Update app.py

app.py

@@ -15,9 +15,13 @@ def ASR(audio):
         # Copy the contents of the uploaded audio file to the temporary file
         temp_audio_file.write(open(audio, "rb").read())
         temp_audio_file.flush()
-        audio_array,sample_rate= torchaudio.load(temp_audio_file.name)
-        inputs = processor(audio_array, sampling_rate=16_000)
-        text = pipe(inputs=inputs.numpy())
+        # Load the audio file using torchaudio
+        waveform, sample_rate = torchaudio.load(temp_audio_file.name)
+
+        # Convert the PyTorch tensor to a NumPy ndarray
+        audio_array = waveform.numpy()
+        #inputs = processor(audio_array, sampling_rate=16_000)
+        text = pipe(audio_array)
         return text
 iface = gr.Interface(fn=ASR, inputs=audio_input, outputs="text")
 iface.launch(share=False)