Update app.py

app.py

@@ -3,6 +3,7 @@ from transformers import pipeline
 import numpy as np
 import os
 from huggingface_hub import login
+import librosa
 import spaces
 
 HF_TOKEN = os.environ.get("HF_TOKEN")
@@ -12,17 +13,24 @@ if HF_TOKEN:
 MODEL_ID = "badrex/ASRwanda"
 transcriber = pipeline("automatic-speech-recognition", model=MODEL_ID)
 
+
 @spaces.GPU
 def transcribe(audio):
     sr, y = audio
+    
     # convert to mono if stereo
     if y.ndim > 1:
         y = y.mean(axis=1)
+    
+    # resample to 16kHz if needed
+    if sr != 16000:
+        y = librosa.resample(y, orig_sr=sr, target_sr=16000)
+    
     y = y.astype(np.float32)
     y /= np.max(np.abs(y))
+    
     return transcriber({"sampling_rate": 16000, "raw": y})["text"]
 
-
 examples = []
 examples_dir = "examples"
 if os.path.exists(examples_dir):