Fix README: correct quantization description, remove mlx-audio version reference

README.md

@@ -93,9 +93,9 @@ pipeline_tag: text-to-speech
 base_model: fishaudio/s2-pro
 ---
 
-# mlx-community/s2-pro-bf16
+# fish-audio-s2-pro-8bit
 
-This model was converted to MLX format from [`fishaudio/s2-pro`](https://huggingface.co/fishaudio/s2-pro) using mlx-audio version **0.4.0**.
+This model is an 8-bit quantized version of [`fishaudio/s2-pro`](https://huggingface.co/fishaudio/s2-pro), quantized using `mlx.nn.quantize()` (group_size=64, bits=8) via [scripts/quantize_fish_s2.py](https://github.com/cs2764/mlx-audio-api/blob/main/scripts/quantize_fish_s2.py).
 
 Refer to the [original model card](https://huggingface.co/fishaudio/s2-pro) for more details on the model.