Update app.py

app.py

@@ -6,7 +6,8 @@ from pathlib import Path
 
 def install_dependencies():
     commands = [
-        "pip install spaces-0.1.0-py3-none-any.whl"
+        "pip install spaces-0.1.0-py3-none-any.whl",
+        "pip install librosa"
     ]
     for cmd in commands:
         os.system(cmd)
@@ -18,6 +19,7 @@ from PIL import Image
 import soundfile as sf
 import torch
 import gradio as gr
+import librosa
 from huggingface_hub import snapshot_download
 
 try:
@@ -40,16 +42,7 @@ pipe = None
 
 def load_audio_for_model(audio_filepath):
     try:
-        wav, sr = sf.read(audio_filepath)
-        if wav.dtype != np.float32:
-            if np.issubdtype(wav.dtype, np.integer):
-                wav = wav.astype("float32") / 32768.0
-            else:
-                wav = wav.astype("float32")
-        
-        if wav.ndim > 1:
-            wav = wav.mean(axis=1)
-            
+        wav, sr = librosa.load(audio_filepath, sr=16000)
         return wav, sr
     except Exception:
         return None, None