Update app.py

app.py

@@ -7,8 +7,10 @@ from pydub import AudioSegment
 import os
 import scipy.signal as signal
 import torch
-
+from pydub.utils import mediainfo
 from pydub.silence import detect_nonsilent  # Correct import
+import pandas as pd
+
 
 hf_token = os.getenv('diarizationToken')
 
@@ -29,26 +31,28 @@ def remove_phone_tonalities(audio, sr):
     filtered_audio = signal.filtfilt(b, a, audio)
     return filtered_audio
 
+
+
 def convert_audio_to_wav(audio_path):
-    # Convert any audio format to WAV using pydub
-    sound = AudioSegment.from_file(audio_path)
-    wav_path = "converted_audio.wav"
-    sound.export(wav_path, format="wav")
-    return wav_path
+    # Check the audio file format before conversion
+    audio_info = mediainfo(audio_path)
+    print(f"Audio file info: {audio_info}")
+    
+    if audio_info['format_name'] not in ['wav', 'mp3', 'flac', 'ogg']:  # Add other valid formats if necessary
+        raise ValueError(f"Unsupported audio format: {audio_info['format_name']}")
+    
+    try:
+        # Convert any audio format to WAV using pydub
+        sound = AudioSegment.from_file(audio_path)
+        wav_path = "converted_audio.wav"
+        sound.export(wav_path, format="wav")
+        return wav_path
+    except Exception as e:
+        print(f"Error converting audio: {e}")
+        raise
 
 
 
-import gradio as gr
-import librosa
-import numpy as np
-import whisperx
-from transformers import pipeline
-from pydub import AudioSegment
-import os
-import scipy.signal as signal
-import torch
-import pandas as pd
-from pydub.silence import detect_nonsilent
 
 hf_token = os.getenv('diarizationToken')
 
@@ -70,6 +74,8 @@ def remove_phone_tonalities(audio, sr):
     filtered_audio = signal.filtfilt(b, a, audio)
     return filtered_audio
 
+
+    
 def process_audio(audio_path):
     print(f"Received audio file: {audio_path}")