Update README.md

README.md

@@ -1,7 +1,8 @@
 SPIRAL: Self-supervised Perturbation-Invariant Representation Learning for Speech Pre-Training
 ======== 
 
-This is the pretrained model of **SPIRAL Base with Multi-Condition Training**, trained with 960-hour LibriSpeech data, and additional AudioSet Data for noise-robustness.
+This is the pretrained model of **SPIRAL Base with Multi-Condition Training**, trained with 960-hour LibriSpeech data, and noise dataset from [ICASSP 2021 DNS Challenge](https://github.com/microsoft/DNS-Challenge/tree/icassp2021-final) for noise robustness.
+
 
 Citation
 ========