Add: Code example in README

README.md

@@ -26,7 +26,17 @@ pipeline_tag: text-to-speech
 ---
 
 # Changes to suno/bark
-This duplicated repo allows to pass a voice preset besides the text input. See `handler.py` for details.
+This duplicated repo allows to pass a voice preset besides the text input when using an inference client. See `handler.py` for details.
+Example:
+```python
+import numpy as np
+from huggingface_hub import InferenceClient
+
+client = InferenceClient(model="<my-inference-url>")
+sentence = "hello my name is suno"
+audio_raw = client.post(json={'inputs': sentence, 'voice_preset': "v2/en_speaker_0",})
+audio = np.array(eval(audio_raw)[0]['generated_audio'][0],)
+```
 
 # Bark