docs: add batch_transcribe example to MLX section

README.md

@@ -607,6 +607,20 @@ text = transcribe(model, "audio.wav", task="sqa", question="What is the speaker
 text = transcribe(model, "audio.wav", task="summarize")
 ```
 
+#### Batch Inference
+
+Process multiple audio files in parallel for higher throughput:
+
+```python
+from mlx_meralion import load_model, batch_transcribe
+
+model = load_model("MERaLiON/MERaLiON-2-10B-MLX")
+
+results = batch_transcribe(model, ["audio1.wav", "audio2.wav", "audio3.wav"])
+for text in results:
+    print(text)
+```
+
 #### CLI
 
 ```bash