diarization-3.1",
         use_auth_token=HF_TOKEN,
     )
-    pipeline.to(torch.device(device))  # ✅ תיקון כאן!
-    print("✅ Model loaded successfully!")
+    pipeline.to(torch.device(device))
+    print("Model loaded successfully!")
 except Exception as e:
-    print(f"❌ Failed to load model: {e}")
+    print(f"Failed to load model: {e}")
     raise
 
 app = FastAPI(
@@ -36,18 +36,18 @@ app = FastAPI(
     version="1.0.0"
 )
 
-# הגבלות
+# Limits
 MAX_FILE_SIZE_MB = 50
 MAX_DURATION_MINUTES = 15
 MAX_CONCURRENT_REQUESTS = 2
 
-# ניהול תור
+# Queue management
 processing_semaphore = asyncio.Semaphore(MAX_CONCURRENT_REQUESTS)
 active_requests = 0
 
 
 def ensure_wav_16k_mono(in_path: str) -> str:
-    """ממיר אודיו ל-WAV 16kHz mono"""
+    """Convert audio to WAV 16kHz mono"""
     out_path = str(Path(in_path).with_suffix(".wav"))
     cmd = [
         "ffmpeg", "-y", "-i", in_path,
@@ -62,7 +62,7 @@ def ensure_wav_16k_mono(in_path: str) -> str:
 
 
 def estimate_duration(file_path: str) -> float:
-    """אומדן אורך קובץ בדקות"""
+    """Estimate file duration in minutes"""
     try:
         file_size_mb = os.path.getsize(file_path) / (1024 * 1024)
         return file_size_mb / 2.0
@@ -72,7 +72,7 @@ def estimate_duration(file_path: str) -> float:
 
 @app.get("/")
 def root():
-    """מידע על ה-API"""
+    """API information"""
     global active_requests
     return {
         "service": "Hebrew Speaker Diarization API",
@@ -99,7 +99,7 @@ def root():
 
 @app.get("/health")
 def health():
-    """בדיקת בריאות"""
+    """Health check"""
     global active_requests
     return {
         "status": "healthy",
@@ -113,13 +113,13 @@ def health():
 @app.post("/diarize")
 async def diarize(file: UploadFile = File(...)):
     """
-    זיהוי דוברים בקובץ אודיו
+    Speaker diarization for audio file
     
     Args:
-        file: קובץ אודיו (MP3, WAV, M4A, וכו')
+        file: Audio file (MP3, WAV, M4A, etc.)
     
     Returns:
-        JSON: רשימת מקטעים עם זיהוי דוברים
+        JSON: List of segments with speaker identification
     """
     global active_requests
     
@@ -160,13 +160,13 @@ async def diarize(file: UploadFile = File(...)):
                         detail=f"File too long: ~{duration:.1f} min (max: {MAX_DURATION_MINUTES} min)"
                     )
                 
-                print(f"🎤 Processing: {file.filename} ({file_size_mb:.1f}MB)")
+                print(f"Processing: {file.filename} ({file_size_mb:.1f}MB)")
                 start_time = datetime.now()
                 
                 annotation = pipeline(wav_path)
                 
                 processing_time = (datetime.now() - start_time).total_seconds()
-                print(f"✅ Done in {processing_time:.1f}s")
+                print(f"Done in {processing_time:.1f}s")
                 
                 segments = []
                 last_segment = None
@@ -209,7 +209,7 @@ async def diarize(file: UploadFile = File(...)):
     except HTTPException:
         raise
     except Exception as e:
-        print(f"❌ Error: {str(e)}")
+        print(f"Error: {str(e)}")
         raise HTTPException(status_code=500, detail=f"Processing error: {str(e)}")
     finally:
         for path in [tmp_path, wav_path]:
@@ -222,21 +222,4 @@ async def diarize(file: UploadFile = File(...)):
 
 if __name__ == "__main__":
     import uvicorn
-    uvicorn.run(app, host="0.0.0.0", port=7860)
-```
-
----
-
-## 🚀 עכשיו זה אמור לעבוד!
-
-1. **עדכן את `app.py`** ב-Space
-2. שמור
-3. המתן לבנייה (~1-2 דקות - כי יש cache)
-4. בדוק `/health`
-
-אחרי התיקון הזה, אמור לראות בלוגים:
-```
-🚀 Loading model on cpu...
-✅ Model loaded successfully!
-INFO:     Started server process
-INFO:     Uvicorn running on http://0.0.0.0:7860
+    uvicorn.run(app, host="0.0.0.0", port=7860)