Update Transformers code example

cc @reach-vb

README.md

@@ -103,22 +103,23 @@ inputs = processor(
 audio_values = model.generate(**inputs, max_new_tokens=256)
 ```
 
-3. Listen to the audio samples either in an ipynb notebook:
+4. Listen to the audio samples either in an ipynb notebook:
 
 ```python
 from IPython.display import Audio
 
 sampling_rate = model.config.audio_encoder.sampling_rate
-Audio(audio_values[0].numpy(), rate=sampling_rate)
+Audio(audio_values[0].cpu().numpy(), rate=sampling_rate)
 ```
 
-Or save them as a `.wav` file using a third-party library, e.g. `scipy`:
+Or save them as a `.wav` file using a third-party library, e.g. `soundfile`:
 
 ```python
-import scipy
+import soundfile as sf
 
 sampling_rate = model.config.audio_encoder.sampling_rate
-scipy.io.wavfile.write("musicgen_out.wav", rate=sampling_rate, data=audio_values[0, 0].numpy())
+audio_values = audio_values.cpu().numpy()
+sf.write("musicgen_out.wav", audio_values[0].T, sampling_rate)
 ```
 
 For more details on using the MusicGen model for inference using the 🤗 Transformers library, refer to the [MusicGen docs](https://huggingface.co/docs/transformers/model_doc/musicgen).