re-clarify the opensource version weight

README.md

@@ -21,6 +21,8 @@ Kangdi Wang<sup>1</sup> · Yusheng Dai<sup>2</sup> · Jin Xu<sup>1†</sup>
 
 [[Demo Page](https://eps-acoustic-revolution-lab.github.io/EAR_VAE2/)] - [[Paper](https://arxiv.org/abs/2608.19843)] - [[Codebase](https://github.com/Eps-Acoustic-Revolution-Lab/EAR_VAE2)]
 
+> **⚠️ Note on open-source weights:**  Due to data licensing constraints, the open-source model weights are **retrained on publicly available datasets** (not the full internal training corpus). Performance may differ from the numbers reported in the paper, which were obtained with the full-scale proprietary training data.
+
 ---
 
 <p align="center">
@@ -126,21 +128,6 @@ python inference.py \
   --config checkpoints/config.json \
   --input input.wav --output output.wav
 ```
-
-
-## Model Details
-
-| Config | Params (M) | Latent dim | Rate (Hz) | Compression |
-|--------|:---:|:---:|:---:|:---:|
-| Small (C0=64) | ~42.6 | 128 | 25 | 1920× |
-
-- **Sample rate**: 48 kHz stereo
-- **STFT**: 3840-point FFT, 1920-sample hop → 25 Hz frame rate
-- **Latent**: 128-d continuous (VAE with KL regularization)
-- **Refiner**: 12-layer banded Transformer (256-d, 1024 intermediate)
-
-> **⚠️ Note on open-source weights:**  Due to data licensing constraints, the open-source model weights are **retrained on publicly available datasets** (not the full internal training corpus). Performance may differ from the numbers reported in the paper, which were obtained with the full-scale proprietary training data.
-
 ---
 
 ## Citation