Update app.py

app.py

@@ -1,6 +1,6 @@
 import gradio as gr
 from transformers import pipeline, WhisperProcessor, WhisperForConditionalGeneration
-import torch
+import torchaudio
 import soundfile as sf
 
 # Load Whisper model and processor
@@ -15,8 +15,13 @@ def transcribe_and_analyze(audio_path):
     # Load audio from the provided file
     audio, sample_rate = sf.read(audio_path)
     
+    # Resample audio to 16000 Hz if necessary
+    if sample_rate != 16000:
+        audio_tensor = torchaudio.functional.resample(torch.tensor(audio), orig_freq=sample_rate, new_freq=16000)
+        audio = audio_tensor.numpy()  # Convert back to numpy array
+    
     # Process audio with Whisper
-    input_features = processor(audio, sampling_rate=sample_rate, return_tensors="pt").input_features
+    input_features = processor(audio, sampling_rate=16000, return_tensors="pt").input_features
     predicted_ids = model.generate(input_features)
     transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0]