Update app.py

app.py

@@ -20,17 +20,16 @@ def transcribe_audio(audio_bytes):
     audio_tensor = torch.tensor(audio_array, dtype=torch.float64) / 32768.0
     
     # Provide inputs to the processor
-    inputs = processor(audio=audio_tensor, sampling_rate=16000, return_tensors="pt")
-    
-    # Generate logits from the model
-    logits = model(**inputs).logits
-    
-    # Decode the predicted IDs to get the transcription
-    predicted_ids = torch.argmax(logits, dim=-1)
-    transcription = processor.decode(predicted_ids[0])
-    
-    return transcription
+    #inputs = processor(audio=audio_tensor, sampling_rate=16000, return_tensors="pt")
+    input_features = processor(audio_tensor, sampling_rate=16000, return_tensors="pt").input_features 
+
+   # generate token ids
+    predicted_ids = model.generate(input_features)
+    # decode token ids to text
+    transcription = processor.batch_decode(predicted_ids, skip_special_tokens=False)
 
+    transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
+    return transcription
 
 # Streamlit app
 st.title("Audio to Text Transcription..")