Standard speaker adaptation algorithms perform poorly on dysarthric speech because of the limited phonemic repertoire of dysarthric speakers. In a previous paper, we proposed the use of "metamodels" to correct dysarthric speech. Here, we report on an improved technique that makes use of a cascade of Weighted Finite-State Transducers (WFSTs) at the confusion-matrix, word and language levels. This approach outperforms both standard MLLR and metamodels.
|Number of pages||4|
|Publication status||Published - Sep 2008|
|Event||9th Annual Conference of the International Speech Communication Association (INTERSPEECH) - Brisbane, Australia|
Duration: 22 Sep 2008 → 26 Sep 2008
|Conference||9th Annual Conference of the International Speech Communication Association (INTERSPEECH)|
|Period||22/09/08 → 26/09/08|