Update utils.py

utils.py

@@ -1918,6 +1918,7 @@ def extract_speech_token(model, feature_extractor, utts, batch_size=128):
                     audio = _resample_buffer[sample_rate](audio)
                     audio = audio.to(audio_device)
                 else:
+                    print(f"{torchaudio.__version__=}")
                     audio = _resample_buffer[sample_rate](audio)
             # if audio.shape[0] > 1:
             #     audio = audio[:1]