Update app.py

app.py

@@ -1,11 +1,23 @@
 import gradio as gr
 from transformers import pipeline
+import torch
+import librosa
 
-asr_pipeline = pipeline("automatic-speech-recognition", model="monadical-labs/whisper-medium.en")
+# Load ASR pipeline
+device = "cuda" if torch.cuda.is_available() else "cpu"
+asr_pipeline = pipeline("automatic-speech-recognition", model="monadical-labs/whisper-medium.en", device=device)
 
 def transcribe(audio):
-    text = asr_pipeline(audio)["text"]
+    if audio is None:
+        return "Error: No audio file received."
+    
+    # Load the audio file correctly
+    audio_data, sr = librosa.load(audio, sr=16000)  # Resample to 16kHz (Whisper requirement)
+    
+    # Process the audio
+    text = asr_pipeline(audio_data)["text"]
     return text
 
+# Create Gradio interface
 demo = gr.Interface(fn=transcribe, inputs=gr.Audio(type="filepath"), outputs="text")
-demo.launch()
+demo.launch()