2013
DOI: 10.1016/j.csl.2012.10.002
|View full text |Cite
|
Sign up to set email alerts
|

Acoustic model adaptation using in-domain background models for dysarthric speech recognition

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

2
43
1
2

Year Published

2014
2014
2021
2021

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 47 publications
(48 citation statements)
references
References 17 publications
2
43
1
2
Order By: Relevance
“…For instance, these results suggest that the variabilities in dysarthric speech can be better accommodated from modelling both typical and dysarthric domains. One such attempt was reported by [28], where background interpolation MAP was implemented to obtain an intermediate prior acoustic model to narrow the gap between two disparate SI systems (typical & dysarthric), albeit, the reported results were no better than those reported by [21]. Our best overall results, as reported in sections 3.3 & 3.4, are based on MLLR-MAP adapted SAT systems.…”
Section: Discussionsupporting
confidence: 67%
“…For instance, these results suggest that the variabilities in dysarthric speech can be better accommodated from modelling both typical and dysarthric domains. One such attempt was reported by [28], where background interpolation MAP was implemented to obtain an intermediate prior acoustic model to narrow the gap between two disparate SI systems (typical & dysarthric), albeit, the reported results were no better than those reported by [21]. Our best overall results, as reported in sections 3.3 & 3.4, are based on MLLR-MAP adapted SAT systems.…”
Section: Discussionsupporting
confidence: 67%
“…When comparing the more recent researches of Morales and Romero (2014), we found that the correlation of speech intelligibility in the recognition accuracy of ASR system was lower than the previous research by Sharma and Hasegawa-Johnson (2012), with both of them uses Nemours speech database. This may indicates that new method such as the one in Morales and Romero (2014) may have improved the recognition accuracy of the ASR system in recognising dysarthric speech with low intelligibility.…”
Section: Specific Factorscontrasting
confidence: 76%
“…For example, Parkinson's disease and amyotrophic lateral sclerosis (ALS) impact the patient's motor functions and therefore impair their speech. Only a few studies have been focused on dysarthric speech recognition [4][5][6]. Recent studies using mixed data from a variety of neurological diseases indicated articulatory data can improve the speech recognition performance [7,8].…”
Section: Introductionmentioning
confidence: 99%