Output details
11 - Computer Science and Informatics
Birmingham City University
Incorporating the Voicing Information into HMM-based Automatic Speech Recognition in Noisy Environments
<22> All current speech technology research in some way exploits the so-called source-filter model of speech. This paper, in a highly reputable speech science and technology journal, presents an advancement of knowledge on automatic estimation of the source characteristics of speech signals that has been omitted in research so far. For the first time this paper presents a method for estimation of the voicing characteristics of speech for individual frequency regions and its incorporation into automatic speech recognition. This work was one of the outcomes of the £115k EPSRC funding and helped to enable another £356k EPSRC funding.