librosa

app.py

@@ -13,6 +13,7 @@ import os
 import re
 
 import pandas as pd
+import librosa
 
 import importlib 
 modeling_MERT = importlib.import_module("MERT-v1-95M.modeling_MERT")
@@ -169,11 +170,11 @@ def model_inference(inputs):
     return df
 
 def convert_audio(inputs):
-    return model_inference(inputs)    
+    audio_data, sample_rate = librosa.load(inputs, sr=None)
+    return model_inference(audio_data)    
 
 def build_audio_flow(title, description, article):
-    # Define file upload element
-    audio_file = gr.File(label="Select Audio File (WAV, FLAC)")
+    audio_file = gr.File(label="Select Audio File")
 
     demo =  gr.Interface(
         fn=convert_audio,