fMLLR based feature-space speaker adaptation of DNN acoustic models
INTERSPEECH, pp. 3630-3634, 2015.
We investigate the problem of speaker adaptation of DNN acoustic models in two settings: the traditional unsupervised adaptation and a supervised adaptation (SuA) where a few minutes of transcribed speech is available. SuA presents additional difficulties when a test speaker’s adaptation information does not match the registered speaker’s...More
Full Text (Upload PDF)
PPT (Upload PPT)