Author:Sahraeian, R; Van Compernolle, D; De Wet, FebeDate:May 2014This paper demonstrates the application of Intrinsic Spectral Analysis (ISA) for low-resource Automatic Speech Recognition (ASR). State-of-the-art speech recognition systems that require large amounts of task specific training data fail to ...Read more
Author:Sahraeian, R; van Compernolle, D; De Wet, FebeDate:Sep 2015Conventional acoustic modeling involves estimating many parameters to effectively model feature distributions. The sparseness of speech and text data, however, degrades the reliability of the estimation process and makes speech recognition a ...Read more
Author:Sahraeian, R; Van Compernolle, D; De Wet, FebeDate:Nov 2015Recently, multilingual deep neural networks (DNNs) have been successfully used to improve under-resourced speech recognizers. Common approaches use either a merged universal phoneme set based on the International Phonetic Alphabet (IPA) or a ...Read more