Hello, I am a phonetic beginner. I would like to use wsj0-2mix data set s1 as the training set of ETESpeechRecognition to improve the separation performance of convtasnet by extracting features. I would like to ask whether this method is feasible. Your advice and help are urgently needed
Hello, I am a phonetic beginner. I would like to use wsj0-2mix data set s1 as the training set of ETESpeechRecognition to improve the separation performance of convtasnet by extracting features. I would like to ask whether this method is feasible. Your advice and help are urgently needed