Hi guyz, I am trying to implement the paper http://www.ncbi.nlm.nih.gov/pubmed/23757540 to learn more about machine learning but it uses XM2VTS dataset for speaker-independent audio-visual speech recognition. which is not free and as currently I neither belong to academia nor have much funds I am stuck .. Can someone please help me find some open source data set that I can use...
[link][comment]