ASR perfomance of enhanced speech data

[NewEricWang]

Hi,
Do anybody try to use the enhanced speech data for ASR?
Can you get performance improvement using the enhancement method?
I failed to get good results.