dc.creator | Gibson, Matthew | |
dc.date.accessioned | 2018-11-24T13:10:51Z | |
dc.date.available | 2010-08-26T13:34:34Z | |
dc.date.available | 2018-11-24T13:10:51Z | |
dc.date.issued | 2009 | |
dc.identifier | http://www.dspace.cam.ac.uk/handle/1810/226336 | |
dc.identifier.uri | http://repository.aust.edu.ng/xmlui/handle/123456789/2835 | |
dc.description.abstract | Hidden Markov model (HMM) -based speech synthesis systems possess several advantages over concatenative synthesis systems. One such advantage is the relative ease with which HMM-based systems are adapted to speakers not present in the training dataset. Speaker adaptation methods used in the field of HMM-based automatic speech recognition (ASR) are adopted for this task. In the case of unsupervised speaker adaptation, previous work has used a supplementary set of acoustic models to firstly estimate the transcription of the adaptation data. By defining a mapping between HMM-based synthesis models and ASR-style models, this paper introduces an approach to the unsupervised speaker adaptation task for HMM-based speech synthesis models which avoids the need for supplementary acoustic models. Further, this enables unsupervised adaptation of HMM-based speech synthesis models without the need to perform linguistic analysis of the estimated transcription of the adaptation data. | |
dc.subject | HMM-based speech synthesis | |
dc.subject | unsupervised speaker adaptation | |
dc.title | Two-pass decision tree construction for unsupervised adaptation of HMM-based synthesis models | |
dc.type | Article | |