Update app.py

app.py

@@ -70,21 +70,22 @@ def convert_to_wav(filepath):
 pipe = pipeline("automatic-speech-recognition", model="NbAiLab/nb-whisper-large", chunk_length_s=30, generate_kwargs={'task': 'transcribe', 'language': 'no'})
 
 @spaces.GPU()
-def transcribe_audio(filepath, batch_size=16, sample_rate=16000):
+def transcribe_audio(filepath, batch_size=16):
     
-    if not isinstance(filepath, str):
-        raise ValueError("Expected a file path string, but received something else.")
-
-    # Ensure the file exists before trying to load it
-    if not Path(filepath).is_file():
-        raise FileNotFoundError(f"The file {filepath} does not exist.")
-    
-    waveform, sample_rate = torchaudio.load("{filepath}")   
+    with tempfile.NamedTemporaryFile(suffix=".wav") as temp_audio_file:
+        # Copy the contents of the uploaded audio file to the temporary file
+        temp_audio_file.write(open(audio, "rb").read())
+        temp_audio_file.flush()
+        # Load the audio file using torchaudio
+        waveform, sample_rate = torchaudio.load(temp_audio_file.name)
+        # Resample the audio to 16kHz
+        resampler = torchaudio.transforms.Resample(sample_rate, 16000)
+        waveform = resampler(waveform)
            
     # --convert to mono
     if waveform.ndim > 1:
         waveform = waveform[0, :]
-
+    # Convert PyTorch tensor NumPy ndarray
     waveform = waveform.numpy()
 
     start_time = time.time()